graphkit-learn/notebooks/run_treepatternkernel.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- This is a regression problem ---\n",
      "\n",
      "1. Loading dataset from file...\n",
      "\n",
      "2. Calculating gram matrices. This could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8798832893371582 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-10, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f54822470>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.7850902080535889 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-09, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5159d048>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.7520360946655273 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-08, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f515ac710>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8086762428283691 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-07, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f51570cc0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8013076782226562 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-06, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f515c0b38>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8211216926574707 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-05, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f514e8748>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8376433849334717 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 0.0001, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f51596e48>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8134620189666748 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 0.001, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f547e3fd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.795604944229126 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 0.01, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f51558320>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.819868803024292 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 0.1, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f514a8320>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.849412441253662 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1.0, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5158a780>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8060426712036133 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 10.0, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f514c0278>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.7692971229553223 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 100.0, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f51444438>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.834604263305664 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1000.0, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f513efc50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8828351497650146 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 10000.0, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f513eb860>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8942792415618896 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 100000.0, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f513fa358>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8877644538879395 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1000000.0, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f51450eb8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8435029983520508 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 10000000.0, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5131fa20>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.76875638961792 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 100000000.0, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f51361a58>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.7975521087646484 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1000000000.0, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f51379128>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.785642385482788 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 10000000000.0, 'kernel_type': 'untiln'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5137d160>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.821126699447632 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-10, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQsAAAECCAYAAADpWvKaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAFs9JREFUeJzt3c2vJGd1x/Hfqer7krHnzpudwTEoQZY3sMgIRsACRUZIYDsLw8YyWcRCSMPC/gPMCjZIbBASUrBkFMtmAY43Fl5YBOQNKwRGshwbBTECAx4ZTwjBxti3u2/3yaKr763urrrPqarbt3tmvh9pNPd2d1U/1S/nPvXUec5j7i4ASMlW3QAA1waCBYAQggWAEIIFgBCCBYAQggWAkJUHCzO728x+aWaXzeyRVbdnWczsVTP7LzN70cxeKG47a2Y/MrNfFf+fWXU7uzCzx83sqpm9XLqt8hht4pvF+/6SmX1odS1vr+aYv2JmV4r3+kUzu7d035eKY/6lmX16Na1uZ6XBwsxySf8m6R5JH5D0OTP7wCrbtGSfcPcL7n6x+P0RSc+7+52Sni9+v5Y9IenuudvqjvEeSXcW/y5JevSY2njUntDiMUvSN4r3+oK7PydJxWf7AUkfLLb5VvEduCasumfxEUmX3f3X7j6Q9JSk+1bcpuN0n6Qni5+flPSZFbalM3f/saQ/zd1cd4z3SfqOT/xE0mkzu+14Wnp0ao65zn2SnnL3vrv/RtJlTb4D14RVB4vbJf2+9PtrxW3XI5f0QzP7uZldKm477+6vFz//QdL51TRtqeqO8Xp/7x8uTq8eL51eXtPHvOpgcSP5uLt/SJPu90Nm9k/lO32Sd39d597fCMdYeFTSHZIuSHpd0tdX25yjsepgcUXS+0q/v7e47brj7leK/69KekaT7ucb06538f/V1bVwaeqO8bp97939DXcfuftY0rd1cKpxTR/zqoPFzyTdaWbvN7NNTQZ/nl1xm46cmd1kZienP0v6lKSXNTnWB4uHPSjp+6tp4VLVHeOzkv61uCryMUlvlk5XrmlzYy+f1eS9libH/ICZbZnZ+zUZ3P3pcbevrd4qn9zd98zsYUn/KSmX9Li7v7LKNi3JeUnPmJk0ec2/6+4/MLOfSXrazL4g6beS7l9hGzszs+9JukvSLWb2mqQvS/qaqo/xOUn3ajLI946kzx97g49AzTHfZWYXNDnlelXSFyXJ3V8xs6cl/ULSnqSH3H20ina3YUxRBxCx6tMQANcIggWAEIIFgBCCBYAQggWAkKUFi6azSUsp0DcMjvnGcL0c81KCRcvZpNfFC9oQx3xjuC6OeVk9ixt9Nilw3VlWBmfV7LqP1j1407Z9227Sjp3dzxCzPJPsIJb56CDRzfK5EgCZLe50lEiMm9/HPKvYpySNxodvt7//dBze7u3o1PZ7ZrPi6p533jjYjqayjn8/Eu3f3jilUyf+bnLMTY+hY9u86nNSYtH3dn6/ifd6e/OUdm6+vT77MfCW22hxc8+Dn5X9DRZv2u3/WYPhX0M7Wlm6d3Eed0mStnVCH9c9+19g21hslm3lsl6D5pb3kR1dfRHr1eyrSduOQiAYVfG69q9CKmDPPHb5Y/HeCzxH12Da0rRtoTbW7aMiWP70xXjNoWUdeXJ2nbs/5u4X3f3ihraW1AwAR2VZfw73Z5NqEiQekPQvqY2mPYpsa0vjfn/2vl5P2iqCytx9lUq9CdteDEa+G9hHlZoehG9tzPxu/WG7/Qe17iFsbhx+/2C57Z7RoLfgmz3ZYG+JjVGo1zDeXnz/s90lt0sHPYrxxsH7ng2bzUGrPF1qcCazlGDRdDapZZnykzv7AWLc7yvb2Znp8vtgII0nL46dPT37fBVfnOSX9dbDa+PWdvf6sQ/G6PxNoceFn3fOsj6ge+fatXvKe/FPX9Zv9mEfnj/RtDkzUuf4eb/dmMXwPduttptKjaVIUjZYHHAY3tysR1417jHeiAfspZ1oF0VKnws92DTpNZR6DNbLpa3Ng8cMBgf7ng8OVX8tE8Ei9aX0zeqXxoLBou25Zfkvx2GWFSyafNmrjDbjx900WDQe0Jsz3kwFi3b7jXzZDzNKtEuqDhbe8CNmHSfDk8EJIIRgASBkpZWypnzsGr/11n7+hPV6kzGK0qmHsvxgDGP+FKPqlKM0mFM9pnF4N772/ppBufnTjranCdHt2p7mpLZremrQZftUfsLMY3vWekxhKnWaMQ6cglWdClWdIjQR2d6Lb2r5lCcbNnveytO4VQ9wtrWfR7G1tT+YeXBfaQwjkDwzEyCqxjTaJuDUfNnmxzhsb0lJU4l2pKTGRNomJrXRZHxktJkp21tuVbfImEjVuIct/2LIfpAoj29UDVgevo+K2xpsz2kIgJC16FlYninb2Tm4GtLvy86enu0d9If7vYHxmZtnd1BxfTx1TT51ibCuixy9tt0/1+5yWnTEv2uXvM7g1Gb6QYcYN0nKbNiNfudkt+xTT2yetzyd6O90uxoSuZpS9VoNTjb7W28VH5lxgwhAzwJACMECQAjBAkDIWoxZSCbb3pKXMji9l89exShfHp0bo6jK188TYxapy3bjmkzE6JhF22zDVJbhVNtsw+Tzd5yUOt6IH3fTMYvUmEPKODEtJh8cfn+drhmc48AwUV6RHdD09agas2iCngWAEIIFgBCCBYAQggWAkPUY4PSx/M23DkrhZfmkHkV5UDPP9pO05hOuqgYzZ9KhK5K2UgOVdffXDWbND5i2TZqKDlxG5jFUSdZ0aDjo2GX7JgOD47x90tRUagBzHBiUrhpU7P6apR8zHXienRvS7Hkqp7Q3+BjRswAQQrAAELIepyFTRd3MypqZ83kXyX0dxMGqPIy26vIz6vIylmXZeRzHoUlOR5P8jbYieQupXI1lmZ5+RHIyavfRMU+FngWAEIIFgBCCBYAQggWAkPUY4Mwy2amd/YV/fLcv3XpmJlfC+nv7xW/mC9dUDTqm8ij6Zw9fc6EuByAbxvIn+mfavbTR8u5N6y9G9U93+/vRqPhNw4lb/dMdi8wk2tY0b2FqsNNuu6nIwGNV2wanmn0GqpYCoPgNgCNHsAAQQrAAEEKwABBCsAAQQrAAEEKwABCyHnkWZvKtDWm3XLA3m1kSsLz26HxeRdUkrlSeRaqWQt2Eq+i1+Gi+xLxRcKLXsvIsjrVgb8P6FJ0L9iY+7W3zLLoXEk6/Dtmw4nXNGn4GRt3yVOhZAAghWAAIIVgACCFYAAghWAAIWY+rIWOX3nxbVlTvVq8n9fdmroBMqntPYtv8lY6qKx/lqx3Vs1IPnz1aNzJeV85u/upK26sV0e3altVLXaVpu4TfwfYNqns3KauXt79aMZXaPnIFq6rN3dsVqCqeLT6/Bbab2UfV6011bwBHjWABIIRgASCEYAEghGABIGQ9roZM9SbN8a3FlVzm54qklK+AHOUCQHVzSo578Z5lzz05DtfkIkMr+sZM2xaZR1Kr6VyS+c07bQ3ghkGwABBCsAAQsh5jFlkmnbpZ6k9S4aw/1Oj8TTPrhmS7e7K9SdZl/9z2zOZV2Yx5//AMzdS6HnVjAtEMy3fPtYvDqTobU/mS6lnsnu24NkeDw8720o8p65/pdsyptjVtz9TwTGwtmTqep4/LBovvi51tlm7rVfUsevHXlJ4FgBCCBYAQggWAEIIFgBCCBYCQTldDzOxVSX+RNJK05+4XzeyspP+Q9A+SXpV0v7v/X7dmAli1o+hZfMLdL7j7xeL3RyQ97+53Snq++B3ANW4ZpyH3SXqy+PlJSZ9ZwnMAOGZdg4VL+qGZ/dzMLhW3nXf314uf/yDpfMfnALAGumZwftzdr5jZ30r6kZn9d/lOd3czq0wRK4LLJUnazk92bAaAZevUs3D3K8X/VyU9I+kjkt4ws9skqfj/as22j7n7RXe/uJmf6NIMAMegdbAws5vM7OT0Z0mfkvSypGclPVg87EFJ3+/aSACr1+U05LykZ8xsup/vuvsPzOxnkp42sy9I+q2k+7s3E8CqtQ4W7v5rSf9Ycfv/Svpko52ZpM2N/Vmn0qQy1njjoHRRtlteRX129lxVlaq8v3DTbDsTfaq6ilLhdT2Cs0fnjTdjj8s7rlVRp20FrqnxYpGzWk1neXZtm6eqTO21XIslMGv0UL3ArNXBYhmvrGHlqxGrqAM4DgQLACEECwAhBAsAIQQLACEECwAhBAsAIQQLACEECwAh67FuiLvsL+9Ixfqk3suV7e7NZm32sv11RObXBKnK1hz3DrLVqtYVSWVi1t1ftS9pMbuw7boe0czMJuuElqUyS9uundFm+yYZmZ51b1sqQzPSnqrHVK3p0UhFduaC4iHlbNHRsNnfeqvKNK2eFF6JngWAEIIFgBCCBYAQggWAEIIFgBCCBYAQggWAEIIFgBCCBYAQggWAEIIFgBCCBYAQggWAkPWYdVrwXjG1bnNx8Yn5dUSS+yrNDq1aV6StupmJdeuMLMuy1yU5Dk1mnTZZj6St0KzT1NojS7I/2zSyxkiNpuuMLGzfaWsANwyCBYAQggWAEIIFgBCCBYAQggWAEIIFgJD1yLPIMvnJE9KgKG09GGrv3E3yUoXurD+SjSbXmAenZpMFqipdp6pr908fHifrqmfng0M327d7tl0eRDT3oHOl6xq757pdi/cGVcezYCXzqeG5jgddVd26bK/d387tc++22m4qy9K5E8Ph4gt7+7k3Gz3PaLx4fG/0RuHt6VkACCFYAAghWAAIIVgACCFYAAghWAAIIVgACCFYAAghWAAIIVgACCFYAAghWAAIIVgACCFYAAghWAAIWY96FpJ8sycbHBQ48J5ptHkQy7L+wbz7+VoT443F2hGpehZ19SoO26ck5YNYvYcma2LMPm/sccuqZ9GkHkXl9k3W1Rg2rPmRqkeRYBuH143wlvUsIvUoDrO1kX4zq+pZbGbxWhSS9G5FPYsm6FkACCFYAAghWAAIIVgACEkGCzN73MyumtnLpdvOmtmPzOxXxf9nitvNzL5pZpfN7CUz+9AyGw/g+ER6Fk9IunvutkckPe/ud0p6vvhdku6RdGfx75KkR4+mmQBWLRks3P3Hkv40d/N9kp4sfn5S0mdKt3/HJ34i6bSZ3XZUjQWwOm3HLM67++vFz3+QdL74+XZJvy897rXiNgDXuM4DnO7ukhpny5jZJTN7wcxeGIze6doMAEvWNli8MT29KP6/Wtx+RdL7So97b3HbAnd/zN0vuvvFzfxEy2YAOC5t072flfSgpK8V/3+/dPvDZvaUpI9KerN0ulLPXfb2u1JepLTmmbL+aCbF2/NsfznD+VTuqtRuzw5SiSuXN0wsQ1iX1l2XDj3/HG3TsaPbtU0nT23XdEnBBQ1SuJscg+dqvbzg/j5S21ugg1yRcl6Vit1EZPu8eN5yavm7e8G5AdN9VKSlW4OM+2SwMLPvSbpL0i1m9pqkL2sSJJ42sy9I+q2k+4uHPyfpXkmXJb0j6fPxpgBYZ8lg4e6fq7nrkxWPdUkPdW0UgPVDBieAEIIFgBCCBYAQggWAEIIFgBCCBYAQggWAEIIFgJC1qe4tSconscs3F5s1X+07pZx+XVepu426quBH+RwRy64efhyaVBJvVDW8rUD18FSF8GWZpnlHKoHXqaoGbg3mgK5HsMgy+c1/IxtMXggb7Gl4/oQ8P/gC5v2xsr3Jgb1zcvZTVvWhS5Xs758+/Mtd90GOzp3on2n34Y4GgWUtBTA813HHTcr1N5zrceLWvzZszKw8P/yLPhy2+zrccesfW2031bN0AKqaB/Lhs79r9Dz98eLxvZTHJwNxGgIghGABIIRgASCEYAEghGABIIRgASCEYAEghGABIIRgASCEYAEghGABIIRgASCEYAEghGABIIRgASBkPepZSBpv95QPDmopeG4ab5brWah039y2FQVdUmuZpgqvVEz9lxSvZ9F6LdJokZe9JRXbaVKPokKT4jDJtUfnpOpRpGz1Fou/lLWtZxGpR3GYmzf6ycdU1bPYqChmc5iqehZN0LMAEEKwABBCsAAQQrAAEEKwABBCsAAQQrAAELIeeRbuyt/alfeK2JVlyvvjmdyKcc/21xGZXxOkKqdiXFpzpCqnIpUvUXd/Xf7E/HO0XtcjmD/ROo8jtV3D3IeF/TfZ3hrkdOTeOg9iKrX9dCGfQ5tRketRlQPRRGT7jXySU1HO6Xh7b6vR82xVfCizBu8BPQsAIQQLACEECwAhBAsAIQQLACEECwAhBAsAIQQLACHrkZQ1lU1i13h7sVnzxXBSyklSVcVx2qormtOxrkjzdiy7uM5xaFBop0lRnbYixXVSBXSWZZqMFSmUU6eqWI6JpCwAR4xgASCEYAEghGABIGQtBjg9M412tpXtTmbFZbt7Gr5nW54dDGhmA5cVk+b6O7aw/bx8ePjAzWAn0aaagcxode/hmXYDch4c9LPBcqp7b597t9P2kZmbU8NhosT6nDtu/WPT5sxIVeFuO3v0Y7f8ptV2U1uWnqL89mhxhuknd15p9Dy7vnh8T2XxAVN6FgBCCBYAQggWAEIIFgBCksHCzB43s6tm9nLptq+Y2RUze7H4d2/pvi+Z2WUz+6WZfXpZDQdwvCI9iyck3V1x+zfc/ULx7zlJMrMPSHpA0geLbb5lZs2GvAGspWSwcPcfS/pTcH/3SXrK3fvu/htJlyV9pEP7AKyJLmMWD5vZS8Vpypnittsl/b70mNeK2wBc49oGi0cl3SHpgqTXJX296Q7M7JKZvWBmLwz33mnZDADHpVWwcPc33H3k7mNJ39bBqcYVSe8rPfS9xW1V+3jM3S+6+8WN3ok2zQBwjFoFCzO7rfTrZyVNr5Q8K+kBM9sys/dLulPST7s1EcA6SM4NMbPvSbpL0i1m9pqkL0u6y8wuSHJJr0r6oiS5+ytm9rSkX0jak/SQu6+mWgiAI5UMFu7+uYqb//2Qx39V0le7NArA+iGDE0AIwQJAyFrUs5CZxhv5fj0LqahxsTlbz6J8X9l4c3GXeaLuRF29iv191hS2zYbBVc4bFKOd0QvWgxgsJzG2ST2KKlsb8eXjm9azSNWjSEkVu21bzyJSj+IwJ/Pd5GOq6llsW7C4SqGqnkUT9CwAhBAsAIQQLACEECwAhBAsAIQQLACEECwAhBAsAISsRVKWjV29v/TlvUns8l6mbOCziVi9g2Ss+QWEqhKwxqV8n6pFiFKLBdUlX9WtXj6f5NV6EaBoslXLnKxUsljTRKku2+cNEteybNw6aWoqtf1Gnp7zWJUYVpUw1URk++kK6OUEsD+Pm5V2qEriYhV1AEduLXoWU9OexXhj8a/TfPp3cl+l3kRVOnhbdWnidenhy7L0dPJj0CS1vEkaeVuRdPJUyviyTHsUkdTwOlU9i8zoWQA4YgQLACFrcRrimWnv5Jay4WQQJxuONLx5a2YwMRu6bDTpMg1Ozsa4qlOD1ADm4FSi+5VV32/BWad2dhB63MLT1jzvvNFwOXH+9nNvdtp+M4sXRms6YPnhs79r2pwZG4m2vb3XbqCy6Wrm8yKzR6sGM//5RLNTkqEvfrZuajAOT88CQAjBAkAIwQJACMECQAjBAkAIwQJACMECQAjBAkAIwQJACMECQAjBAkDIWswNkSbzQabTyj3PZCOXlVL5Pbf9uSLzM4mrZhaX55VUzR2x1BSGUU3xm7raLnNzOrxm+5RRcDtrOUU9NfdkNO729+PdBtvnDaaob2Yj9cfdPq6p7bey9DT4qvklXVf6imw/nT9SnkdSNdfjMBu2+OE1xT+n9CwAhBAsAIQQLACEECwAhBAsAISYNxxRXUojzP5H0l8l/XHVbTlmt4hjvhGs8zH/vbvfGnngWgQLSTKzF9z94qrbcZw45hvD9XLMnIYACCFYAAhZp2Dx2KobsAIc843hujjmtRmzALDe1qlnAWCNESwAhBAsAIQQLACEECwAhPw/LvSDOWuiMpAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f51608e10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.769596576690674 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-09, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f513da518>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.7997679710388184 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-08, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f513f4828>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.760547399520874 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-07, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f515f0470>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.6406989097595215 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-06, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f515301d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.7307281494140625 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-05, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQsAAAECCAYAAADpWvKaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAFs9JREFUeJzt3c2vJGd1x/Hfqer7krHnzpudwTEoQZY3sMgIRsACRUZIYDsLw8YyWcRCSMPC/gPMCjZIbBASUrBkFMtmAY43Fl5YBOQNKwRGshwbBTECAx4ZTwjBxti3u2/3yaKr763urrrPqarbt3tmvh9pNPd2d1U/1S/nPvXUec5j7i4ASMlW3QAA1waCBYAQggWAEIIFgBCCBYAQggWAkJUHCzO728x+aWaXzeyRVbdnWczsVTP7LzN70cxeKG47a2Y/MrNfFf+fWXU7uzCzx83sqpm9XLqt8hht4pvF+/6SmX1odS1vr+aYv2JmV4r3+kUzu7d035eKY/6lmX16Na1uZ6XBwsxySf8m6R5JH5D0OTP7wCrbtGSfcPcL7n6x+P0RSc+7+52Sni9+v5Y9IenuudvqjvEeSXcW/y5JevSY2njUntDiMUvSN4r3+oK7PydJxWf7AUkfLLb5VvEduCasumfxEUmX3f3X7j6Q9JSk+1bcpuN0n6Qni5+flPSZFbalM3f/saQ/zd1cd4z3SfqOT/xE0mkzu+14Wnp0ao65zn2SnnL3vrv/RtJlTb4D14RVB4vbJf2+9PtrxW3XI5f0QzP7uZldKm477+6vFz//QdL51TRtqeqO8Xp/7x8uTq8eL51eXtPHvOpgcSP5uLt/SJPu90Nm9k/lO32Sd39d597fCMdYeFTSHZIuSHpd0tdX25yjsepgcUXS+0q/v7e47brj7leK/69KekaT7ucb06538f/V1bVwaeqO8bp97939DXcfuftY0rd1cKpxTR/zqoPFzyTdaWbvN7NNTQZ/nl1xm46cmd1kZienP0v6lKSXNTnWB4uHPSjp+6tp4VLVHeOzkv61uCryMUlvlk5XrmlzYy+f1eS9libH/ICZbZnZ+zUZ3P3pcbevrd4qn9zd98zsYUn/KSmX9Li7v7LKNi3JeUnPmJk0ec2/6+4/MLOfSXrazL4g6beS7l9hGzszs+9JukvSLWb2mqQvS/qaqo/xOUn3ajLI946kzx97g49AzTHfZWYXNDnlelXSFyXJ3V8xs6cl/ULSnqSH3H20ina3YUxRBxCx6tMQANcIggWAEIIFgBCCBYAQggWAkKUFi6azSUsp0DcMjvnGcL0c81KCRcvZpNfFC9oQx3xjuC6OeVk9ixt9Nilw3VlWBmfV7LqP1j1407Z9227Sjp3dzxCzPJPsIJb56CDRzfK5EgCZLe50lEiMm9/HPKvYpySNxodvt7//dBze7u3o1PZ7ZrPi6p533jjYjqayjn8/Eu3f3jilUyf+bnLMTY+hY9u86nNSYtH3dn6/ifd6e/OUdm6+vT77MfCW22hxc8+Dn5X9DRZv2u3/WYPhX0M7Wlm6d3Eed0mStnVCH9c9+19g21hslm3lsl6D5pb3kR1dfRHr1eyrSduOQiAYVfG69q9CKmDPPHb5Y/HeCzxH12Da0rRtoTbW7aMiWP70xXjNoWUdeXJ2nbs/5u4X3f3ihraW1AwAR2VZfw73Z5NqEiQekPQvqY2mPYpsa0vjfn/2vl5P2iqCytx9lUq9CdteDEa+G9hHlZoehG9tzPxu/WG7/Qe17iFsbhx+/2C57Z7RoLfgmz3ZYG+JjVGo1zDeXnz/s90lt0sHPYrxxsH7ng2bzUGrPF1qcCazlGDRdDapZZnykzv7AWLc7yvb2Znp8vtgII0nL46dPT37fBVfnOSX9dbDa+PWdvf6sQ/G6PxNoceFn3fOsj6ge+fatXvKe/FPX9Zv9mEfnj/RtDkzUuf4eb/dmMXwPduttptKjaVIUjZYHHAY3tysR1417jHeiAfspZ1oF0VKnws92DTpNZR6DNbLpa3Ng8cMBgf7ng8OVX8tE8Ei9aX0zeqXxoLBou25Zfkvx2GWFSyafNmrjDbjx900WDQe0Jsz3kwFi3b7jXzZDzNKtEuqDhbe8CNmHSfDk8EJIIRgASBkpZWypnzsGr/11n7+hPV6kzGK0qmHsvxgDGP+FKPqlKM0mFM9pnF4N772/ppBufnTjranCdHt2p7mpLZremrQZftUfsLMY3vWekxhKnWaMQ6cglWdClWdIjQR2d6Lb2r5lCcbNnveytO4VQ9wtrWfR7G1tT+YeXBfaQwjkDwzEyCqxjTaJuDUfNnmxzhsb0lJU4l2pKTGRNomJrXRZHxktJkp21tuVbfImEjVuIct/2LIfpAoj29UDVgevo+K2xpsz2kIgJC16FlYninb2Tm4GtLvy86enu0d9If7vYHxmZtnd1BxfTx1TT51ibCuixy9tt0/1+5yWnTEv2uXvM7g1Gb6QYcYN0nKbNiNfudkt+xTT2yetzyd6O90uxoSuZpS9VoNTjb7W28VH5lxgwhAzwJACMECQAjBAkDIWoxZSCbb3pKXMji9l89exShfHp0bo6jK188TYxapy3bjmkzE6JhF22zDVJbhVNtsw+Tzd5yUOt6IH3fTMYvUmEPKODEtJh8cfn+drhmc48AwUV6RHdD09agas2iCngWAEIIFgBCCBYAQggWAkPUY4PSx/M23DkrhZfmkHkV5UDPP9pO05hOuqgYzZ9KhK5K2UgOVdffXDWbND5i2TZqKDlxG5jFUSdZ0aDjo2GX7JgOD47x90tRUagBzHBiUrhpU7P6apR8zHXienRvS7Hkqp7Q3+BjRswAQQrAAELIepyFTRd3MypqZ83kXyX0dxMGqPIy26vIz6vIylmXZeRzHoUlOR5P8jbYieQupXI1lmZ5+RHIyavfRMU+FngWAEIIFgBCCBYAQggWAkPUY4Mwy2amd/YV/fLcv3XpmJlfC+nv7xW/mC9dUDTqm8ij6Zw9fc6EuByAbxvIn+mfavbTR8u5N6y9G9U93+/vRqPhNw4lb/dMdi8wk2tY0b2FqsNNuu6nIwGNV2wanmn0GqpYCoPgNgCNHsAAQQrAAEEKwABBCsAAQQrAAEEKwABCyHnkWZvKtDWm3XLA3m1kSsLz26HxeRdUkrlSeRaqWQt2Eq+i1+Gi+xLxRcKLXsvIsjrVgb8P6FJ0L9iY+7W3zLLoXEk6/Dtmw4nXNGn4GRt3yVOhZAAghWAAIIVgACCFYAAghWAAIWY+rIWOX3nxbVlTvVq8n9fdmroBMqntPYtv8lY6qKx/lqx3Vs1IPnz1aNzJeV85u/upK26sV0e3altVLXaVpu4TfwfYNqns3KauXt79aMZXaPnIFq6rN3dsVqCqeLT6/Bbab2UfV6011bwBHjWABIIRgASCEYAEghGABIGQ9roZM9SbN8a3FlVzm54qklK+AHOUCQHVzSo578Z5lzz05DtfkIkMr+sZM2xaZR1Kr6VyS+c07bQ3ghkGwABBCsAAQsh5jFlkmnbpZ6k9S4aw/1Oj8TTPrhmS7e7K9SdZl/9z2zOZV2Yx5//AMzdS6HnVjAtEMy3fPtYvDqTobU/mS6lnsnu24NkeDw8720o8p65/pdsyptjVtz9TwTGwtmTqep4/LBovvi51tlm7rVfUsevHXlJ4FgBCCBYAQggWAEIIFgBCCBYCQTldDzOxVSX+RNJK05+4XzeyspP+Q9A+SXpV0v7v/X7dmAli1o+hZfMLdL7j7xeL3RyQ97+53Snq++B3ANW4ZpyH3SXqy+PlJSZ9ZwnMAOGZdg4VL+qGZ/dzMLhW3nXf314uf/yDpfMfnALAGumZwftzdr5jZ30r6kZn9d/lOd3czq0wRK4LLJUnazk92bAaAZevUs3D3K8X/VyU9I+kjkt4ws9skqfj/as22j7n7RXe/uJmf6NIMAMegdbAws5vM7OT0Z0mfkvSypGclPVg87EFJ3+/aSACr1+U05LykZ8xsup/vuvsPzOxnkp42sy9I+q2k+7s3E8CqtQ4W7v5rSf9Ycfv/Svpko52ZpM2N/Vmn0qQy1njjoHRRtlteRX129lxVlaq8v3DTbDsTfaq6ilLhdT2Cs0fnjTdjj8s7rlVRp20FrqnxYpGzWk1neXZtm6eqTO21XIslMGv0UL3ArNXBYhmvrGHlqxGrqAM4DgQLACEECwAhBAsAIQQLACEECwAhBAsAIQQLACEECwAh67FuiLvsL+9Ixfqk3suV7e7NZm32sv11RObXBKnK1hz3DrLVqtYVSWVi1t1ftS9pMbuw7boe0czMJuuElqUyS9uundFm+yYZmZ51b1sqQzPSnqrHVK3p0UhFduaC4iHlbNHRsNnfeqvKNK2eFF6JngWAEIIFgBCCBYAQggWAEIIFgBCCBYAQggWAEIIFgBCCBYAQggWAEIIFgBCCBYAQggWAkPWYdVrwXjG1bnNx8Yn5dUSS+yrNDq1aV6StupmJdeuMLMuy1yU5Dk1mnTZZj6St0KzT1NojS7I/2zSyxkiNpuuMLGzfaWsANwyCBYAQggWAEIIFgBCCBYAQggWAEIIFgJD1yLPIMvnJE9KgKG09GGrv3E3yUoXurD+SjSbXmAenZpMFqipdp6pr908fHifrqmfng0M327d7tl0eRDT3oHOl6xq757pdi/cGVcezYCXzqeG5jgddVd26bK/d387tc++22m4qy9K5E8Ph4gt7+7k3Gz3PaLx4fG/0RuHt6VkACCFYAAghWAAIIVgACCFYAAghWAAIIVgACCFYAAghWAAIIVgACCFYAAghWAAIIVgACCFYAAghWAAIWY96FpJ8sycbHBQ48J5ptHkQy7L+wbz7+VoT443F2hGpehZ19SoO26ck5YNYvYcma2LMPm/sccuqZ9GkHkXl9k3W1Rg2rPmRqkeRYBuH143wlvUsIvUoDrO1kX4zq+pZbGbxWhSS9G5FPYsm6FkACCFYAAghWAAIIVgACEkGCzN73MyumtnLpdvOmtmPzOxXxf9nitvNzL5pZpfN7CUz+9AyGw/g+ER6Fk9IunvutkckPe/ud0p6vvhdku6RdGfx75KkR4+mmQBWLRks3P3Hkv40d/N9kp4sfn5S0mdKt3/HJ34i6bSZ3XZUjQWwOm3HLM67++vFz3+QdL74+XZJvy897rXiNgDXuM4DnO7ukhpny5jZJTN7wcxeGIze6doMAEvWNli8MT29KP6/Wtx+RdL7So97b3HbAnd/zN0vuvvFzfxEy2YAOC5t072flfSgpK8V/3+/dPvDZvaUpI9KerN0ulLPXfb2u1JepLTmmbL+aCbF2/NsfznD+VTuqtRuzw5SiSuXN0wsQ1iX1l2XDj3/HG3TsaPbtU0nT23XdEnBBQ1SuJscg+dqvbzg/j5S21ugg1yRcl6Vit1EZPu8eN5yavm7e8G5AdN9VKSlW4OM+2SwMLPvSbpL0i1m9pqkL2sSJJ42sy9I+q2k+4uHPyfpXkmXJb0j6fPxpgBYZ8lg4e6fq7nrkxWPdUkPdW0UgPVDBieAEIIFgBCCBYAQggWAEIIFgBCCBYAQggWAEIIFgJC1qe4tSconscs3F5s1X+07pZx+XVepu426quBH+RwRy64efhyaVBJvVDW8rUD18FSF8GWZpnlHKoHXqaoGbg3mgK5HsMgy+c1/IxtMXggb7Gl4/oQ8P/gC5v2xsr3Jgb1zcvZTVvWhS5Xs758+/Mtd90GOzp3on2n34Y4GgWUtBTA813HHTcr1N5zrceLWvzZszKw8P/yLPhy2+zrccesfW2031bN0AKqaB/Lhs79r9Dz98eLxvZTHJwNxGgIghGABIIRgASCEYAEghGABIIRgASCEYAEghGABIIRgASCEYAEghGABIIRgASCEYAEghGABIIRgASBkPepZSBpv95QPDmopeG4ab5brWah039y2FQVdUmuZpgqvVEz9lxSvZ9F6LdJokZe9JRXbaVKPokKT4jDJtUfnpOpRpGz1Fou/lLWtZxGpR3GYmzf6ycdU1bPYqChmc5iqehZN0LMAEEKwABBCsAAQQrAAEEKwABBCsAAQQrAAELIeeRbuyt/alfeK2JVlyvvjmdyKcc/21xGZXxOkKqdiXFpzpCqnIpUvUXd/Xf7E/HO0XtcjmD/ROo8jtV3D3IeF/TfZ3hrkdOTeOg9iKrX9dCGfQ5tRketRlQPRRGT7jXySU1HO6Xh7b6vR82xVfCizBu8BPQsAIQQLACEECwAhBAsAIQQLACEECwAhBAsAIQQLACHrkZQ1lU1i13h7sVnzxXBSyklSVcVx2qormtOxrkjzdiy7uM5xaFBop0lRnbYixXVSBXSWZZqMFSmUU6eqWI6JpCwAR4xgASCEYAEghGABIGQtBjg9M412tpXtTmbFZbt7Gr5nW54dDGhmA5cVk+b6O7aw/bx8ePjAzWAn0aaagcxode/hmXYDch4c9LPBcqp7b597t9P2kZmbU8NhosT6nDtu/WPT5sxIVeFuO3v0Y7f8ptV2U1uWnqL89mhxhuknd15p9Dy7vnh8T2XxAVN6FgBCCBYAQggWAEIIFgBCksHCzB43s6tm9nLptq+Y2RUze7H4d2/pvi+Z2WUz+6WZfXpZDQdwvCI9iyck3V1x+zfc/ULx7zlJMrMPSHpA0geLbb5lZs2GvAGspWSwcPcfS/pTcH/3SXrK3fvu/htJlyV9pEP7AKyJLmMWD5vZS8Vpypnittsl/b70mNeK2wBc49oGi0cl3SHpgqTXJX296Q7M7JKZvWBmLwz33mnZDADHpVWwcPc33H3k7mNJ39bBqcYVSe8rPfS9xW1V+3jM3S+6+8WN3ok2zQBwjFoFCzO7rfTrZyVNr5Q8K+kBM9sys/dLulPST7s1EcA6SM4NMbPvSbpL0i1m9pqkL0u6y8wuSHJJr0r6oiS5+ytm9rSkX0jak/SQu6+mWgiAI5UMFu7+uYqb//2Qx39V0le7NArA+iGDE0AIwQJAyFrUs5CZxhv5fj0LqahxsTlbz6J8X9l4c3GXeaLuRF29iv191hS2zYbBVc4bFKOd0QvWgxgsJzG2ST2KKlsb8eXjm9azSNWjSEkVu21bzyJSj+IwJ/Pd5GOq6llsW7C4SqGqnkUT9CwAhBAsAIQQLACEECwAhBAsAIQQLACEECwAhBAsAISsRVKWjV29v/TlvUns8l6mbOCziVi9g2Ss+QWEqhKwxqV8n6pFiFKLBdUlX9WtXj6f5NV6EaBoslXLnKxUsljTRKku2+cNEteybNw6aWoqtf1Gnp7zWJUYVpUw1URk++kK6OUEsD+Pm5V2qEriYhV1AEduLXoWU9OexXhj8a/TfPp3cl+l3kRVOnhbdWnidenhy7L0dPJj0CS1vEkaeVuRdPJUyviyTHsUkdTwOlU9i8zoWQA4YgQLACFrcRrimWnv5Jay4WQQJxuONLx5a2YwMRu6bDTpMg1Ozsa4qlOD1ADm4FSi+5VV32/BWad2dhB63MLT1jzvvNFwOXH+9nNvdtp+M4sXRms6YPnhs79r2pwZG4m2vb3XbqCy6Wrm8yKzR6sGM//5RLNTkqEvfrZuajAOT88CQAjBAkAIwQJACMECQAjBAkAIwQJACMECQAjBAkAIwQJACMECQAjBAkDIWswNkSbzQabTyj3PZCOXlVL5Pbf9uSLzM4mrZhaX55VUzR2x1BSGUU3xm7raLnNzOrxm+5RRcDtrOUU9NfdkNO729+PdBtvnDaaob2Yj9cfdPq6p7bey9DT4qvklXVf6imw/nT9SnkdSNdfjMBu2+OE1xT+n9CwAhBAsAIQQLACEECwAhBAsAISYNxxRXUojzP5H0l8l/XHVbTlmt4hjvhGs8zH/vbvfGnngWgQLSTKzF9z94qrbcZw45hvD9XLMnIYACCFYAAhZp2Dx2KobsAIc843hujjmtRmzALDe1qlnAWCNESwAhBAsAIQQLACEECwAhPw/LvSDOWuiMpAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5142ca90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.7127954959869385 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 0.0001, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f51485748>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.7299909591674805 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 0.001, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQsAAAECCAYAAADpWvKaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAFs9JREFUeJzt3c2vJGd1x/Hfqer7krHnzpudwTEoQZY3sMgIRsACRUZIYDsLw8YyWcRCSMPC/gPMCjZIbBASUrBkFMtmAY43Fl5YBOQNKwRGshwbBTECAx4ZTwjBxti3u2/3yaKr763urrrPqarbt3tmvh9pNPd2d1U/1S/nPvXUec5j7i4ASMlW3QAA1waCBYAQggWAEIIFgBCCBYAQggWAkJUHCzO728x+aWaXzeyRVbdnWczsVTP7LzN70cxeKG47a2Y/MrNfFf+fWXU7uzCzx83sqpm9XLqt8hht4pvF+/6SmX1odS1vr+aYv2JmV4r3+kUzu7d035eKY/6lmX16Na1uZ6XBwsxySf8m6R5JH5D0OTP7wCrbtGSfcPcL7n6x+P0RSc+7+52Sni9+v5Y9IenuudvqjvEeSXcW/y5JevSY2njUntDiMUvSN4r3+oK7PydJxWf7AUkfLLb5VvEduCasumfxEUmX3f3X7j6Q9JSk+1bcpuN0n6Qni5+flPSZFbalM3f/saQ/zd1cd4z3SfqOT/xE0mkzu+14Wnp0ao65zn2SnnL3vrv/RtJlTb4D14RVB4vbJf2+9PtrxW3XI5f0QzP7uZldKm477+6vFz//QdL51TRtqeqO8Xp/7x8uTq8eL51eXtPHvOpgcSP5uLt/SJPu90Nm9k/lO32Sd39d597fCMdYeFTSHZIuSHpd0tdX25yjsepgcUXS+0q/v7e47brj7leK/69KekaT7ucb06538f/V1bVwaeqO8bp97939DXcfuftY0rd1cKpxTR/zqoPFzyTdaWbvN7NNTQZ/nl1xm46cmd1kZienP0v6lKSXNTnWB4uHPSjp+6tp4VLVHeOzkv61uCryMUlvlk5XrmlzYy+f1eS9libH/ICZbZnZ+zUZ3P3pcbevrd4qn9zd98zsYUn/KSmX9Li7v7LKNi3JeUnPmJk0ec2/6+4/MLOfSXrazL4g6beS7l9hGzszs+9JukvSLWb2mqQvS/qaqo/xOUn3ajLI946kzx97g49AzTHfZWYXNDnlelXSFyXJ3V8xs6cl/ULSnqSH3H20ina3YUxRBxCx6tMQANcIggWAEIIFgBCCBYAQggWAkKUFi6azSUsp0DcMjvnGcL0c81KCRcvZpNfFC9oQx3xjuC6OeVk9ixt9Nilw3VlWBmfV7LqP1j1407Z9227Sjp3dzxCzPJPsIJb56CDRzfK5EgCZLe50lEiMm9/HPKvYpySNxodvt7//dBze7u3o1PZ7ZrPi6p533jjYjqayjn8/Eu3f3jilUyf+bnLMTY+hY9u86nNSYtH3dn6/ifd6e/OUdm6+vT77MfCW22hxc8+Dn5X9DRZv2u3/WYPhX0M7Wlm6d3Eed0mStnVCH9c9+19g21hslm3lsl6D5pb3kR1dfRHr1eyrSduOQiAYVfG69q9CKmDPPHb5Y/HeCzxH12Da0rRtoTbW7aMiWP70xXjNoWUdeXJ2nbs/5u4X3f3ihraW1AwAR2VZfw73Z5NqEiQekPQvqY2mPYpsa0vjfn/2vl5P2iqCytx9lUq9CdteDEa+G9hHlZoehG9tzPxu/WG7/Qe17iFsbhx+/2C57Z7RoLfgmz3ZYG+JjVGo1zDeXnz/s90lt0sHPYrxxsH7ng2bzUGrPF1qcCazlGDRdDapZZnykzv7AWLc7yvb2Znp8vtgII0nL46dPT37fBVfnOSX9dbDa+PWdvf6sQ/G6PxNoceFn3fOsj6ge+fatXvKe/FPX9Zv9mEfnj/RtDkzUuf4eb/dmMXwPduttptKjaVIUjZYHHAY3tysR1417jHeiAfspZ1oF0VKnws92DTpNZR6DNbLpa3Ng8cMBgf7ng8OVX8tE8Ei9aX0zeqXxoLBou25Zfkvx2GWFSyafNmrjDbjx900WDQe0Jsz3kwFi3b7jXzZDzNKtEuqDhbe8CNmHSfDk8EJIIRgASBkpZWypnzsGr/11n7+hPV6kzGK0qmHsvxgDGP+FKPqlKM0mFM9pnF4N772/ppBufnTjranCdHt2p7mpLZremrQZftUfsLMY3vWekxhKnWaMQ6cglWdClWdIjQR2d6Lb2r5lCcbNnveytO4VQ9wtrWfR7G1tT+YeXBfaQwjkDwzEyCqxjTaJuDUfNnmxzhsb0lJU4l2pKTGRNomJrXRZHxktJkp21tuVbfImEjVuIct/2LIfpAoj29UDVgevo+K2xpsz2kIgJC16FlYninb2Tm4GtLvy86enu0d9If7vYHxmZtnd1BxfTx1TT51ibCuixy9tt0/1+5yWnTEv2uXvM7g1Gb6QYcYN0nKbNiNfudkt+xTT2yetzyd6O90uxoSuZpS9VoNTjb7W28VH5lxgwhAzwJACMECQAjBAkDIWoxZSCbb3pKXMji9l89exShfHp0bo6jK188TYxapy3bjmkzE6JhF22zDVJbhVNtsw+Tzd5yUOt6IH3fTMYvUmEPKODEtJh8cfn+drhmc48AwUV6RHdD09agas2iCngWAEIIFgBCCBYAQggWAkPUY4PSx/M23DkrhZfmkHkV5UDPP9pO05hOuqgYzZ9KhK5K2UgOVdffXDWbND5i2TZqKDlxG5jFUSdZ0aDjo2GX7JgOD47x90tRUagBzHBiUrhpU7P6apR8zHXienRvS7Hkqp7Q3+BjRswAQQrAAELIepyFTRd3MypqZ83kXyX0dxMGqPIy26vIz6vIylmXZeRzHoUlOR5P8jbYieQupXI1lmZ5+RHIyavfRMU+FngWAEIIFgBCCBYAQggWAkPUY4Mwy2amd/YV/fLcv3XpmJlfC+nv7xW/mC9dUDTqm8ij6Zw9fc6EuByAbxvIn+mfavbTR8u5N6y9G9U93+/vRqPhNw4lb/dMdi8wk2tY0b2FqsNNuu6nIwGNV2wanmn0GqpYCoPgNgCNHsAAQQrAAEEKwABBCsAAQQrAAEEKwABCyHnkWZvKtDWm3XLA3m1kSsLz26HxeRdUkrlSeRaqWQt2Eq+i1+Gi+xLxRcKLXsvIsjrVgb8P6FJ0L9iY+7W3zLLoXEk6/Dtmw4nXNGn4GRt3yVOhZAAghWAAIIVgACCFYAAghWAAIWY+rIWOX3nxbVlTvVq8n9fdmroBMqntPYtv8lY6qKx/lqx3Vs1IPnz1aNzJeV85u/upK26sV0e3altVLXaVpu4TfwfYNqns3KauXt79aMZXaPnIFq6rN3dsVqCqeLT6/Bbab2UfV6011bwBHjWABIIRgASCEYAEghGABIGQ9roZM9SbN8a3FlVzm54qklK+AHOUCQHVzSo578Z5lzz05DtfkIkMr+sZM2xaZR1Kr6VyS+c07bQ3ghkGwABBCsAAQsh5jFlkmnbpZ6k9S4aw/1Oj8TTPrhmS7e7K9SdZl/9z2zOZV2Yx5//AMzdS6HnVjAtEMy3fPtYvDqTobU/mS6lnsnu24NkeDw8720o8p65/pdsyptjVtz9TwTGwtmTqep4/LBovvi51tlm7rVfUsevHXlJ4FgBCCBYAQggWAEIIFgBCCBYCQTldDzOxVSX+RNJK05+4XzeyspP+Q9A+SXpV0v7v/X7dmAli1o+hZfMLdL7j7xeL3RyQ97+53Snq++B3ANW4ZpyH3SXqy+PlJSZ9ZwnMAOGZdg4VL+qGZ/dzMLhW3nXf314uf/yDpfMfnALAGumZwftzdr5jZ30r6kZn9d/lOd3czq0wRK4LLJUnazk92bAaAZevUs3D3K8X/VyU9I+kjkt4ws9skqfj/as22j7n7RXe/uJmf6NIMAMegdbAws5vM7OT0Z0mfkvSypGclPVg87EFJ3+/aSACr1+U05LykZ8xsup/vuvsPzOxnkp42sy9I+q2k+7s3E8CqtQ4W7v5rSf9Ycfv/Svpko52ZpM2N/Vmn0qQy1njjoHRRtlteRX129lxVlaq8v3DTbDsTfaq6ilLhdT2Cs0fnjTdjj8s7rlVRp20FrqnxYpGzWk1neXZtm6eqTO21XIslMGv0UL3ArNXBYhmvrGHlqxGrqAM4DgQLACEECwAhBAsAIQQLACEECwAhBAsAIQQLACEECwAh67FuiLvsL+9Ixfqk3suV7e7NZm32sv11RObXBKnK1hz3DrLVqtYVSWVi1t1ftS9pMbuw7boe0czMJuuElqUyS9uundFm+yYZmZ51b1sqQzPSnqrHVK3p0UhFduaC4iHlbNHRsNnfeqvKNK2eFF6JngWAEIIFgBCCBYAQggWAEIIFgBCCBYAQggWAEIIFgBCCBYAQggWAEIIFgBCCBYAQggWAkPWYdVrwXjG1bnNx8Yn5dUSS+yrNDq1aV6StupmJdeuMLMuy1yU5Dk1mnTZZj6St0KzT1NojS7I/2zSyxkiNpuuMLGzfaWsANwyCBYAQggWAEIIFgBCCBYAQggWAEIIFgJD1yLPIMvnJE9KgKG09GGrv3E3yUoXurD+SjSbXmAenZpMFqipdp6pr908fHifrqmfng0M327d7tl0eRDT3oHOl6xq757pdi/cGVcezYCXzqeG5jgddVd26bK/d387tc++22m4qy9K5E8Ph4gt7+7k3Gz3PaLx4fG/0RuHt6VkACCFYAAghWAAIIVgACCFYAAghWAAIIVgACCFYAAghWAAIIVgACCFYAAghWAAIIVgACCFYAAghWAAIWY96FpJ8sycbHBQ48J5ptHkQy7L+wbz7+VoT443F2hGpehZ19SoO26ck5YNYvYcma2LMPm/sccuqZ9GkHkXl9k3W1Rg2rPmRqkeRYBuH143wlvUsIvUoDrO1kX4zq+pZbGbxWhSS9G5FPYsm6FkACCFYAAghWAAIIVgACEkGCzN73MyumtnLpdvOmtmPzOxXxf9nitvNzL5pZpfN7CUz+9AyGw/g+ER6Fk9IunvutkckPe/ud0p6vvhdku6RdGfx75KkR4+mmQBWLRks3P3Hkv40d/N9kp4sfn5S0mdKt3/HJ34i6bSZ3XZUjQWwOm3HLM67++vFz3+QdL74+XZJvy897rXiNgDXuM4DnO7ukhpny5jZJTN7wcxeGIze6doMAEvWNli8MT29KP6/Wtx+RdL7So97b3HbAnd/zN0vuvvFzfxEy2YAOC5t072flfSgpK8V/3+/dPvDZvaUpI9KerN0ulLPXfb2u1JepLTmmbL+aCbF2/NsfznD+VTuqtRuzw5SiSuXN0wsQ1iX1l2XDj3/HG3TsaPbtU0nT23XdEnBBQ1SuJscg+dqvbzg/j5S21ugg1yRcl6Vit1EZPu8eN5yavm7e8G5AdN9VKSlW4OM+2SwMLPvSbpL0i1m9pqkL2sSJJ42sy9I+q2k+4uHPyfpXkmXJb0j6fPxpgBYZ8lg4e6fq7nrkxWPdUkPdW0UgPVDBieAEIIFgBCCBYAQggWAEIIFgBCCBYAQggWAEIIFgJC1qe4tSconscs3F5s1X+07pZx+XVepu426quBH+RwRy64efhyaVBJvVDW8rUD18FSF8GWZpnlHKoHXqaoGbg3mgK5HsMgy+c1/IxtMXggb7Gl4/oQ8P/gC5v2xsr3Jgb1zcvZTVvWhS5Xs758+/Mtd90GOzp3on2n34Y4GgWUtBTA813HHTcr1N5zrceLWvzZszKw8P/yLPhy2+zrccesfW2031bN0AKqaB/Lhs79r9Dz98eLxvZTHJwNxGgIghGABIIRgASCEYAEghGABIIRgASCEYAEghGABIIRgASCEYAEghGABIIRgASCEYAEghGABIIRgASBkPepZSBpv95QPDmopeG4ab5brWah039y2FQVdUmuZpgqvVEz9lxSvZ9F6LdJokZe9JRXbaVKPokKT4jDJtUfnpOpRpGz1Fou/lLWtZxGpR3GYmzf6ycdU1bPYqChmc5iqehZN0LMAEEKwABBCsAAQQrAAEEKwABBCsAAQQrAAELIeeRbuyt/alfeK2JVlyvvjmdyKcc/21xGZXxOkKqdiXFpzpCqnIpUvUXd/Xf7E/HO0XtcjmD/ROo8jtV3D3IeF/TfZ3hrkdOTeOg9iKrX9dCGfQ5tRketRlQPRRGT7jXySU1HO6Xh7b6vR82xVfCizBu8BPQsAIQQLACEECwAhBAsAIQQLACEECwAhBAsAIQQLACHrkZQ1lU1i13h7sVnzxXBSyklSVcVx2qormtOxrkjzdiy7uM5xaFBop0lRnbYixXVSBXSWZZqMFSmUU6eqWI6JpCwAR4xgASCEYAEghGABIGQtBjg9M412tpXtTmbFZbt7Gr5nW54dDGhmA5cVk+b6O7aw/bx8ePjAzWAn0aaagcxode/hmXYDch4c9LPBcqp7b597t9P2kZmbU8NhosT6nDtu/WPT5sxIVeFuO3v0Y7f8ptV2U1uWnqL89mhxhuknd15p9Dy7vnh8T2XxAVN6FgBCCBYAQggWAEIIFgBCksHCzB43s6tm9nLptq+Y2RUze7H4d2/pvi+Z2WUz+6WZfXpZDQdwvCI9iyck3V1x+zfc/ULx7zlJMrMPSHpA0geLbb5lZs2GvAGspWSwcPcfS/pTcH/3SXrK3fvu/htJlyV9pEP7AKyJLmMWD5vZS8Vpypnittsl/b70mNeK2wBc49oGi0cl3SHpgqTXJX296Q7M7JKZvWBmLwz33mnZDADHpVWwcPc33H3k7mNJ39bBqcYVSe8rPfS9xW1V+3jM3S+6+8WN3ok2zQBwjFoFCzO7rfTrZyVNr5Q8K+kBM9sys/dLulPST7s1EcA6SM4NMbPvSbpL0i1m9pqkL0u6y8wuSHJJr0r6oiS5+ytm9rSkX0jak/SQu6+mWgiAI5UMFu7+uYqb//2Qx39V0le7NArA+iGDE0AIwQJAyFrUs5CZxhv5fj0LqahxsTlbz6J8X9l4c3GXeaLuRF29iv191hS2zYbBVc4bFKOd0QvWgxgsJzG2ST2KKlsb8eXjm9azSNWjSEkVu21bzyJSj+IwJ/Pd5GOq6llsW7C4SqGqnkUT9CwAhBAsAIQQLACEECwAhBAsAIQQLACEECwAhBAsAISsRVKWjV29v/TlvUns8l6mbOCziVi9g2Ss+QWEqhKwxqV8n6pFiFKLBdUlX9WtXj6f5NV6EaBoslXLnKxUsljTRKku2+cNEteybNw6aWoqtf1Gnp7zWJUYVpUw1URk++kK6OUEsD+Pm5V2qEriYhV1AEduLXoWU9OexXhj8a/TfPp3cl+l3kRVOnhbdWnidenhy7L0dPJj0CS1vEkaeVuRdPJUyviyTHsUkdTwOlU9i8zoWQA4YgQLACFrcRrimWnv5Jay4WQQJxuONLx5a2YwMRu6bDTpMg1Ozsa4qlOD1ADm4FSi+5VV32/BWad2dhB63MLT1jzvvNFwOXH+9nNvdtp+M4sXRms6YPnhs79r2pwZG4m2vb3XbqCy6Wrm8yKzR6sGM//5RLNTkqEvfrZuajAOT88CQAjBAkAIwQJACMECQAjBAkAIwQJACMECQAjBAkAIwQJACMECQAjBAkDIWswNkSbzQabTyj3PZCOXlVL5Pbf9uSLzM4mrZhaX55VUzR2x1BSGUU3xm7raLnNzOrxm+5RRcDtrOUU9NfdkNO729+PdBtvnDaaob2Yj9cfdPq6p7bey9DT4qvklXVf6imw/nT9SnkdSNdfjMBu2+OE1xT+n9CwAhBAsAIQQLACEECwAhBAsAISYNxxRXUojzP5H0l8l/XHVbTlmt4hjvhGs8zH/vbvfGnngWgQLSTKzF9z94qrbcZw45hvD9XLMnIYACCFYAAhZp2Dx2KobsAIc843hujjmtRmzALDe1qlnAWCNESwAhBAsAIQQLACEECwAhPw/LvSDOWuiMpAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5147eba8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.726623773574829 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 0.01, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5133a240>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.7834739685058594 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 0.1, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5154d320>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.799546480178833 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1.0, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f51333160>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.86911940574646 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 10.0, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f514e2f98>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.633831262588501 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 100.0, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f51468cf8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.7458419799804688 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1000.0, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f513e2a20>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.9190473556518555 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 10000.0, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5133dcc0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.8917624950408936 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 100000.0, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f514bf320>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.869596004486084 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1000000.0, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5154d828>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.684450626373291 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 10000000.0, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f514586a0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.7181267738342285 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 100000000.0, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f514bc240>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.7365922927856445 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1000000000.0, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f513a3748>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 3.6689658164978027 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 10000000000.0, 'kernel_type': 'size'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f514fefd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.7944748401641846 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-10, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5144c588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.7583632469177246 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-09, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f515eb4a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8586950302124023 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-08, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f514f6588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8722550868988037 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-07, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f513a3828>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.833204746246338 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-06, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f51462e10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.9087231159210205 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1e-05, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5145bc18>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8667919635772705 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 0.0001, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f51410fd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.7739126682281494 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 0.001, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f514bf978>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.842292070388794 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 0.01, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f513e9390>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8595993518829346 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 0.1, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQsAAAECCAYAAADpWvKaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAFs9JREFUeJzt3c2vJGd1x/Hfqer7krHnzpudwTEoQZY3sMgIRsACRUZIYDsLw8YyWcRCSMPC/gPMCjZIbBASUrBkFMtmAY43Fl5YBOQNKwRGshwbBTECAx4ZTwjBxti3u2/3yaKr763urrrPqarbt3tmvh9pNPd2d1U/1S/nPvXUec5j7i4ASMlW3QAA1waCBYAQggWAEIIFgBCCBYAQggWAkJUHCzO728x+aWaXzeyRVbdnWczsVTP7LzN70cxeKG47a2Y/MrNfFf+fWXU7uzCzx83sqpm9XLqt8hht4pvF+/6SmX1odS1vr+aYv2JmV4r3+kUzu7d035eKY/6lmX16Na1uZ6XBwsxySf8m6R5JH5D0OTP7wCrbtGSfcPcL7n6x+P0RSc+7+52Sni9+v5Y9IenuudvqjvEeSXcW/y5JevSY2njUntDiMUvSN4r3+oK7PydJxWf7AUkfLLb5VvEduCasumfxEUmX3f3X7j6Q9JSk+1bcpuN0n6Qni5+flPSZFbalM3f/saQ/zd1cd4z3SfqOT/xE0mkzu+14Wnp0ao65zn2SnnL3vrv/RtJlTb4D14RVB4vbJf2+9PtrxW3XI5f0QzP7uZldKm477+6vFz//QdL51TRtqeqO8Xp/7x8uTq8eL51eXtPHvOpgcSP5uLt/SJPu90Nm9k/lO32Sd39d597fCMdYeFTSHZIuSHpd0tdX25yjsepgcUXS+0q/v7e47brj7leK/69KekaT7ucb06538f/V1bVwaeqO8bp97939DXcfuftY0rd1cKpxTR/zqoPFzyTdaWbvN7NNTQZ/nl1xm46cmd1kZienP0v6lKSXNTnWB4uHPSjp+6tp4VLVHeOzkv61uCryMUlvlk5XrmlzYy+f1eS9libH/ICZbZnZ+zUZ3P3pcbevrd4qn9zd98zsYUn/KSmX9Li7v7LKNi3JeUnPmJk0ec2/6+4/MLOfSXrazL4g6beS7l9hGzszs+9JukvSLWb2mqQvS/qaqo/xOUn3ajLI946kzx97g49AzTHfZWYXNDnlelXSFyXJ3V8xs6cl/ULSnqSH3H20ina3YUxRBxCx6tMQANcIggWAEIIFgBCCBYAQggWAkKUFi6azSUsp0DcMjvnGcL0c81KCRcvZpNfFC9oQx3xjuC6OeVk9ixt9Nilw3VlWBmfV7LqP1j1407Z9227Sjp3dzxCzPJPsIJb56CDRzfK5EgCZLe50lEiMm9/HPKvYpySNxodvt7//dBze7u3o1PZ7ZrPi6p533jjYjqayjn8/Eu3f3jilUyf+bnLMTY+hY9u86nNSYtH3dn6/ifd6e/OUdm6+vT77MfCW22hxc8+Dn5X9DRZv2u3/WYPhX0M7Wlm6d3Eed0mStnVCH9c9+19g21hslm3lsl6D5pb3kR1dfRHr1eyrSduOQiAYVfG69q9CKmDPPHb5Y/HeCzxH12Da0rRtoTbW7aMiWP70xXjNoWUdeXJ2nbs/5u4X3f3ihraW1AwAR2VZfw73Z5NqEiQekPQvqY2mPYpsa0vjfn/2vl5P2iqCytx9lUq9CdteDEa+G9hHlZoehG9tzPxu/WG7/Qe17iFsbhx+/2C57Z7RoLfgmz3ZYG+JjVGo1zDeXnz/s90lt0sHPYrxxsH7ng2bzUGrPF1qcCazlGDRdDapZZnykzv7AWLc7yvb2Znp8vtgII0nL46dPT37fBVfnOSX9dbDa+PWdvf6sQ/G6PxNoceFn3fOsj6ge+fatXvKe/FPX9Zv9mEfnj/RtDkzUuf4eb/dmMXwPduttptKjaVIUjZYHHAY3tysR1417jHeiAfspZ1oF0VKnws92DTpNZR6DNbLpa3Ng8cMBgf7ng8OVX8tE8Ei9aX0zeqXxoLBou25Zfkvx2GWFSyafNmrjDbjx900WDQe0Jsz3kwFi3b7jXzZDzNKtEuqDhbe8CNmHSfDk8EJIIRgASBkpZWypnzsGr/11n7+hPV6kzGK0qmHsvxgDGP+FKPqlKM0mFM9pnF4N772/ppBufnTjranCdHt2p7mpLZremrQZftUfsLMY3vWekxhKnWaMQ6cglWdClWdIjQR2d6Lb2r5lCcbNnveytO4VQ9wtrWfR7G1tT+YeXBfaQwjkDwzEyCqxjTaJuDUfNnmxzhsb0lJU4l2pKTGRNomJrXRZHxktJkp21tuVbfImEjVuIct/2LIfpAoj29UDVgevo+K2xpsz2kIgJC16FlYninb2Tm4GtLvy86enu0d9If7vYHxmZtnd1BxfTx1TT51ibCuixy9tt0/1+5yWnTEv2uXvM7g1Gb6QYcYN0nKbNiNfudkt+xTT2yetzyd6O90uxoSuZpS9VoNTjb7W28VH5lxgwhAzwJACMECQAjBAkDIWoxZSCbb3pKXMji9l89exShfHp0bo6jK188TYxapy3bjmkzE6JhF22zDVJbhVNtsw+Tzd5yUOt6IH3fTMYvUmEPKODEtJh8cfn+drhmc48AwUV6RHdD09agas2iCngWAEIIFgBCCBYAQggWAkPUY4PSx/M23DkrhZfmkHkV5UDPP9pO05hOuqgYzZ9KhK5K2UgOVdffXDWbND5i2TZqKDlxG5jFUSdZ0aDjo2GX7JgOD47x90tRUagBzHBiUrhpU7P6apR8zHXienRvS7Hkqp7Q3+BjRswAQQrAAELIepyFTRd3MypqZ83kXyX0dxMGqPIy26vIz6vIylmXZeRzHoUlOR5P8jbYieQupXI1lmZ5+RHIyavfRMU+FngWAEIIFgBCCBYAQggWAkPUY4Mwy2amd/YV/fLcv3XpmJlfC+nv7xW/mC9dUDTqm8ij6Zw9fc6EuByAbxvIn+mfavbTR8u5N6y9G9U93+/vRqPhNw4lb/dMdi8wk2tY0b2FqsNNuu6nIwGNV2wanmn0GqpYCoPgNgCNHsAAQQrAAEEKwABBCsAAQQrAAEEKwABCyHnkWZvKtDWm3XLA3m1kSsLz26HxeRdUkrlSeRaqWQt2Eq+i1+Gi+xLxRcKLXsvIsjrVgb8P6FJ0L9iY+7W3zLLoXEk6/Dtmw4nXNGn4GRt3yVOhZAAghWAAIIVgACCFYAAghWAAIWY+rIWOX3nxbVlTvVq8n9fdmroBMqntPYtv8lY6qKx/lqx3Vs1IPnz1aNzJeV85u/upK26sV0e3altVLXaVpu4TfwfYNqns3KauXt79aMZXaPnIFq6rN3dsVqCqeLT6/Bbab2UfV6011bwBHjWABIIRgASCEYAEghGABIGQ9roZM9SbN8a3FlVzm54qklK+AHOUCQHVzSo578Z5lzz05DtfkIkMr+sZM2xaZR1Kr6VyS+c07bQ3ghkGwABBCsAAQsh5jFlkmnbpZ6k9S4aw/1Oj8TTPrhmS7e7K9SdZl/9z2zOZV2Yx5//AMzdS6HnVjAtEMy3fPtYvDqTobU/mS6lnsnu24NkeDw8720o8p65/pdsyptjVtz9TwTGwtmTqep4/LBovvi51tlm7rVfUsevHXlJ4FgBCCBYAQggWAEIIFgBCCBYCQTldDzOxVSX+RNJK05+4XzeyspP+Q9A+SXpV0v7v/X7dmAli1o+hZfMLdL7j7xeL3RyQ97+53Snq++B3ANW4ZpyH3SXqy+PlJSZ9ZwnMAOGZdg4VL+qGZ/dzMLhW3nXf314uf/yDpfMfnALAGumZwftzdr5jZ30r6kZn9d/lOd3czq0wRK4LLJUnazk92bAaAZevUs3D3K8X/VyU9I+kjkt4ws9skqfj/as22j7n7RXe/uJmf6NIMAMegdbAws5vM7OT0Z0mfkvSypGclPVg87EFJ3+/aSACr1+U05LykZ8xsup/vuvsPzOxnkp42sy9I+q2k+7s3E8CqtQ4W7v5rSf9Ycfv/Svpko52ZpM2N/Vmn0qQy1njjoHRRtlteRX129lxVlaq8v3DTbDsTfaq6ilLhdT2Cs0fnjTdjj8s7rlVRp20FrqnxYpGzWk1neXZtm6eqTO21XIslMGv0UL3ArNXBYhmvrGHlqxGrqAM4DgQLACEECwAhBAsAIQQLACEECwAhBAsAIQQLACEECwAh67FuiLvsL+9Ixfqk3suV7e7NZm32sv11RObXBKnK1hz3DrLVqtYVSWVi1t1ftS9pMbuw7boe0czMJuuElqUyS9uundFm+yYZmZ51b1sqQzPSnqrHVK3p0UhFduaC4iHlbNHRsNnfeqvKNK2eFF6JngWAEIIFgBCCBYAQggWAEIIFgBCCBYAQggWAEIIFgBCCBYAQggWAEIIFgBCCBYAQggWAkPWYdVrwXjG1bnNx8Yn5dUSS+yrNDq1aV6StupmJdeuMLMuy1yU5Dk1mnTZZj6St0KzT1NojS7I/2zSyxkiNpuuMLGzfaWsANwyCBYAQggWAEIIFgBCCBYAQggWAEIIFgJD1yLPIMvnJE9KgKG09GGrv3E3yUoXurD+SjSbXmAenZpMFqipdp6pr908fHifrqmfng0M327d7tl0eRDT3oHOl6xq757pdi/cGVcezYCXzqeG5jgddVd26bK/d387tc++22m4qy9K5E8Ph4gt7+7k3Gz3PaLx4fG/0RuHt6VkACCFYAAghWAAIIVgACCFYAAghWAAIIVgACCFYAAghWAAIIVgACCFYAAghWAAIIVgACCFYAAghWAAIWY96FpJ8sycbHBQ48J5ptHkQy7L+wbz7+VoT443F2hGpehZ19SoO26ck5YNYvYcma2LMPm/sccuqZ9GkHkXl9k3W1Rg2rPmRqkeRYBuH143wlvUsIvUoDrO1kX4zq+pZbGbxWhSS9G5FPYsm6FkACCFYAAghWAAIIVgACEkGCzN73MyumtnLpdvOmtmPzOxXxf9nitvNzL5pZpfN7CUz+9AyGw/g+ER6Fk9IunvutkckPe/ud0p6vvhdku6RdGfx75KkR4+mmQBWLRks3P3Hkv40d/N9kp4sfn5S0mdKt3/HJ34i6bSZ3XZUjQWwOm3HLM67++vFz3+QdL74+XZJvy897rXiNgDXuM4DnO7ukhpny5jZJTN7wcxeGIze6doMAEvWNli8MT29KP6/Wtx+RdL7So97b3HbAnd/zN0vuvvFzfxEy2YAOC5t072flfSgpK8V/3+/dPvDZvaUpI9KerN0ulLPXfb2u1JepLTmmbL+aCbF2/NsfznD+VTuqtRuzw5SiSuXN0wsQ1iX1l2XDj3/HG3TsaPbtU0nT23XdEnBBQ1SuJscg+dqvbzg/j5S21ugg1yRcl6Vit1EZPu8eN5yavm7e8G5AdN9VKSlW4OM+2SwMLPvSbpL0i1m9pqkL2sSJJ42sy9I+q2k+4uHPyfpXkmXJb0j6fPxpgBYZ8lg4e6fq7nrkxWPdUkPdW0UgPVDBieAEIIFgBCCBYAQggWAEIIFgBCCBYAQggWAEIIFgJC1qe4tSconscs3F5s1X+07pZx+XVepu426quBH+RwRy64efhyaVBJvVDW8rUD18FSF8GWZpnlHKoHXqaoGbg3mgK5HsMgy+c1/IxtMXggb7Gl4/oQ8P/gC5v2xsr3Jgb1zcvZTVvWhS5Xs758+/Mtd90GOzp3on2n34Y4GgWUtBTA813HHTcr1N5zrceLWvzZszKw8P/yLPhy2+zrccesfW2031bN0AKqaB/Lhs79r9Dz98eLxvZTHJwNxGgIghGABIIRgASCEYAEghGABIIRgASCEYAEghGABIIRgASCEYAEghGABIIRgASCEYAEghGABIIRgASBkPepZSBpv95QPDmopeG4ab5brWah039y2FQVdUmuZpgqvVEz9lxSvZ9F6LdJokZe9JRXbaVKPokKT4jDJtUfnpOpRpGz1Fou/lLWtZxGpR3GYmzf6ycdU1bPYqChmc5iqehZN0LMAEEKwABBCsAAQQrAAEEKwABBCsAAQQrAAELIeeRbuyt/alfeK2JVlyvvjmdyKcc/21xGZXxOkKqdiXFpzpCqnIpUvUXd/Xf7E/HO0XtcjmD/ROo8jtV3D3IeF/TfZ3hrkdOTeOg9iKrX9dCGfQ5tRketRlQPRRGT7jXySU1HO6Xh7b6vR82xVfCizBu8BPQsAIQQLACEECwAhBAsAIQQLACEECwAhBAsAIQQLACHrkZQ1lU1i13h7sVnzxXBSyklSVcVx2qormtOxrkjzdiy7uM5xaFBop0lRnbYixXVSBXSWZZqMFSmUU6eqWI6JpCwAR4xgASCEYAEghGABIGQtBjg9M412tpXtTmbFZbt7Gr5nW54dDGhmA5cVk+b6O7aw/bx8ePjAzWAn0aaagcxode/hmXYDch4c9LPBcqp7b597t9P2kZmbU8NhosT6nDtu/WPT5sxIVeFuO3v0Y7f8ptV2U1uWnqL89mhxhuknd15p9Dy7vnh8T2XxAVN6FgBCCBYAQggWAEIIFgBCksHCzB43s6tm9nLptq+Y2RUze7H4d2/pvi+Z2WUz+6WZfXpZDQdwvCI9iyck3V1x+zfc/ULx7zlJMrMPSHpA0geLbb5lZs2GvAGspWSwcPcfS/pTcH/3SXrK3fvu/htJlyV9pEP7AKyJLmMWD5vZS8Vpypnittsl/b70mNeK2wBc49oGi0cl3SHpgqTXJX296Q7M7JKZvWBmLwz33mnZDADHpVWwcPc33H3k7mNJ39bBqcYVSe8rPfS9xW1V+3jM3S+6+8WN3ok2zQBwjFoFCzO7rfTrZyVNr5Q8K+kBM9sys/dLulPST7s1EcA6SM4NMbPvSbpL0i1m9pqkL0u6y8wuSHJJr0r6oiS5+ytm9rSkX0jak/SQu6+mWgiAI5UMFu7+uYqb//2Qx39V0le7NArA+iGDE0AIwQJAyFrUs5CZxhv5fj0LqahxsTlbz6J8X9l4c3GXeaLuRF29iv191hS2zYbBVc4bFKOd0QvWgxgsJzG2ST2KKlsb8eXjm9azSNWjSEkVu21bzyJSj+IwJ/Pd5GOq6llsW7C4SqGqnkUT9CwAhBAsAIQQLACEECwAhBAsAIQQLACEECwAhBAsAISsRVKWjV29v/TlvUns8l6mbOCziVi9g2Ss+QWEqhKwxqV8n6pFiFKLBdUlX9WtXj6f5NV6EaBoslXLnKxUsljTRKku2+cNEteybNw6aWoqtf1Gnp7zWJUYVpUw1URk++kK6OUEsD+Pm5V2qEriYhV1AEduLXoWU9OexXhj8a/TfPp3cl+l3kRVOnhbdWnidenhy7L0dPJj0CS1vEkaeVuRdPJUyviyTHsUkdTwOlU9i8zoWQA4YgQLACFrcRrimWnv5Jay4WQQJxuONLx5a2YwMRu6bDTpMg1Ozsa4qlOD1ADm4FSi+5VV32/BWad2dhB63MLT1jzvvNFwOXH+9nNvdtp+M4sXRms6YPnhs79r2pwZG4m2vb3XbqCy6Wrm8yKzR6sGM//5RLNTkqEvfrZuajAOT88CQAjBAkAIwQJACMECQAjBAkAIwQJACMECQAjBAkAIwQJACMECQAjBAkDIWswNkSbzQabTyj3PZCOXlVL5Pbf9uSLzM4mrZhaX55VUzR2x1BSGUU3xm7raLnNzOrxm+5RRcDtrOUU9NfdkNO729+PdBtvnDaaob2Yj9cfdPq6p7bey9DT4qvklXVf6imw/nT9SnkdSNdfjMBu2+OE1xT+n9CwAhBAsAIQQLACEECwAhBAsAISYNxxRXUojzP5H0l8l/XHVbTlmt4hjvhGs8zH/vbvfGnngWgQLSTKzF9z94qrbcZw45hvD9XLMnIYACCFYAAhZp2Dx2KobsAIc843hujjmtRmzALDe1qlnAWCNESwAhBAsAIQQLACEECwAhPw/LvSDOWuiMpAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5148dba8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.818037509918213 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1.0, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5143df60>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.862872838973999 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 10.0, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f51373f60>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8323884010314941 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 100.0, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f514cb7f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8291311264038086 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1000.0, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f514964a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.9067535400390625 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 10000.0, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5139afd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.953721284866333 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 100000.0, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5151f978>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.846060037612915 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1000000.0, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5147a9e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.7710473537445068 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 10000000.0, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5155af98>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8904602527618408 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 100000000.0, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f51545da0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8758275508880615 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 1000000000.0, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f515ad160>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 1.8019354343414307 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 1.0, 'lmda': 10000000000.0, 'kernel_type': 'branching'} is: \n",
      "[[  5.   6.   4. ...  20.  20.  20.]\n",
      " [  6.   8.   4. ...  20.  20.  20.]\n",
      " [  4.   4.   5. ...  21.  21.  21.]\n",
      " ...\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]\n",
      " [ 20.  20.  21. ... 101. 101. 101.]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f513b7320>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 35.20880174636841 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 2.0, 'lmda': 1e-10, 'kernel_type': 'untiln'} is: \n",
      "[[ 13.          14.           4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [ 14.          20.           4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [  4.           4.          13.         ...  29.          29.\n",
      "   29.        ]\n",
      " ...\n",
      " [ 20.          20.          29.         ... 365.00000001 365.00000001\n",
      "  365.00000001]\n",
      " [ 20.          20.          29.         ... 365.00000001 365.00000001\n",
      "  365.00000001]\n",
      " [ 20.          20.          29.         ... 365.00000001 365.00000001\n",
      "  365.00000002]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f51439b38>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 34.336108684539795 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 2.0, 'lmda': 1e-09, 'kernel_type': 'untiln'} is: \n",
      "[[ 13.          14.           4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [ 14.          20.00000001   4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [  4.           4.          13.         ...  29.          29.\n",
      "   29.        ]\n",
      " ...\n",
      " [ 20.          20.          29.         ... 365.00000015 365.00000015\n",
      "  365.00000015]\n",
      " [ 20.          20.          29.         ... 365.00000015 365.00000015\n",
      "  365.00000015]\n",
      " [ 20.          20.          29.         ... 365.00000015 365.00000015\n",
      "  365.0000002 ]]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQsAAAECCAYAAADpWvKaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAHnlJREFUeJzt3c2P5OZ9J/Dv7ynWW79Mz4znVdI4HikTK7LsVZyB7YNhKMgifjmsnIsh72GFwIBysP8A++RcAuQSBAg2MaBgBSuH2KuLYR2EJIYuPiyytrIwHDlerweJHGk8Uvf09Et1V3V1FfnbA8kudhVZ/JGs7qqZ+X6AQfewWOTDl/7xIfk8v0dUFUREedy8C0BE9wcGCyIyYbAgIhMGCyIyYbAgIhMGCyIymXuwEJHPicgvROSWiHx93uU5KSLytoj8i4j8RETejKadF5EfiMgvo5/n5l3OKkTkZRFZF5G3EtNSt1FCfxkd95+KyMfnV/LyMrb5T0TkdnSsfyIiX0h89o1om38hIp+dT6nLmWuwEJEagL8C8HkATwH4sog8Nc8ynbDfU9VnVPVm9P+vA3hDVW8AeCP6//3s2wA+NzYtaxs/D+BG9O9FAN86pTLO2rcxuc0A8BfRsX5GVV8HgOjcfh7AR6Lv/HX0N3BfmHfN4hMAbqnqv6nqIYDvAnhuzmU6Tc8BeCX6/RUAX5xjWSpT1R8CuDc2OWsbnwPwtxr6JwBnReTq6ZR0djK2OctzAL6rqn1V/XcAtxD+DdwX5h0sHgXwTuL/70bTHkQK4B9F5J9F5MVo2mVVvRP9/h6Ay/Mp2onK2sYH/dh/Lbq9ejlxe3lfb/O8g8XD5NOq+nGE1e+vishnkh9q2O7+gW57/zBsY+RbAJ4A8AyAOwD+fL7FmY15B4vbAK4l/v9YNO2Bo6q3o5/rAL6HsPr5flz1jn6uz6+EJyZrGx/YY6+q76uqr6oBgL/B6Fbjvt7meQeLHwO4ISLXRaSB8OHPa3Mu08yJyLKIrMa/A/gDAG8h3NYXotleAPD9+ZTwRGVt42sA/lv0VuRTAHYStyv3tbFnL3+I8FgD4TY/LyJNEbmO8OHuj067fGV581y5qg5F5GsA/gFADcDLqvqzeZbphFwG8D0RAcJ9/neq+vci8mMAr4rIVwD8CsCX5ljGykTkOwCeBXBBRN4F8E0Af4b0bXwdwBcQPuTrAvijUy/wDGRs87Mi8gzCW663AfwxAKjqz0TkVQD/CmAI4Kuq6s+j3GUIu6gTkcW8b0OI6D7BYEFEJgwWRGTCYEFEJgwWRGRyYsGiaG/SRBPohwa3+eHwoGzziQSLkr1JH4gdWhC3+eHwQGzzSdUsHvbepEQPnJNqwZnWu+6TWTM3pKUtt4wzcv6ohZg4B4iE/wDocIhjnyU5iT6Qo0k6GGIaqeWkEXCSOlmHtgZ3E2VM0XIrWPMuHm8VZ/heWI7p21eW1CpeP3LK36qtYq15Odxm4760LjuXCJB+WENFyxPL2Wet2grWGpeyWz+KYHrBAKQd76LHKpgsQi/o4DA4yFl5aG7NvaP7uBcBoIUlfFo/D9duhR86h6DXCxvLRtvnltuACKTdgn9v+/jCdOwnAHdmaXKdrdbR78H2TqlyuzOtjA/k2Dr8re30+WbErU5uXxXSagIAgp3dmS53Qj2xzqx9mUJardLHzLyONUN55PgfqLSaJ14uAJCl9tH6YlrwWCXP/9j/uvuq+fsndRuS27tOVV9S1ZuqerOOJohosZ1UzeKoNynCIPE8gP+a+62omimNOtDrHf9MBNJoTER2C2k2wp9Vq9hA5u0J6o1oZaYa3dzE+2JC1Sp+GUX21WmUz3BuSaN+fELW+TBjEu+rKvuh4rl5IsGiaG9ScQ5uaRlBtxtO6PVQO3s2PBBeWETd2YX2etDhEN6lC2MLiHaCN9oc7XRGn0f3e0Hv4GhS7ZHpSam0nrFrOvvTv3d4CABwHyyZAClrvWOk0y23/CzRPpLr13JmnE7rBVJK7vXy54kNfehvfahweY5xgE65YNT2+rmLmLjrH/oIPnK9UrFUBFqb/ofs7R1G6wtGq37sfKH1uP7kcw/dqafMmVGGQmsrIEpS+rp1fmk0gG7iD8AJpD7akGMHaTxCjkd70wpzomzVGkLZ78+5ZqIV11/k+4XXVLFyUSiQFVlu1X3mldywU64MzjWfxZFEDeJIXKNYWw1/3t3M/n70FFvXVo4mBe/lJJ3KefKtKxkPEDt705cbcf1D03zjgnbGbcIYPaGHaq5V7fmRf3Ylf6bYpv0hsAJwXsU/ducQnGlnf74+5Rybwqsa4J3AX53+cFXu3J2YVg/OFlqNbE0+EJXDgfn7bO5NRCYLkfzmjJzXT7r/DNeOor4I4Ee1hehe2q2ujmog/fR7y2TbCreauMKlPdM4zLnyZ7RjkJXl9Pnj9iDR1U/69ohtWe+4zJpPnqyrYHzVLnClSV18gbYKujzlKj8+b70GOZxB25Ip5QuWDa9Oxy6vWnNwhzNIdpV4FpEmWAlrnEHilqXoeoPm5I3Ej//PX2G3c3ux21mkik5kaTSg429DEs8wTOEt+UdR5pmGZbkJmQ9ET0rZqu9pl3OKIvf6KlL8GUdRhnr2ST33yF1vvK+q3PJUvI9YnDMHgMSNssRNtlBMPMMIijZ4SnmmoRvl7k+lnX41FD+8MgTROvS9jVLLt3JlA2BGy9YgasQmv67WmKxIPdUVOPH9syulnylYueCMYabjf3HBmXbq84RZ8wZh2fy10fnnvVPsHNNzk9snOTWaY2UotLYTIs7Brawca5npXbpwvPl2rwcFEGxtw117JH+hvZRbleTJ9psfzClURg2iN/32xUWvVgdPPz59+VX1qt0ujIur+P3fqVjuAhe+2oG9Gu36Q+x/6okSBbIrUp6j7wwCdD5T7dWppbZQ6wfRz1E43n6y2HqT3435v7aHAD7gJCITBgsiMmGwICKThXhmAUR9FvYTLThFwrcYlldx8cO+Il2MT7oFZ1mn1NcgU8XV6wJ3jQkaAvFPoKlAxXMl8AAp8fb1tPc1axZEZLIYNQvnIJnNvaNXkdOaN8evRpcSTZXHc14Axxo85b0y8lfTmz17GzmvFqPGZK5kAyJ/xdbc2+3OuCNZVO5a195QKs1w2X5KeVsFtmHoo27cN1nUCfx29vWxebdAx7ZYEKDRqtj2QgTDpenX7fb7YSfI5HkbeMUa5jU6k2/Q3MD+6pQ1CyIyYbAgIpOFuA3R4RDD9btwy6NqVZyPIu49Wjt3dvQgKa3BFXD81uNsTmu8nByd3p30KnLwgYzljj3kKvsgrb4xPV9GbHB5rdTyc1V8wNrYzs8JEetfWbUvWGbzQK++l33cu48V6DGboDNoAe51p98O7F1L9JuK11twfxycn+z74v+Lvb7AmgURmTBYEJHJQtyGjJNmY7KrdpF2F+PKtMPILFxWr9Ooa3qBjjnzoI3064MM55CqoEA1elHabwSN2dxuFhanQKiwH6ruQ9YsiMhkoWoW8bgGUnPHkusCON7uIqurclbimJR2GM7YXmKijFk1hyiJULAUtgXw1kuOv5Gx3nGuXbaLevrlxV8Ol2d9wJqpQO2tViA3xHDZK9cOYlyQXfOz5NfQsQfAftuhuWl/qJslt0aq4d/GcHm0z1qbxVI3Hq5OnjOuQI2SNQsiMmGwICKThbgNkZpD7czKsWHgao9cDh/qxLcQUc5M3dicTFwTp+NLVuVS2lEkbz0Or1+aXqaUcSGB/Oaxtd3w9mn/yYtT58ter22+Is10LWrdcH9tf7TYWBTjsvZbGlegRbzXC7DxuxXblsj0DltFyhPzDgJs3SjXPiMpryNZfLtQS9ydb/52sab5tZS7luGP7E89WbMgIhMGCyIyYbAgIhMGCyIyYbAgIhMGCyIyYbAgIpOFaGcB58KRvvZHzXm17gEiR2N6yp3EqOjjzXKj/ydT4aXmo0g0pc5rD+C30neNt92Z+r14rFB3WK4dhG9M0da4V26U9iyuHzYy8PrVRlEfNu3Xn+a2fRtc30f9XNX0dYDfyG5X0L5XvKGF6wc4XK12zVUBgpyMge270SBDg9F5O1wq9udb35s854u0LWHNgohMGCyIyITBgohMGCyIyITBgohMFuJtiA59+HfvwbWj7MNOgE6UhKWzBwCQix8ARCDtNrSX/hQ9OQDQsSzcKb1S83ptZr31GJzPGNglXocfLldKZluzJlLpX5jM1GyS8TIg7vVoSQAzTXPHnvyme8k+aJAEisCrnluvvp993PeuGP4cxl/E+YBf7QUSgOO9SdPsPVo7Wl/ML3gK9M9O7j+/wLhNrFkQkQmDBRGZMFgQkQmDBRGZMFgQkclCvA05EqdZr6c8ohWB1r2jtw25kmNCxinndQYDwmS8LQjqYdx1s1iHqRzlvhbU0794auVOKjLIUMUxWE0MqwjG/mLcKe82nePlnTULIjJZqJpFPMgQRI6yecfUC2sHwdoKXCdjIJycgYHiAYCAURbuTFHv0YllZdRs4itz0AjL6XXK9QqNe3/mzrdc7tDJcHrNqLmdvt1Wrm9vZ+G37A0Uhk1XqlfoONfPrpkGnmHgprHd5zcEzbvVM60ne5Omrzc8r5JtOpo542SNG6xMHnv2OiWimVuImoU4B7eyDH9rFCrdBx8FROD60RW6H17x9L0NDJ5+PHU57nAUJtMGrE0OKZg3rkdWPoq8lplxjWLnxvL0GTPUrBf2GT9jaHTC7d18qmTL0IjLu0ImWMdIAYD6vmLjYyWHbDxaIeCm7d8Su7S+p9h6svo1N+8KH+8rrzuqHWx9rFhNSw4ny1mkFShrFkRkwmBBRCYMFkRkwmBBRCYMFkRkUultiIi8DaADwAcwVNWbInIewP8E8CEAbwP4kqpuVSsmEc3bLGoWv6eqz6jqzej/XwfwhqreAPBG9H8ius+dxG3IcwBeiX5/BcAXT2AdRHTKqgYLBfCPIvLPIvJiNO2yqt6Jfn8PwOWK6yCiBVC1BeenVfW2iFwC8AMR+b/JD1VVRdLbPEbB5UUAaLmVisUgopNWqWahqrejn+sAvgfgEwDeF5GrABD9XM/47kuqelNVbzakWhNjIjp5pYOFiCyLyGr8O4A/APAWgNcAvBDN9gKA71ctJBHNX5XbkMsAvidhMhgPwN+p6t+LyI8BvCoiXwHwKwBfql5MIpq30sFCVf8NwH9Kmb4J4PcLLcwJZKkNdBMjn0ejqAftMAeFW5/SVCPKouSvjPJV1DdScl4cG0V9epGyRjPPG9cjzkdh7j06ZjwTU5bWveo5FJK8XlTug2qPsbTA15u79m6eXjeA16v4PF6AYMpA7M2t4t1O6z3FQadiuRygOQPEN6IO2V5vVMbDs8VGlfc6k/kshPksiGjWGCyIyITBgohMGCyIyITBgohMGCyIyITBgohMGCyIyITBgohMFmLcEB368O9uwi0tHU2TTtiaU7d3wp9XLobjiDTqQC+9eaTbHbUAHVxem/y8PRp3wg2mt4Bs3EsfUax/IaPTW9Q47miksJLjelhbZvY+UKz13pGM8Txrh7Wpn1s1CrTK7F60X6vcQMytW6ep72WXr3fJsPFjs/QHgqBZfQyX5HggabpXwnUkxxcZtoutd7A2OVqc1u3LYM2CiEwYLIjIhMGCiEwYLIjIhMGCiEwYLIjIhMGCiEwYLIjIhMGCiEwYLIjIhMGCiEwYLIjIhMGCiEwWotfpOGk2UiZKOJbIoMBABxFtRDFxULFLJZDZKzOohx/IcAbrqFCOPIGX/kU3qN5z8kSdxm41rGO856tMduQ8WVX2Q8V9yJoFEZkwWBCRCYMFEZkwWBCRCYMFEZks1NsQaTXDX5wDhmNvPbwwR2RwZglymPFGxE9/NC3D8Em/vzzKwVnrTn+rEo+GPrGsjKffLsq5GdTD+NvolBvlPB7NPM9RzsyCst56xKN4FxnZPI3XtW/3YMn+eF69cqOcj6v3spfht4rn4AxqQPNe9Vc13pRyhesJ16GjUxit9WLrHa6mjKLu25fBmgURmSxEzUJqDm51BcHO7mja9WtQEbiotqGHYUZv+fU2+r/z+NgCwh+1bns0zU1GzPrG/tHv2x89P7VMXr+ZOl1leiRubofl3HwqIwt4jtqB8ZDMuN1BXKPY+nC160etb/++FtiGxq5i++mKNQsBagfZK1UUX369I9j/6EGVUgECaC+nphjt1truaL4zN7YKraY/mDy3pGVvKMKaBRGZMFgQkQmDBRGZMFgQkQmDBRGZMFgQkQmDBRGZMFgQkQmDBRGZMFgQkQmDBRGZMFgQkQmDBRGZMFgQkQmDBRGZLEQ+C4iDtNvQbu9oktZrUBH4Z1cAALWtTjgdmMjlEOdFGC6PNqex3Z9cz3DUd1+C6bkLhs30ONrcmd7/3/XDz8uOw6HGI9KomNFqXJzhqkg+ijRFc1RY1fcBr1stiYfmbFpjt/jy6/vAwV49f8YpVPL3g7dVi9Y3KuPO7lKh9QQp5QwG9uPNmgURmTBYEJEJgwURmTBYEJFJbrAQkZdFZF1E3kpMOy8iPxCRX0Y/z0XTRUT+UkRuichPReTjJ1l4Ijo9lprFtwF8bmza1wG8oao3ALwR/R8APg/gRvTvRQDfmk0xiWjecoOFqv4QwL2xyc8BeCX6/RUAX0xM/1sN/ROAsyJydVaFJaL5KfvM4rKq3ol+fw/A5ej3RwG8k5jv3WgaEd3nKj/gVFUFio/OIiIvisibIvLmYdDL/wIRzVXZYPF+fHsR/VyPpt8GcC0x32PRtAmq+pKq3lTVmw3XTpuFiBZI2eberwF4AcCfRT+/n5j+NRH5LoBPAthJ3K5k830Em/cg7WjIPxFgrxe26t7cBgDoI5fC4QxFUDtIb3LtbXWPfu9fWR19EA9vWB8N/eZyxh9ubh+mTu9eaqR/IVqHHw23KOXGRTYPTNy9ONu33vEgxUWaa6cp0oT74IJ9ZYNVIKjNYGDkTvY6Dy4ZDtrY1wcHAm2UPNgJyWEJ0/hXw+4Lw4PRfEvt9HM0S/NMd2LaRsM+fGFusBCR7wB4FsAFEXkXwDcRBolXReQrAH4F4EvR7K8D+AKAWwC6AP7IXBIiWmi5wUJVv5zx0e+nzKsAvlq1UES0eNiCk4hMGCyIyITBgohMGCyIyITBgohMGCyIyITBgohMGCyIyGQxsntHpBU193buKBN33MA3me3b9TPaaieydyeb5aZl//Z605voxlm6J8qYkRVcXbiSOCt4fb9c0+Q4y3YeNyjZLjvja3FW8apZw+v79nkHq/nzxFSAxk7FtuiYXj5/KX/5aRnCvc3qf0bJrN1pDlpRZu5EJvDuxnKhdRysDCam+b69vrAYwcIJZHkZwfbO0ST9rQ8BDnBe2BZeDofheb6+if1PPZG6mPrKqN9GWh+H5t1R79aN312bWqT6ufS2+oE3/aC274WBbONj5dLDez3bwQtmfOSaW+FJuP10tWBRJF1/kb4ejR1B/+nqvZO7+1N2nFd82922h6tPrefPOG0ZotjutabOc64WXry2d0YB4rMf/nmh9ewOJtfxD017/xLehhCRCYMFEZkwWBCRCYMFEZkwWBCRCYMFEZkwWBCRCYMFEZkwWBCRCYMFEZkwWBCRCYMFEZkwWBCRCYMFEZkwWBCRyWLks4BAGnVocihGFya8gcuPZ0EjGqfTFUiOkjdr1TwrZb9fPb9LNRXXn5YcZlHUPB++zP6Ud1ItB0jTG5ba7Q7Vx34tYjGChRPAO55sRmvhWRecCUdYd7vZiU/ED3ea3x6dqfW9lGxawSgLleSMB+s30g9ffT8vw1b4uZtMSmQSTB8fd1SOvdmeKPVeuLzawelFq2mDFE/Mu5+TuMbAFw/ws9fpOsadn1DvOGzuLVUpFkTyM1Z1dtrRL6OkSrc6FwqtZ2N/MrNWb2jfpwt8HSCiRcJgQUQmDBZEZMJgQUQmDBZEZMJgQUQmi/HqdOgjuHsP0o7GNRCH2l4//H19EwAQXLsSjiMSnEHtIP29Z3JckO5jKxOfq4xem7mMcYpi8fgf4/auZOyyaNGBF73aKvlmMx6/I0/v0mwHGfJbUVuViu/uG7v2ch1csg2oBEQDAJUY12PctNejcrlfeHn9FQ/nGyXfkydsd6cPGPTI5W0AwO5a82jahVaBEZ0APL66OTHt3TrHDSGiGWOwICITBgsiMmGwICITBgsiMmGwICITBgsiMmGwICKTxWiUFZMwdkmjPtk0qEAynHGlkuNkyVhE4E3/fOZKrifIOuLzSLpTYJ2LklSn5h1vEHgSyXTSxAl2qhymtGQ5RZa3IIeAiBYdgwURmTBYEJHJQj2zkFbUScYJMDx+b5jMyVkbZHRACtKnp+Xo9A5suTQnl5Uxf3Q7GOfuLJsjM86Fmac/KHf3mlX+OPdnkbyYaeoF+jYNCub7dNvVT9d6J/v62F/JX/7EMwpfsHVvstNiYYncmmk222GezyAYlf/nm5cKreLi8uTBOfDt+3QxgoVzcKsrCLZ3jiYFH7kOFYEX9RR1h+FZLnfuovOZ68e/H83TaI16FGpK58Lm5qhX4daN6Qf4cDX9pPKbqZNH67gbBpmtJ8tV2g6mnMxJQXO2CXub98J9uP/Rg0rLOdibftInacPe69Tb9HD1qfUyRTriRKcm1y3Te3Tr3gr+y9M/rVIs1BDgV93zU+dZq4fH5RfbowDxjd98vdB6NoeT5/wvC0R33oYQkQmDBRGZMFgQkQmDBRGZ5AYLEXlZRNZF5K3EtD8Rkdsi8pPo3xcSn31DRG6JyC9E5LMnVXAiOl2WmsW3AXwuZfpfqOoz0b/XAUBEngLwPICPRN/5axEpPiYcES2c3GChqj8EcM+4vOcAfFdV+6r67wBuAfhEhfIR0YKo8sziayLy0+g25Vw07VEA7yTmeTeaRkT3ubLB4lsAngDwDIA7AP686AJE5EUReVNE3jwMskdIJ6LFUCpYqOr7quqragDgbzC61bgN4Fpi1seiaWnLeElVb6rqzYZrlykGEZ2iUsFCRK4m/vuHAOI3Ja8BeF5EmiJyHcANAD+qVkQiWgS5fUNE5DsAngVwQUTeBfBNAM+KyDMIx916G8AfA4Cq/kxEXgXwrwCGAL6qqhldl4jofpIbLFT1yymT/8eU+f8UwJ9WKRQRLR624CQiEwYLIjJZjHwWAqB+PA+CikA9FybCmZj/+LRRstzZZZ3VeSSwBeYfvitut8ps82wkuYrLbnrDWZ4iR2qw5+VI064NKm/baViQYCGAd7xVuNbCo+qvtgAArpudmCTO/jRcSmTC6k4eQBmOpmVljIoFjfTptZzcMLVBeNDdcPp8WdKS9qTxurM9670oQ5f2Tq91fm3Xvq76vmC716q0PgHg+9nReLu7XHyhnXpu4po8ThTdYcYJF/n13hoAYGNnlMDmrd61rNlTvXMwWc794P+Zvz/v6xgR3ScYLIjIhMGCiEwYLIjIhMGCiEwYLIjIhMGCiEwWo53F0Edw9x5kKeyqLiLw9g7D3+/cBQD416+Egw4NzqCWMVpY+/1RI4i9a4lu73FLHB29p3fD6Y1g2nfT17H3aE7bgCiLoJRsp9PYts3XvTLbRjxB1K6l6uXD27K3nfCv9vNnihy06jhXq94nsbOTnQ7hkcv5O3+88dRme+loAKAq4nYUWW5e/A8AwLvLZ4+mXa1vFVrH0+13Jqb9oNY1f581CyIyWYyaRWQ01qkDhscvzYHnABH4a23U+ulX1WQLzWS73rjp9nB5dNWztsScWEfGxU2jsBsPb1i2haVnHOu0bAvRrObcGrW2L9KqMk19377dw4MC6xLF9k6JFpbjpowpuruWMzYlJndfELhjQwqWlWyZmSauURz4o/L/qPNEoXVca02m0u0F/2H+PmsWRGTCYEFEJotxG1JzcCvL0J3do0nDx84DDqgHYfUrHkXde2cD208eH0U9vs0IvKWJaUmtzcOj3zd/e3rez+FS+q7xc/oyNaNnZFsfK3efcHjWVjUftmf7gLO1Hu6wMzeKPTQbt7ObPUr5uKX2Yf5Mke7GMj774Z+XKdIRB8WtzoXMzy+07COKx36+eanwaOZp8jqFxQ8zk7ce//3R/11oHTspibFf9ezJslmzICITBgsiMmGwICITBgsiMmGwICITBgsiMmGwICITBgsiMmGwICITBgsiMmGwICKTxegbEii014O0oo4XInD9sG+FbIX9RYIPXAUcoOfOZHZRb3RGAxEdnB914oj7iRyujrr31nK6JdT30tfRPzu9C/ZgJfxcDsvFYa9j6+I9WCuZCCZj8cPV8IP+oNopEexldwEf1zxjT7xysDLA7qDaIEMAsLGf3c398dXN3O87HD8vLi7vY3M4vXu5RdoAQElx4ppkN/O0vh7TrLnJ/lC1AvUF1iyIyITBgohMGCyIyITBgohMGCyIyERUZ5txqVQhRDYA7AO4O++ynLIL4DY/DBZ5m39DVS9aZlyIYAEAIvKmqt6cdzlOE7f54fCgbDNvQ4jIhMGCiEwWKVi8NO8CzAG3+eHwQGzzwjyzIKLFtkg1CyJaYAwWRGTCYEFEJgwWRGTCYEFEJv8fTiV/vJn+AcQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f513b7be0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 34.09339237213135 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 2.0, 'lmda': 1e-08, 'kernel_type': 'untiln'} is: \n",
      "[[ 13.00000004  14.           4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [ 14.          20.00000008   4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [  4.           4.          13.00000004 ...  29.00000004  29.00000004\n",
      "   29.00000004]\n",
      " ...\n",
      " [ 20.          20.          29.00000004 ... 365.00000148 365.00000148\n",
      "  365.00000148]\n",
      " [ 20.          20.          29.00000004 ... 365.00000148 365.00000148\n",
      "  365.00000148]\n",
      " [ 20.          20.          29.00000004 ... 365.00000148 365.00000148\n",
      "  365.00000202]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f513c05f8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 34.172189235687256 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 2.0, 'lmda': 1e-07, 'kernel_type': 'untiln'} is: \n",
      "[[ 13.0000004  14.          4.        ...  20.         20.\n",
      "   20.       ]\n",
      " [ 14.         20.0000008   4.        ...  20.         20.\n",
      "   20.       ]\n",
      " [  4.          4.         13.0000004 ...  29.0000004  29.0000004\n",
      "   29.0000004]\n",
      " ...\n",
      " [ 20.         20.         29.0000004 ... 365.0000148 365.0000148\n",
      "  365.0000148]\n",
      " [ 20.         20.         29.0000004 ... 365.0000148 365.0000148\n",
      "  365.0000148]\n",
      " [ 20.         20.         29.0000004 ... 365.0000148 365.0000148\n",
      "  365.0000202]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f9418cd30>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 34.612823486328125 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 2.0, 'lmda': 1e-06, 'kernel_type': 'untiln'} is: \n",
      "[[ 13.000004  14.         4.       ...  20.        20.        20.      ]\n",
      " [ 14.        20.000008   4.       ...  20.        20.        20.      ]\n",
      " [  4.         4.        13.000004 ...  29.000004  29.000004  29.000004]\n",
      " ...\n",
      " [ 20.        20.        29.000004 ... 365.000148 365.000148 365.000148]\n",
      " [ 20.        20.        29.000004 ... 365.000148 365.000148 365.000148]\n",
      " [ 20.        20.        29.000004 ... 365.000148 365.000148 365.000202]]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQsAAAECCAYAAADpWvKaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAHnlJREFUeJzt3c2P5OZ9J/Dv7ynWW79Mz4znVdI4HikTK7LsVZyB7YNhKMgifjmsnIsh72GFwIBysP8A++RcAuQSBAg2MaBgBSuH2KuLYR2EJIYuPiyytrIwHDlerweJHGk8Uvf09Et1V3V1FfnbA8kudhVZ/JGs7qqZ+X6AQfewWOTDl/7xIfk8v0dUFUREedy8C0BE9wcGCyIyYbAgIhMGCyIyYbAgIhMGCyIymXuwEJHPicgvROSWiHx93uU5KSLytoj8i4j8RETejKadF5EfiMgvo5/n5l3OKkTkZRFZF5G3EtNSt1FCfxkd95+KyMfnV/LyMrb5T0TkdnSsfyIiX0h89o1om38hIp+dT6nLmWuwEJEagL8C8HkATwH4sog8Nc8ynbDfU9VnVPVm9P+vA3hDVW8AeCP6//3s2wA+NzYtaxs/D+BG9O9FAN86pTLO2rcxuc0A8BfRsX5GVV8HgOjcfh7AR6Lv/HX0N3BfmHfN4hMAbqnqv6nqIYDvAnhuzmU6Tc8BeCX6/RUAX5xjWSpT1R8CuDc2OWsbnwPwtxr6JwBnReTq6ZR0djK2OctzAL6rqn1V/XcAtxD+DdwX5h0sHgXwTuL/70bTHkQK4B9F5J9F5MVo2mVVvRP9/h6Ay/Mp2onK2sYH/dh/Lbq9ejlxe3lfb/O8g8XD5NOq+nGE1e+vishnkh9q2O7+gW57/zBsY+RbAJ4A8AyAOwD+fL7FmY15B4vbAK4l/v9YNO2Bo6q3o5/rAL6HsPr5flz1jn6uz6+EJyZrGx/YY6+q76uqr6oBgL/B6Fbjvt7meQeLHwO4ISLXRaSB8OHPa3Mu08yJyLKIrMa/A/gDAG8h3NYXotleAPD9+ZTwRGVt42sA/lv0VuRTAHYStyv3tbFnL3+I8FgD4TY/LyJNEbmO8OHuj067fGV581y5qg5F5GsA/gFADcDLqvqzeZbphFwG8D0RAcJ9/neq+vci8mMAr4rIVwD8CsCX5ljGykTkOwCeBXBBRN4F8E0Af4b0bXwdwBcQPuTrAvijUy/wDGRs87Mi8gzCW663AfwxAKjqz0TkVQD/CmAI4Kuq6s+j3GUIu6gTkcW8b0OI6D7BYEFEJgwWRGTCYEFEJgwWRGRyYsGiaG/SRBPohwa3+eHwoGzziQSLkr1JH4gdWhC3+eHwQGzzSdUsHvbepEQPnJNqwZnWu+6TWTM3pKUtt4wzcv6ohZg4B4iE/wDocIhjnyU5iT6Qo0k6GGIaqeWkEXCSOlmHtgZ3E2VM0XIrWPMuHm8VZ/heWI7p21eW1CpeP3LK36qtYq15Odxm4760LjuXCJB+WENFyxPL2Wet2grWGpeyWz+KYHrBAKQd76LHKpgsQi/o4DA4yFl5aG7NvaP7uBcBoIUlfFo/D9duhR86h6DXCxvLRtvnltuACKTdgn9v+/jCdOwnAHdmaXKdrdbR78H2TqlyuzOtjA/k2Dr8re30+WbErU5uXxXSagIAgp3dmS53Qj2xzqx9mUJardLHzLyONUN55PgfqLSaJ14uAJCl9tH6YlrwWCXP/9j/uvuq+fsndRuS27tOVV9S1ZuqerOOJohosZ1UzeKoNynCIPE8gP+a+62omimNOtDrHf9MBNJoTER2C2k2wp9Vq9hA5u0J6o1oZaYa3dzE+2JC1Sp+GUX21WmUz3BuSaN+fELW+TBjEu+rKvuh4rl5IsGiaG9ScQ5uaRlBtxtO6PVQO3s2PBBeWETd2YX2etDhEN6lC2MLiHaCN9oc7XRGn0f3e0Hv4GhS7ZHpSam0nrFrOvvTv3d4CABwHyyZAClrvWOk0y23/CzRPpLr13JmnE7rBVJK7vXy54kNfehvfahweY5xgE65YNT2+rmLmLjrH/oIPnK9UrFUBFqb/ofs7R1G6wtGq37sfKH1uP7kcw/dqafMmVGGQmsrIEpS+rp1fmk0gG7iD8AJpD7akGMHaTxCjkd70wpzomzVGkLZ78+5ZqIV11/k+4XXVLFyUSiQFVlu1X3mldywU64MzjWfxZFEDeJIXKNYWw1/3t3M/n70FFvXVo4mBe/lJJ3KefKtKxkPEDt705cbcf1D03zjgnbGbcIYPaGHaq5V7fmRf3Ylf6bYpv0hsAJwXsU/ducQnGlnf74+5Rybwqsa4J3AX53+cFXu3J2YVg/OFlqNbE0+EJXDgfn7bO5NRCYLkfzmjJzXT7r/DNeOor4I4Ee1hehe2q2ujmog/fR7y2TbCreauMKlPdM4zLnyZ7RjkJXl9Pnj9iDR1U/69ohtWe+4zJpPnqyrYHzVLnClSV18gbYKujzlKj8+b70GOZxB25Ip5QuWDa9Oxy6vWnNwhzNIdpV4FpEmWAlrnEHilqXoeoPm5I3Ej//PX2G3c3ux21mkik5kaTSg429DEs8wTOEt+UdR5pmGZbkJmQ9ET0rZqu9pl3OKIvf6KlL8GUdRhnr2ST33yF1vvK+q3PJUvI9YnDMHgMSNssRNtlBMPMMIijZ4SnmmoRvl7k+lnX41FD+8MgTROvS9jVLLt3JlA2BGy9YgasQmv67WmKxIPdUVOPH9syulnylYueCMYabjf3HBmXbq84RZ8wZh2fy10fnnvVPsHNNzk9snOTWaY2UotLYTIs7Brawca5npXbpwvPl2rwcFEGxtw117JH+hvZRbleTJ9psfzClURg2iN/32xUWvVgdPPz59+VX1qt0ujIur+P3fqVjuAhe+2oG9Gu36Q+x/6okSBbIrUp6j7wwCdD5T7dWppbZQ6wfRz1E43n6y2HqT3435v7aHAD7gJCITBgsiMmGwICKThXhmAUR9FvYTLThFwrcYlldx8cO+Il2MT7oFZ1mn1NcgU8XV6wJ3jQkaAvFPoKlAxXMl8AAp8fb1tPc1axZEZLIYNQvnIJnNvaNXkdOaN8evRpcSTZXHc14Axxo85b0y8lfTmz17GzmvFqPGZK5kAyJ/xdbc2+3OuCNZVO5a195QKs1w2X5KeVsFtmHoo27cN1nUCfx29vWxebdAx7ZYEKDRqtj2QgTDpenX7fb7YSfI5HkbeMUa5jU6k2/Q3MD+6pQ1CyIyYbAgIpOFuA3R4RDD9btwy6NqVZyPIu49Wjt3dvQgKa3BFXD81uNsTmu8nByd3p30KnLwgYzljj3kKvsgrb4xPV9GbHB5rdTyc1V8wNrYzs8JEetfWbUvWGbzQK++l33cu48V6DGboDNoAe51p98O7F1L9JuK11twfxycn+z74v+Lvb7AmgURmTBYEJHJQtyGjJNmY7KrdpF2F+PKtMPILFxWr9Ooa3qBjjnzoI3064MM55CqoEA1elHabwSN2dxuFhanQKiwH6ruQ9YsiMhkoWoW8bgGUnPHkusCON7uIqurclbimJR2GM7YXmKijFk1hyiJULAUtgXw1kuOv5Gx3nGuXbaLevrlxV8Ol2d9wJqpQO2tViA3xHDZK9cOYlyQXfOz5NfQsQfAftuhuWl/qJslt0aq4d/GcHm0z1qbxVI3Hq5OnjOuQI2SNQsiMmGwICKThbgNkZpD7czKsWHgao9cDh/qxLcQUc5M3dicTFwTp+NLVuVS2lEkbz0Or1+aXqaUcSGB/Oaxtd3w9mn/yYtT58ter22+Is10LWrdcH9tf7TYWBTjsvZbGlegRbzXC7DxuxXblsj0DltFyhPzDgJs3SjXPiMpryNZfLtQS9ydb/52sab5tZS7luGP7E89WbMgIhMGCyIyYbAgIhMGCyIyYbAgIhMGCyIyYbAgIpOFaGcB58KRvvZHzXm17gEiR2N6yp3EqOjjzXKj/ydT4aXmo0g0pc5rD+C30neNt92Z+r14rFB3WK4dhG9M0da4V26U9iyuHzYy8PrVRlEfNu3Xn+a2fRtc30f9XNX0dYDfyG5X0L5XvKGF6wc4XK12zVUBgpyMge270SBDg9F5O1wq9udb35s854u0LWHNgohMGCyIyITBgohMGCyIyITBgohMFuJtiA59+HfvwbWj7MNOgE6UhKWzBwCQix8ARCDtNrSX/hQ9OQDQsSzcKb1S83ptZr31GJzPGNglXocfLldKZluzJlLpX5jM1GyS8TIg7vVoSQAzTXPHnvyme8k+aJAEisCrnluvvp993PeuGP4cxl/E+YBf7QUSgOO9SdPsPVo7Wl/ML3gK9M9O7j+/wLhNrFkQkQmDBRGZMFgQkQmDBRGZMFgQkclCvA05EqdZr6c8ohWB1r2jtw25kmNCxinndQYDwmS8LQjqYdx1s1iHqRzlvhbU0794auVOKjLIUMUxWE0MqwjG/mLcKe82nePlnTULIjJZqJpFPMgQRI6yecfUC2sHwdoKXCdjIJycgYHiAYCAURbuTFHv0YllZdRs4itz0AjL6XXK9QqNe3/mzrdc7tDJcHrNqLmdvt1Wrm9vZ+G37A0Uhk1XqlfoONfPrpkGnmHgprHd5zcEzbvVM60ne5Omrzc8r5JtOpo542SNG6xMHnv2OiWimVuImoU4B7eyDH9rFCrdBx8FROD60RW6H17x9L0NDJ5+PHU57nAUJtMGrE0OKZg3rkdWPoq8lplxjWLnxvL0GTPUrBf2GT9jaHTC7d18qmTL0IjLu0ImWMdIAYD6vmLjYyWHbDxaIeCm7d8Su7S+p9h6svo1N+8KH+8rrzuqHWx9rFhNSw4ny1mkFShrFkRkwmBBRCYMFkRkwmBBRCYMFkRkUultiIi8DaADwAcwVNWbInIewP8E8CEAbwP4kqpuVSsmEc3bLGoWv6eqz6jqzej/XwfwhqreAPBG9H8ius+dxG3IcwBeiX5/BcAXT2AdRHTKqgYLBfCPIvLPIvJiNO2yqt6Jfn8PwOWK6yCiBVC1BeenVfW2iFwC8AMR+b/JD1VVRdLbPEbB5UUAaLmVisUgopNWqWahqrejn+sAvgfgEwDeF5GrABD9XM/47kuqelNVbzakWhNjIjp5pYOFiCyLyGr8O4A/APAWgNcAvBDN9gKA71ctJBHNX5XbkMsAvidhMhgPwN+p6t+LyI8BvCoiXwHwKwBfql5MIpq30sFCVf8NwH9Kmb4J4PcLLcwJZKkNdBMjn0ejqAftMAeFW5/SVCPKouSvjPJV1DdScl4cG0V9epGyRjPPG9cjzkdh7j06ZjwTU5bWveo5FJK8XlTug2qPsbTA15u79m6eXjeA16v4PF6AYMpA7M2t4t1O6z3FQadiuRygOQPEN6IO2V5vVMbDs8VGlfc6k/kshPksiGjWGCyIyITBgohMGCyIyITBgohMGCyIyITBgohMGCyIyITBgohMFmLcEB368O9uwi0tHU2TTtiaU7d3wp9XLobjiDTqQC+9eaTbHbUAHVxem/y8PRp3wg2mt4Bs3EsfUax/IaPTW9Q47miksJLjelhbZvY+UKz13pGM8Txrh7Wpn1s1CrTK7F60X6vcQMytW6ep72WXr3fJsPFjs/QHgqBZfQyX5HggabpXwnUkxxcZtoutd7A2OVqc1u3LYM2CiEwYLIjIhMGCiEwYLIjIhMGCiEwYLIjIhMGCiEwYLIjIhMGCiEwYLIjIhMGCiEwYLIjIhMGCiEwWotfpOGk2UiZKOJbIoMBABxFtRDFxULFLJZDZKzOohx/IcAbrqFCOPIGX/kU3qN5z8kSdxm41rGO856tMduQ8WVX2Q8V9yJoFEZkwWBCRCYMFEZkwWBCRCYMFEZks1NsQaTXDX5wDhmNvPbwwR2RwZglymPFGxE9/NC3D8Em/vzzKwVnrTn+rEo+GPrGsjKffLsq5GdTD+NvolBvlPB7NPM9RzsyCst56xKN4FxnZPI3XtW/3YMn+eF69cqOcj6v3spfht4rn4AxqQPNe9Vc13pRyhesJ16GjUxit9WLrHa6mjKLu25fBmgURmSxEzUJqDm51BcHO7mja9WtQEbiotqGHYUZv+fU2+r/z+NgCwh+1bns0zU1GzPrG/tHv2x89P7VMXr+ZOl1leiRubofl3HwqIwt4jtqB8ZDMuN1BXKPY+nC160etb/++FtiGxq5i++mKNQsBagfZK1UUX369I9j/6EGVUgECaC+nphjt1truaL4zN7YKraY/mDy3pGVvKMKaBRGZMFgQkQmDBRGZMFgQkQmDBRGZMFgQkQmDBRGZMFgQkQmDBRGZMFgQkQmDBRGZMFgQkQmDBRGZMFgQkQmDBRGZLEQ+C4iDtNvQbu9oktZrUBH4Z1cAALWtTjgdmMjlEOdFGC6PNqex3Z9cz3DUd1+C6bkLhs30ONrcmd7/3/XDz8uOw6HGI9KomNFqXJzhqkg+ijRFc1RY1fcBr1stiYfmbFpjt/jy6/vAwV49f8YpVPL3g7dVi9Y3KuPO7lKh9QQp5QwG9uPNmgURmTBYEJEJgwURmTBYEJFJbrAQkZdFZF1E3kpMOy8iPxCRX0Y/z0XTRUT+UkRuichPReTjJ1l4Ijo9lprFtwF8bmza1wG8oao3ALwR/R8APg/gRvTvRQDfmk0xiWjecoOFqv4QwL2xyc8BeCX6/RUAX0xM/1sN/ROAsyJydVaFJaL5KfvM4rKq3ol+fw/A5ej3RwG8k5jv3WgaEd3nKj/gVFUFio/OIiIvisibIvLmYdDL/wIRzVXZYPF+fHsR/VyPpt8GcC0x32PRtAmq+pKq3lTVmw3XTpuFiBZI2eberwF4AcCfRT+/n5j+NRH5LoBPAthJ3K5k830Em/cg7WjIPxFgrxe26t7cBgDoI5fC4QxFUDtIb3LtbXWPfu9fWR19EA9vWB8N/eZyxh9ubh+mTu9eaqR/IVqHHw23KOXGRTYPTNy9ONu33vEgxUWaa6cp0oT74IJ9ZYNVIKjNYGDkTvY6Dy4ZDtrY1wcHAm2UPNgJyWEJ0/hXw+4Lw4PRfEvt9HM0S/NMd2LaRsM+fGFusBCR7wB4FsAFEXkXwDcRBolXReQrAH4F4EvR7K8D+AKAWwC6AP7IXBIiWmi5wUJVv5zx0e+nzKsAvlq1UES0eNiCk4hMGCyIyITBgohMGCyIyITBgohMGCyIyITBgohMGCyIyGQxsntHpBU193buKBN33MA3me3b9TPaaieydyeb5aZl//Z605voxlm6J8qYkRVcXbiSOCt4fb9c0+Q4y3YeNyjZLjvja3FW8apZw+v79nkHq/nzxFSAxk7FtuiYXj5/KX/5aRnCvc3qf0bJrN1pDlpRZu5EJvDuxnKhdRysDCam+b69vrAYwcIJZHkZwfbO0ST9rQ8BDnBe2BZeDofheb6+if1PPZG6mPrKqN9GWh+H5t1R79aN312bWqT6ufS2+oE3/aC274WBbONj5dLDez3bwQtmfOSaW+FJuP10tWBRJF1/kb4ejR1B/+nqvZO7+1N2nFd82922h6tPrefPOG0ZotjutabOc64WXry2d0YB4rMf/nmh9ewOJtfxD017/xLehhCRCYMFEZkwWBCRCYMFEZkwWBCRCYMFEZkwWBCRCYMFEZkwWBCRCYMFEZkwWBCRCYMFEZkwWBCRCYMFEZkwWBCRyWLks4BAGnVocihGFya8gcuPZ0EjGqfTFUiOkjdr1TwrZb9fPb9LNRXXn5YcZlHUPB++zP6Ud1ItB0jTG5ba7Q7Vx34tYjGChRPAO55sRmvhWRecCUdYd7vZiU/ED3ea3x6dqfW9lGxawSgLleSMB+s30g9ffT8vw1b4uZtMSmQSTB8fd1SOvdmeKPVeuLzawelFq2mDFE/Mu5+TuMbAFw/ws9fpOsadn1DvOGzuLVUpFkTyM1Z1dtrRL6OkSrc6FwqtZ2N/MrNWb2jfpwt8HSCiRcJgQUQmDBZEZMJgQUQmDBZEZMJgQUQmi/HqdOgjuHsP0o7GNRCH2l4//H19EwAQXLsSjiMSnEHtIP29Z3JckO5jKxOfq4xem7mMcYpi8fgf4/auZOyyaNGBF73aKvlmMx6/I0/v0mwHGfJbUVuViu/uG7v2ch1csg2oBEQDAJUY12PctNejcrlfeHn9FQ/nGyXfkydsd6cPGPTI5W0AwO5a82jahVaBEZ0APL66OTHt3TrHDSGiGWOwICITBgsiMmGwICITBgsiMmGwICITBgsiMmGwICKTxWiUFZMwdkmjPtk0qEAynHGlkuNkyVhE4E3/fOZKrifIOuLzSLpTYJ2LklSn5h1vEHgSyXTSxAl2qhymtGQ5RZa3IIeAiBYdgwURmTBYEJHJQj2zkFbUScYJMDx+b5jMyVkbZHRACtKnp+Xo9A5suTQnl5Uxf3Q7GOfuLJsjM86Fmac/KHf3mlX+OPdnkbyYaeoF+jYNCub7dNvVT9d6J/v62F/JX/7EMwpfsHVvstNiYYncmmk222GezyAYlf/nm5cKreLi8uTBOfDt+3QxgoVzcKsrCLZ3jiYFH7kOFYEX9RR1h+FZLnfuovOZ68e/H83TaI16FGpK58Lm5qhX4daN6Qf4cDX9pPKbqZNH67gbBpmtJ8tV2g6mnMxJQXO2CXub98J9uP/Rg0rLOdibftInacPe69Tb9HD1qfUyRTriRKcm1y3Te3Tr3gr+y9M/rVIs1BDgV93zU+dZq4fH5RfbowDxjd98vdB6NoeT5/wvC0R33oYQkQmDBRGZMFgQkQmDBRGZ5AYLEXlZRNZF5K3EtD8Rkdsi8pPo3xcSn31DRG6JyC9E5LMnVXAiOl2WmsW3AXwuZfpfqOoz0b/XAUBEngLwPICPRN/5axEpPiYcES2c3GChqj8EcM+4vOcAfFdV+6r67wBuAfhEhfIR0YKo8sziayLy0+g25Vw07VEA7yTmeTeaRkT3ubLB4lsAngDwDIA7AP686AJE5EUReVNE3jwMskdIJ6LFUCpYqOr7quqragDgbzC61bgN4Fpi1seiaWnLeElVb6rqzYZrlykGEZ2iUsFCRK4m/vuHAOI3Ja8BeF5EmiJyHcANAD+qVkQiWgS5fUNE5DsAngVwQUTeBfBNAM+KyDMIx916G8AfA4Cq/kxEXgXwrwCGAL6qqhldl4jofpIbLFT1yymT/8eU+f8UwJ9WKRQRLR624CQiEwYLIjJZjHwWAqB+PA+CikA9FybCmZj/+LRRstzZZZ3VeSSwBeYfvitut8ps82wkuYrLbnrDWZ4iR2qw5+VI064NKm/baViQYCGAd7xVuNbCo+qvtgAArpudmCTO/jRcSmTC6k4eQBmOpmVljIoFjfTptZzcMLVBeNDdcPp8WdKS9qTxurM9670oQ5f2Tq91fm3Xvq76vmC716q0PgHg+9nReLu7XHyhnXpu4po8ThTdYcYJF/n13hoAYGNnlMDmrd61rNlTvXMwWc794P+Zvz/v6xgR3ScYLIjIhMGCiEwYLIjIhMGCiEwYLIjIhMGCiEwWo53F0Edw9x5kKeyqLiLw9g7D3+/cBQD416+Egw4NzqCWMVpY+/1RI4i9a4lu73FLHB29p3fD6Y1g2nfT17H3aE7bgCiLoJRsp9PYts3XvTLbRjxB1K6l6uXD27K3nfCv9vNnihy06jhXq94nsbOTnQ7hkcv5O3+88dRme+loAKAq4nYUWW5e/A8AwLvLZ4+mXa1vFVrH0+13Jqb9oNY1f581CyIyWYyaRWQ01qkDhscvzYHnABH4a23U+ulX1WQLzWS73rjp9nB5dNWztsScWEfGxU2jsBsPb1i2haVnHOu0bAvRrObcGrW2L9KqMk19377dw4MC6xLF9k6JFpbjpowpuruWMzYlJndfELhjQwqWlWyZmSauURz4o/L/qPNEoXVca02m0u0F/2H+PmsWRGTCYEFEJotxG1JzcCvL0J3do0nDx84DDqgHYfUrHkXde2cD208eH0U9vs0IvKWJaUmtzcOj3zd/e3rez+FS+q7xc/oyNaNnZFsfK3efcHjWVjUftmf7gLO1Hu6wMzeKPTQbt7ObPUr5uKX2Yf5Mke7GMj774Z+XKdIRB8WtzoXMzy+07COKx36+eanwaOZp8jqFxQ8zk7ce//3R/11oHTspibFf9ezJslmzICITBgsiMmGwICITBgsiMmGwICITBgsiMmGwICITBgsiMmGwICITBgsiMmGwICKTxegbEii014O0oo4XInD9sG+FbIX9RYIPXAUcoOfOZHZRb3RGAxEdnB914oj7iRyujrr31nK6JdT30tfRPzu9C/ZgJfxcDsvFYa9j6+I9WCuZCCZj8cPV8IP+oNopEexldwEf1zxjT7xysDLA7qDaIEMAsLGf3c398dXN3O87HD8vLi7vY3M4vXu5RdoAQElx4ppkN/O0vh7TrLnJ/lC1AvUF1iyIyITBgohMGCyIyITBgohMGCyIyERUZ5txqVQhRDYA7AO4O++ynLIL4DY/DBZ5m39DVS9aZlyIYAEAIvKmqt6cdzlOE7f54fCgbDNvQ4jIhMGCiEwWKVi8NO8CzAG3+eHwQGzzwjyzIKLFtkg1CyJaYAwWRGTCYEFEJgwWRGTCYEFEJv8fTiV/vJn+AcQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f941e0f28>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 35.64344334602356 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 2.0, 'lmda': 1e-05, 'kernel_type': 'untiln'} is: \n",
      "[[ 13.00004     14.           4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [ 14.          20.00008      4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [  4.           4.          13.00004    ...  29.00004     29.00004\n",
      "   29.00004   ]\n",
      " ...\n",
      " [ 20.          20.          29.00004    ... 365.00148001 365.00148001\n",
      "  365.00148   ]\n",
      " [ 20.          20.          29.00004    ... 365.00148001 365.00148001\n",
      "  365.00148   ]\n",
      " [ 20.          20.          29.00004    ... 365.00148    365.00148\n",
      "  365.00202   ]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f98ab2f60>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 35.96017336845398 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 2.0, 'lmda': 0.0001, 'kernel_type': 'untiln'} is: \n",
      "[[ 13.0004      14.           4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [ 14.          20.0008       4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [  4.           4.          13.0004     ...  29.0004      29.0004\n",
      "   29.0004    ]\n",
      " ...\n",
      " [ 20.          20.          29.0004     ... 365.01480072 365.01480072\n",
      "  365.0148    ]\n",
      " [ 20.          20.          29.0004     ... 365.01480072 365.01480072\n",
      "  365.0148    ]\n",
      " [ 20.          20.          29.0004     ... 365.0148     365.0148\n",
      "  365.0202    ]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f9418ccc0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 34.550148725509644 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 2.0, 'lmda': 0.001, 'kernel_type': 'untiln'} is: \n",
      "[[ 13.004     14.         4.       ...  20.        20.        20.      ]\n",
      " [ 14.        20.008      4.       ...  20.        20.        20.      ]\n",
      " [  4.         4.        13.004    ...  29.004     29.004     29.004   ]\n",
      " ...\n",
      " [ 20.        20.        29.004    ... 365.148072 365.148072 365.148   ]\n",
      " [ 20.        20.        29.004    ... 365.148072 365.148072 365.148   ]\n",
      " [ 20.        20.        29.004    ... 365.148    365.148    365.202   ]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f9c3193c8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 34.316476345062256 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 2.0, 'lmda': 0.01, 'kernel_type': 'untiln'} is: \n",
      "[[ 13.04    14.       4.     ...  20.      20.      20.    ]\n",
      " [ 14.      20.08     4.     ...  20.      20.      20.    ]\n",
      " [  4.       4.      13.04   ...  29.04    29.04    29.04  ]\n",
      " ...\n",
      " [ 20.      20.      29.04   ... 366.4872 366.4872 366.48  ]\n",
      " [ 20.      20.      29.04   ... 366.4872 366.4872 366.48  ]\n",
      " [ 20.      20.      29.04   ... 366.48   366.48   367.02  ]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f54894e10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 183 built in 34.235636711120605 seconds ---\n",
      "\n",
      "gram matrix with parameters {'h': 2.0, 'lmda': 0.1, 'kernel_type': 'untiln'} is: \n",
      "[[ 13.4   14.     4.   ...  20.    20.    20.  ]\n",
      " [ 14.    20.8    4.   ...  20.    20.    20.  ]\n",
      " [  4.     4.    13.4  ...  29.4   29.4   29.4 ]\n",
      " ...\n",
      " [ 20.    20.    29.4  ... 380.52 380.52 379.8 ]\n",
      " [ 20.    20.    29.4  ... 380.52 380.52 379.8 ]\n",
      " [ 20.    20.    29.4  ... 379.8  379.8  385.2 ]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2f5483e898>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%load_ext line_profiler\n",
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import sys\n",
    "sys.path.insert(0, \"../\")\n",
    "from pygraph.utils.model_selection_precomputed import model_selection_for_precomputed_kernel\n",
    "from pygraph.kernels.treePatternKernel import treepatternkernel\n",
    "\n",
    "datafile = '../../../../datasets/acyclic/Acyclic/dataset_bps.ds'\n",
    "estimator = treepatternkernel\n",
    "param_grid_precomputed = {'h': np.linspace(1, 10, 10), \n",
    "                          'lmda': np.logspace(-10, 10, num = 21, base = 10),\n",
    "                          'kernel_type': ['untiln', 'size', 'branching']}\n",
    "param_grid = {'alpha': np.logspace(-10, 10, num = 41, base = 10)}\n",
    "\n",
    "model_selection_for_precomputed_kernel(datafile, estimator, param_grid_precomputed, param_grid, \n",
    "                                       'regression', NUM_TRIALS=30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- This is a regression problem ---\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 1e-10 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 37.36956548690796 seconds ---\n",
      "[[ 13.          14.           4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [ 14.          20.           4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [  4.           4.          13.         ...  29.          29.\n",
      "   29.        ]\n",
      " ...\n",
      " [ 20.          20.          29.         ... 365.00000001 365.00000001\n",
      "  365.00000001]\n",
      " [ 20.          20.          29.         ... 365.00000001 365.00000001\n",
      "  365.00000001]\n",
      " [ 20.          20.          29.         ... 365.00000001 365.00000001\n",
      "  365.00000002]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  91%|█████████▏| 914/1000 [00:01<00:00, 751.28it/s]\n",
      " Mean performance on train set: 5.993535\n",
      "With standard deviation: 0.356922\n",
      "\n",
      " Mean performance on test set: 7.464904\n",
      "With standard deviation: 1.718585\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:01<00:00, 795.88it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 1e-09 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 37.47467517852783 seconds ---\n",
      "[[ 13.          14.           4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [ 14.          20.00000001   4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [  4.           4.          13.         ...  29.          29.\n",
      "   29.        ]\n",
      " ...\n",
      " [ 20.          20.          29.         ... 365.00000015 365.00000015\n",
      "  365.00000015]\n",
      " [ 20.          20.          29.         ... 365.00000015 365.00000015\n",
      "  365.00000015]\n",
      " [ 20.          20.          29.         ... 365.00000015 365.00000015\n",
      "  365.0000002 ]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  98%|█████████▊| 975/1000 [00:01<00:00, 654.33it/s]\n",
      " Mean performance on train set: 5.963041\n",
      "With standard deviation: 0.374107\n",
      "\n",
      " Mean performance on test set: 7.375105\n",
      "With standard deviation: 1.769252\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:01<00:00, 711.24it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 1e-08 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 37.32968211174011 seconds ---\n",
      "[[ 13.00000004  14.           4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [ 14.          20.00000008   4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [  4.           4.          13.00000004 ...  29.00000004  29.00000004\n",
      "   29.00000004]\n",
      " ...\n",
      " [ 20.          20.          29.00000004 ... 365.00000148 365.00000148\n",
      "  365.00000148]\n",
      " [ 20.          20.          29.00000004 ... 365.00000148 365.00000148\n",
      "  365.00000148]\n",
      " [ 20.          20.          29.00000004 ... 365.00000148 365.00000148\n",
      "  365.00000202]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  97%|█████████▋| 967/1000 [00:01<00:00, 809.48it/s]\n",
      " Mean performance on train set: 5.965110\n",
      "With standard deviation: 0.378249\n",
      "\n",
      " Mean performance on test set: 7.350689\n",
      "With standard deviation: 1.780556\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:01<00:00, 786.78it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 1e-07 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 37.74151062965393 seconds ---\n",
      "[[ 13.0000004  14.          4.        ...  20.         20.\n",
      "   20.       ]\n",
      " [ 14.         20.0000008   4.        ...  20.         20.\n",
      "   20.       ]\n",
      " [  4.          4.         13.0000004 ...  29.0000004  29.0000004\n",
      "   29.0000004]\n",
      " ...\n",
      " [ 20.         20.         29.0000004 ... 365.0000148 365.0000148\n",
      "  365.0000148]\n",
      " [ 20.         20.         29.0000004 ... 365.0000148 365.0000148\n",
      "  365.0000148]\n",
      " [ 20.         20.         29.0000004 ... 365.0000148 365.0000148\n",
      "  365.0000202]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  95%|█████████▌| 954/1000 [00:01<00:00, 735.76it/s]\n",
      " Mean performance on train set: 5.966982\n",
      "With standard deviation: 0.382093\n",
      "\n",
      " Mean performance on test set: 7.350999\n",
      "With standard deviation: 1.781470\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:01<00:00, 804.24it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 1e-06 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 37.52131748199463 seconds ---\n",
      "[[ 13.000004  14.         4.       ...  20.        20.        20.      ]\n",
      " [ 14.        20.000008   4.       ...  20.        20.        20.      ]\n",
      " [  4.         4.        13.000004 ...  29.000004  29.000004  29.000004]\n",
      " ...\n",
      " [ 20.        20.        29.000004 ... 365.000148 365.000148 365.000148]\n",
      " [ 20.        20.        29.000004 ... 365.000148 365.000148 365.000148]\n",
      " [ 20.        20.        29.000004 ... 365.000148 365.000148 365.000202]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  97%|█████████▋| 970/1000 [00:01<00:00, 759.32it/s]\n",
      " Mean performance on train set: 5.969758\n",
      "With standard deviation: 0.386318\n",
      "\n",
      " Mean performance on test set: 7.351225\n",
      "With standard deviation: 1.780522\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:01<00:00, 783.42it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 1e-05 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 37.67099857330322 seconds ---\n",
      "[[ 13.00004     14.           4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [ 14.          20.00008      4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [  4.           4.          13.00004    ...  29.00004     29.00004\n",
      "   29.00004   ]\n",
      " ...\n",
      " [ 20.          20.          29.00004    ... 365.00148001 365.00148001\n",
      "  365.00148   ]\n",
      " [ 20.          20.          29.00004    ... 365.00148001 365.00148001\n",
      "  365.00148   ]\n",
      " [ 20.          20.          29.00004    ... 365.00148    365.00148\n",
      "  365.00202   ]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:01<00:00, 801.70it/s]\n",
      " Mean performance on train set: 5.970557\n",
      "With standard deviation: 0.390719\n",
      "\n",
      " Mean performance on test set: 7.348129\n",
      "With standard deviation: 1.780293\n",
      "\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 0.0001 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 36.80127692222595 seconds ---\n",
      "[[ 13.0004      14.           4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [ 14.          20.0008       4.         ...  20.          20.\n",
      "   20.        ]\n",
      " [  4.           4.          13.0004     ...  29.0004      29.0004\n",
      "   29.0004    ]\n",
      " ...\n",
      " [ 20.          20.          29.0004     ... 365.01480072 365.01480072\n",
      "  365.0148    ]\n",
      " [ 20.          20.          29.0004     ... 365.01480072 365.01480072\n",
      "  365.0148    ]\n",
      " [ 20.          20.          29.0004     ... 365.0148     365.0148\n",
      "  365.0202    ]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  98%|█████████▊| 980/1000 [00:01<00:00, 889.41it/s]\n",
      " Mean performance on train set: 5.942495\n",
      "With standard deviation: 0.331983\n",
      "\n",
      " Mean performance on test set: 7.349836\n",
      "With standard deviation: 1.781100\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:01<00:00, 883.76it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 0.001 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 35.8681423664093 seconds ---\n",
      "[[ 13.004     14.         4.       ...  20.        20.        20.      ]\n",
      " [ 14.        20.008      4.       ...  20.        20.        20.      ]\n",
      " [  4.         4.        13.004    ...  29.004     29.004     29.004   ]\n",
      " ...\n",
      " [ 20.        20.        29.004    ... 365.148072 365.148072 365.148   ]\n",
      " [ 20.        20.        29.004    ... 365.148072 365.148072 365.148   ]\n",
      " [ 20.        20.        29.004    ... 365.148    365.148    365.202   ]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  99%|█████████▉| 988/1000 [00:01<00:00, 886.54it/s]\n",
      " Mean performance on train set: 5.933395\n",
      "With standard deviation: 0.324965\n",
      "\n",
      " Mean performance on test set: 7.357745\n",
      "With standard deviation: 1.780977\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:01<00:00, 888.00it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 0.01 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 36.001843214035034 seconds ---\n",
      "[[ 13.04    14.       4.     ...  20.      20.      20.    ]\n",
      " [ 14.      20.08     4.     ...  20.      20.      20.    ]\n",
      " [  4.       4.      13.04   ...  29.04    29.04    29.04  ]\n",
      " ...\n",
      " [ 20.      20.      29.04   ... 366.4872 366.4872 366.48  ]\n",
      " [ 20.      20.      29.04   ... 366.4872 366.4872 366.48  ]\n",
      " [ 20.      20.      29.04   ... 366.48   366.48   367.02  ]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  98%|█████████▊| 978/1000 [00:01<00:00, 863.94it/s]\n",
      " Mean performance on train set: 5.940695\n",
      "With standard deviation: 0.347431\n",
      "\n",
      " Mean performance on test set: 7.374269\n",
      "With standard deviation: 1.791145\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:01<00:00, 878.96it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 0.1 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 36.37146854400635 seconds ---\n",
      "[[ 13.4   14.     4.   ...  20.    20.    20.  ]\n",
      " [ 14.    20.8    4.   ...  20.    20.    20.  ]\n",
      " [  4.     4.    13.4  ...  29.4   29.4   29.4 ]\n",
      " ...\n",
      " [ 20.    20.    29.4  ... 380.52 380.52 379.8 ]\n",
      " [ 20.    20.    29.4  ... 380.52 380.52 379.8 ]\n",
      " [ 20.    20.    29.4  ... 379.8  379.8  385.2 ]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  99%|█████████▉| 993/1000 [00:01<00:00, 860.40it/s]\n",
      " Mean performance on train set: 6.427114\n",
      "With standard deviation: 1.293674\n",
      "\n",
      " Mean performance on test set: 7.329299\n",
      "With standard deviation: 1.913634\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:01<00:00, 883.01it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 1.0 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 37.83972358703613 seconds ---\n",
      "[[ 17.  14.   4. ...  20.  20.  20.]\n",
      " [ 14.  28.   4. ...  20.  20.  20.]\n",
      " [  4.   4.  17. ...  33.  33.  33.]\n",
      " ...\n",
      " [ 20.  20.  33. ... 585. 585. 513.]\n",
      " [ 20.  20.  33. ... 585. 585. 513.]\n",
      " [ 20.  20.  33. ... 513. 513. 567.]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  98%|█████████▊| 979/1000 [00:01<00:00, 616.77it/s]\n",
      " Mean performance on train set: 6.624254\n",
      "With standard deviation: 1.224196\n",
      "\n",
      " Mean performance on test set: 7.271336\n",
      "With standard deviation: 2.207735\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:01<00:00, 630.82it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 10.0 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 37.50818395614624 seconds ---\n",
      "[[5.300e+01 1.400e+01 4.000e+00 ... 2.000e+01 2.000e+01 2.000e+01]\n",
      " [1.400e+01 1.000e+02 4.000e+00 ... 2.000e+01 2.000e+01 2.000e+01]\n",
      " [4.000e+00 4.000e+00 5.300e+01 ... 6.900e+01 6.900e+01 6.900e+01]\n",
      " ...\n",
      " [2.000e+01 2.000e+01 6.900e+01 ... 9.045e+03 9.045e+03 1.845e+03]\n",
      " [2.000e+01 2.000e+01 6.900e+01 ... 9.045e+03 9.045e+03 1.845e+03]\n",
      " [2.000e+01 2.000e+01 6.900e+01 ... 1.845e+03 1.845e+03 2.385e+03]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:   8%|▊         | 77/1000 [00:00<00:01, 764.71it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.069543502626658e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.1303298666315776e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.70249458866672e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.63992169055093e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.438093960487116e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.0002169262936346e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.5920339281975188e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.5874866272574162e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.0599424240471626e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.468773818521402e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.958334441043603e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  27%|██▋       | 267/1000 [00:00<00:01, 654.82it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/sklearn/linear_model/ridge.py:154: UserWarning: Singular matrix in solving dual problem. Using least-squares solution instead.\n",
      "  warnings.warn(\"Singular matrix in solving dual problem. Using \"\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.055618175730539e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.7159074038024934e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.895455126720251e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.400306511546424e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.206478316049589e-19 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.8083631222444177e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.49051280863482e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.339852738992424e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.277544863160196e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.458523723353626e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  48%|████▊     | 477/1000 [00:00<00:00, 685.70it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.6380760737666547e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.3843421259537676e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.890544546973404e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.054758730954765e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.9172765626494813e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.4455093698440067e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.2914256710839066e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.65667341282596e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.852926745577629e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.285092924342139e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  69%|██████▉   | 694/1000 [00:01<00:00, 712.57it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.700250453064005e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.6205193931367065e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.4925504318417794e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.0111387119813346e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.343123723749221e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.0143662852277667e-16 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.206690575125046e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.049999246995425e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.4232350203422674e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.350008400303505e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.251763015291957e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  90%|█████████ | 902/1000 [00:01<00:00, 666.85it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.924869742342744e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.4010401637647583e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.5117924740400373e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.626753798403599e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.911227588173856e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.0660043401009468e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.3099139652029694e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.2680602391853274e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.396574210735164e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  97%|█████████▋| 969/1000 [00:01<00:00, 600.11it/s]\n",
      " Mean performance on train set: 6.816974\n",
      "With standard deviation: 1.501822\n",
      "\n",
      " Mean performance on test set: 7.497870\n",
      "With standard deviation: 2.368148\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:01<00:00, 648.87it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 100.0 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 37.09455919265747 seconds ---\n",
      "[[4.13000e+02 1.40000e+01 4.00000e+00 ... 2.00000e+01 2.00000e+01\n",
      "  2.00000e+01]\n",
      " [1.40000e+01 8.20000e+02 4.00000e+00 ... 2.00000e+01 2.00000e+01\n",
      "  2.00000e+01]\n",
      " [4.00000e+00 4.00000e+00 4.13000e+02 ... 4.29000e+02 4.29000e+02\n",
      "  4.29000e+02]\n",
      " ...\n",
      " [2.00000e+01 2.00000e+01 4.29000e+02 ... 7.35165e+05 7.35165e+05\n",
      "  1.51650e+04]\n",
      " [2.00000e+01 2.00000e+01 4.29000e+02 ... 7.35165e+05 7.35165e+05\n",
      "  1.51650e+04]\n",
      " [2.00000e+01 2.00000e+01 4.29000e+02 ... 1.51650e+04 1.51650e+04\n",
      "  2.05650e+04]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  10%|▉         | 97/1000 [00:00<00:02, 436.93it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.249229588791739e-20 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.6092761314568358e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.6033357497241564e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.0333531111165975e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.073851980749357e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.497880470461594e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.0349745182117167e-16 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  17%|█▋        | 169/1000 [00:00<00:02, 372.64it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.227908145504113e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.1894008132724887e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.9131474526752795e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.0421770253846576e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.864916618602575e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.663676730244888e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  31%|███▏      | 314/1000 [00:00<00:01, 429.29it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.6989200751598342e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.9988176582222278e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.173259131422707e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.069621878854856e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.087601566853754e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.1827598831940232e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.428244298929586e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.1120965359644164e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.8329496119281176e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.5128147762765525e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.173155329882729e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  50%|█████     | 501/1000 [00:01<00:00, 516.57it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.874621241781873e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.9047959204426696e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.03000007539236e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.8198639503150797e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.679200342495213e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.6211471280327221e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.74280825574767e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.685372827008377e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.9723233156997277e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.142362330339379e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.0042133764798303e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.987833375253946e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  71%|███████   | 706/1000 [00:01<00:00, 471.97it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.63949546549065e-19 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.76170805410039e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.3721058293845662e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.178277242767302e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.4883373934010664e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.526360275338589e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.827383891217367e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  81%|████████  | 807/1000 [00:01<00:00, 457.30it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.052622499085628e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.70793549450487e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.8190124240850417e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.349104192126423e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.132340452050677e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  91%|█████████ | 909/1000 [00:01<00:00, 451.90it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.78695142234395e-20 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.8765535280551442e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.9917255115528226e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.781650263544808e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.581768670551366e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.0493867289518776e-20 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.9787087068181396e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.756012232435961e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.800283208793992e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.468606690086715e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.119459703249427e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  96%|█████████▌| 956/1000 [00:02<00:00, 445.96it/s]\n",
      " Mean performance on train set: 6.687664\n",
      "With standard deviation: 1.348089\n",
      "\n",
      " Mean performance on test set: 7.428867\n",
      "With standard deviation: 2.647892\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:02<00:00, 467.65it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 1000.0 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 37.517051219940186 seconds ---\n",
      "[[4.0130000e+03 1.4000000e+01 4.0000000e+00 ... 2.0000000e+01\n",
      "  2.0000000e+01 2.0000000e+01]\n",
      " [1.4000000e+01 8.0200000e+03 4.0000000e+00 ... 2.0000000e+01\n",
      "  2.0000000e+01 2.0000000e+01]\n",
      " [4.0000000e+00 4.0000000e+00 4.0130000e+03 ... 4.0290000e+03\n",
      "  4.0290000e+03 4.0290000e+03]\n",
      " ...\n",
      " [2.0000000e+01 2.0000000e+01 4.0290000e+03 ... 7.2148365e+07\n",
      "  7.2148365e+07 1.4836500e+05]\n",
      " [2.0000000e+01 2.0000000e+01 4.0290000e+03 ... 7.2148365e+07\n",
      "  7.2148365e+07 1.4836500e+05]\n",
      " [2.0000000e+01 2.0000000e+01 4.0290000e+03 ... 1.4836500e+05\n",
      "  1.4836500e+05 2.0236500e+05]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  10%|█         | 102/1000 [00:00<00:02, 330.46it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.374017095746491e-19 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.687507275679712e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.7645414168071277e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.8090152927008474e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.474194561968185e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.130856876335615e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  19%|█▉        | 188/1000 [00:00<00:02, 348.40it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.7794644716713837e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.3130943734340723e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.7505566440337117e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.3279670378456666e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.318466984022222e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.418445519765442e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  28%|██▊       | 285/1000 [00:00<00:01, 380.47it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.1356199672921913e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.6464620684950592e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.179430869121561e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.473887201835687e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.570987725305032e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.854640029504099e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  38%|███▊      | 385/1000 [00:00<00:01, 389.24it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.3781828327775562e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.955507306233033e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.109947315270106e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.9378280303294975e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.866942525478256e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  49%|████▉     | 494/1000 [00:01<00:01, 398.83it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.636014964778956e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.6552163232757833e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.312233993243073e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.2794244316598437e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.406028628818668e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  60%|██████    | 604/1000 [00:01<00:00, 407.30it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.184686103929999e-20 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.1281587780183657e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.367013528660628e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.155116554595105e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.443029464120917e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.4923079446940085e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.750703061909557e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  71%|███████   | 707/1000 [00:01<00:00, 402.63it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.4970615125032324e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.39494402062226e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.818999641865095e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.051275910233908e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.688197813410084e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  81%|████████  | 806/1000 [00:01<00:00, 404.11it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.43659662072146e-20 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.4284104102664825e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.900304923444742e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.1112688931900636e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.941184656304436e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.746476431972804e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  91%|█████████ | 906/1000 [00:02<00:00, 407.57it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.327409345420052e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.0589341144557062e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.29154188313992e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.225020130252359e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.327720201864263e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  95%|█████████▌| 951/1000 [00:02<00:00, 373.87it/s]\n",
      " Mean performance on train set: 6.819058\n",
      "With standard deviation: 1.410085\n",
      "\n",
      " Mean performance on test set: 7.249143\n",
      "With standard deviation: 2.655536\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:02<00:00, 414.03it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.193651783291256e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.19984276961351e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.0689487149937185e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.8956869823870564e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.790887064559792e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 10000.0 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 37.01269268989563 seconds ---\n",
      "[[4.00130000e+04 1.40000000e+01 4.00000000e+00 ... 2.00000000e+01\n",
      "  2.00000000e+01 2.00000000e+01]\n",
      " [1.40000000e+01 8.00200000e+04 4.00000000e+00 ... 2.00000000e+01\n",
      "  2.00000000e+01 2.00000000e+01]\n",
      " [4.00000000e+00 4.00000000e+00 4.00130000e+04 ... 4.00290000e+04\n",
      "  4.00290000e+04 4.00290000e+04]\n",
      " ...\n",
      " [2.00000000e+01 2.00000000e+01 4.00290000e+04 ... 7.20148036e+09\n",
      "  7.20148036e+09 1.48036500e+06]\n",
      " [2.00000000e+01 2.00000000e+01 4.00290000e+04 ... 7.20148036e+09\n",
      "  7.20148036e+09 1.48036500e+06]\n",
      " [2.00000000e+01 2.00000000e+01 4.00290000e+04 ... 1.48036500e+06\n",
      "  1.48036500e+06 2.02036500e+06]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  11%|█▏        | 114/1000 [00:00<00:02, 303.84it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.6978549111114387e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.1828172674052679e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.547017114313022e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.990786470945978e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.8300974250606965e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.727582936838558e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  21%|██        | 208/1000 [00:00<00:02, 342.60it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.569354559683504e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.885865920757663e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.006883677471783e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.2491482969368813e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.596699463334369e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.74024827114128e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.0002891123411383e-16 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  31%|███       | 311/1000 [00:00<00:01, 371.07it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.478112292477647e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.3626069490970097e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.751107853461428e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.007160521891646e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.592912372477283e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  41%|████▏     | 414/1000 [00:01<00:01, 394.40it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.286966904954778e-19 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.330065717080935e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.0639972759819077e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.3250852692883386e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.32507703774876e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.514317701906229e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  51%|█████▏    | 514/1000 [00:01<00:01, 411.17it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.5186106207732215e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.2309907116861648e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.582391203608073e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.638503426307468e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.0580872068659216e-16 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  62%|██████▏   | 615/1000 [00:01<00:00, 424.96it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.633316922861427e-19 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.253166879599146e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.4688155478710103e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.3252803222906435e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.713726552669558e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.934042177466841e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.468592067289146e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  72%|███████▏  | 716/1000 [00:01<00:00, 433.84it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.66227030386163e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.5342528466878185e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.620745355332433e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.5797394734563764e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.109079490079661e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.548426603146641e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.0484899604694826e-16 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  82%|████████▏ | 816/1000 [00:01<00:00, 439.46it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.080827240476694e-20 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.8279322282295696e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.0410721959495632e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.343375151645726e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.351177509861134e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.370583478449445e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  92%|█████████▏| 916/1000 [00:02<00:00, 437.07it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.3677978521118296e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.0227997187914302e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.557088603475233e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.657645350184021e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.990628121216557e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  96%|█████████▌| 962/1000 [00:02<00:00, 393.51it/s]\n",
      " Mean performance on train set: 6.934306\n",
      "With standard deviation: 1.384412\n",
      "\n",
      " Mean performance on test set: 7.081832\n",
      "With standard deviation: 2.624800\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:02<00:00, 420.16it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.2964047050969517e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.089708488833387e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.3234351109582e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.293818265362604e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.44555950404844e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 100000.0 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 35.835275411605835 seconds ---\n",
      "[[4.0001300e+05 1.4000000e+01 4.0000000e+00 ... 2.0000000e+01\n",
      "  2.0000000e+01 2.0000000e+01]\n",
      " [1.4000000e+01 8.0002000e+05 4.0000000e+00 ... 2.0000000e+01\n",
      "  2.0000000e+01 2.0000000e+01]\n",
      " [4.0000000e+00 4.0000000e+00 4.0001300e+05 ... 4.0002900e+05\n",
      "  4.0002900e+05 4.0002900e+05]\n",
      " ...\n",
      " [2.0000000e+01 2.0000000e+01 4.0002900e+05 ... 7.2001480e+11\n",
      "  7.2001480e+11 1.4800365e+07]\n",
      " [2.0000000e+01 2.0000000e+01 4.0002900e+05 ... 7.2001480e+11\n",
      "  7.2001480e+11 1.4800365e+07]\n",
      " [2.0000000e+01 2.0000000e+01 4.0002900e+05 ... 1.4800365e+07\n",
      "  1.4800365e+07 2.0200365e+07]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  13%|█▎        | 126/1000 [00:00<00:03, 261.95it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.0191112815027622e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.232307074954237e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.097885039345644e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.727869526025791e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  21%|██        | 208/1000 [00:00<00:03, 262.44it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.2411064955279154e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.149899322677468e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.642478117522784e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  30%|███       | 305/1000 [00:01<00:02, 265.15it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.785067039039337e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.225934530879337e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.562790184640486e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  41%|████▏     | 414/1000 [00:01<00:01, 297.05it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.135758789917749e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.564182826681079e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.542436590136228e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  50%|█████     | 500/1000 [00:01<00:01, 319.27it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.195315905739342e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.379873880613873e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.630975349505465e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.518071272961898e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  61%|██████▏   | 613/1000 [00:02<00:01, 303.76it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.9460047081015216e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.839254463570563e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.318583704180543e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  73%|███████▎  | 728/1000 [00:02<00:00, 300.34it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.3984244112232524e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.3291156014339405e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.707621088224988e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.0690013288979288e-16 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  81%|████████▏ | 813/1000 [00:02<00:00, 321.97it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.315773170165585e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.138544398203078e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.386508127676171e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  89%|████████▉ | 891/1000 [00:02<00:00, 317.18it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.6842499515474312e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.528434456947986e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.856433851414765e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  96%|█████████▌| 957/1000 [00:03<00:00, 259.36it/s]\n",
      " Mean performance on train set: 9.394995\n",
      "With standard deviation: 1.047066\n",
      "\n",
      " Mean performance on test set: 8.237631\n",
      "With standard deviation: 3.665300\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:03<00:00, 299.25it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 1000000.0 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.056791553686018e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.644703618966645e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.15242050721053e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 37.77732253074646 seconds ---\n",
      "[[4.00001300e+06 1.40000000e+01 4.00000000e+00 ... 2.00000000e+01\n",
      "  2.00000000e+01 2.00000000e+01]\n",
      " [1.40000000e+01 8.00002000e+06 4.00000000e+00 ... 2.00000000e+01\n",
      "  2.00000000e+01 2.00000000e+01]\n",
      " [4.00000000e+00 4.00000000e+00 4.00001300e+06 ... 4.00002900e+06\n",
      "  4.00002900e+06 4.00002900e+06]\n",
      " ...\n",
      " [2.00000000e+01 2.00000000e+01 4.00002900e+06 ... 7.20001480e+13\n",
      "  7.20001480e+13 1.48000365e+08]\n",
      " [2.00000000e+01 2.00000000e+01 4.00002900e+06 ... 7.20001480e+13\n",
      "  7.20001480e+13 1.48000365e+08]\n",
      " [2.00000000e+01 2.00000000e+01 4.00002900e+06 ... 1.48000365e+08\n",
      "  1.48000365e+08 2.02000365e+08]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  14%|█▎        | 135/1000 [00:00<00:03, 253.06it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.741195162637844e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.5965964498458038e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.983361443347492e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.7922291165206923e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.685526054240851e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  23%|██▎       | 230/1000 [00:00<00:02, 265.75it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.6848871984797616e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.8811603375005575e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.7514073450053307e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.5976314128410034e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.790988115471154e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.477178586927344e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.035215871851048e-16 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  33%|███▎      | 332/1000 [00:01<00:02, 277.35it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.287317576627726e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.886325173924881e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.8372812242318245e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.523017625167697e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.309735186090854e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  43%|████▎     | 431/1000 [00:01<00:02, 258.66it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.2092887442597021e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.444285941342485e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.6650957319224102e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.778343353100153e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.067496740668901e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.832366528737191e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  53%|█████▎    | 534/1000 [00:01<00:01, 252.20it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.2368084459111367e-19 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.625050966790768e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.786992563738048e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.5350081600164477e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.2823286559994256e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.880362578796432e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.1100444346816681e-16 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  64%|██████▎   | 637/1000 [00:02<00:01, 260.77it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.1335408080135885e-20 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.4688002952482946e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.6678722631357644e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.288251804550535e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.895017980474164e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  74%|███████▍  | 741/1000 [00:02<00:00, 276.66it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.209157188853246e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.4519382674684447e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.5748326244710203e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.852436562697074e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.536506069365062e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.969036612017947e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  81%|████████▏ | 813/1000 [00:02<00:00, 297.62it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.309383987394578e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.565067260805818e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.516194197490843e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.041064934861363e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.499601220689098e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.051151158798192e-16 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  92%|█████████▏| 915/1000 [00:03<00:00, 298.37it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.7472428919435347e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.703770718809819e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.2818242346374262e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.102554831016506e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.599188283622467e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.0647003359802031e-16 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  98%|█████████▊| 977/1000 [00:03<00:00, 278.57it/s]\n",
      " Mean performance on train set: 9.635245\n",
      "With standard deviation: 0.687560\n",
      "\n",
      " Mean performance on test set: 8.529828\n",
      "With standard deviation: 3.580591\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:03<00:00, 284.18it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 10000000.0 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.502409510736216e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.0439081147173944e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.4602149061556115e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.9479003259500843e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.34621741763787e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.0245542507930726e-16 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 37.46095633506775 seconds ---\n",
      "[[4.00000130e+07 1.40000000e+01 4.00000000e+00 ... 2.00000000e+01\n",
      "  2.00000000e+01 2.00000000e+01]\n",
      " [1.40000000e+01 8.00000200e+07 4.00000000e+00 ... 2.00000000e+01\n",
      "  2.00000000e+01 2.00000000e+01]\n",
      " [4.00000000e+00 4.00000000e+00 4.00000130e+07 ... 4.00000290e+07\n",
      "  4.00000290e+07 4.00000290e+07]\n",
      " ...\n",
      " [2.00000000e+01 2.00000000e+01 4.00000290e+07 ... 7.20000148e+15\n",
      "  7.20000148e+15 1.48000036e+09]\n",
      " [2.00000000e+01 2.00000000e+01 4.00000290e+07 ... 7.20000148e+15\n",
      "  7.20000148e+15 1.48000036e+09]\n",
      " [2.00000000e+01 2.00000000e+01 4.00000290e+07 ... 1.48000036e+09\n",
      "  1.48000036e+09 2.02000036e+09]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  12%|█▏        | 122/1000 [00:00<00:03, 231.16it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.200122265640733e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.627297027618617e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.893611596005168e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.1052758523976415e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.158312302718829e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.675650470006637e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  23%|██▎       | 227/1000 [00:00<00:03, 253.57it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.2230286797750079e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.0790289882373515e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.6170735381557016e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.90989130271085e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.4735492461286675e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.544548737558878e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  34%|███▍      | 344/1000 [00:01<00:02, 274.36it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.0569387542185164e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.8929851773550792e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.762490810329375e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.031994076021703e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.380795983197089e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  44%|████▎     | 437/1000 [00:01<00:02, 281.37it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.159845331824398e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.592173545119868e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.0031182975801337e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.315102488294594e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.427690586128571e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.36815306216013e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  52%|█████▎    | 525/1000 [00:01<00:01, 277.96it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.096509347533013e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.2037051514926243e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.641239932076709e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.911654384898199e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.195245830759744e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  61%|██████    | 606/1000 [00:02<00:01, 236.45it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.179739003035368e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.121167134816686e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 2.201750997217992e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.6143396276266097e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.876412427833992e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 8.158416120381036e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  73%|███████▎  | 727/1000 [00:02<00:01, 211.65it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.355737066205781e-20 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 6.778261189640049e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.992328403436096e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.166526583123927e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.566565138343654e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.272056103177315e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  82%|████████▏ | 822/1000 [00:03<00:00, 223.94it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.2435735437353417e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.471379508084743e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.686971341479104e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.827108937015577e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  93%|█████████▎| 931/1000 [00:03<00:00, 255.06it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.2357280081107672e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.559072238694825e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.45488311322705e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.6616151717441874e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.807701664283496e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "calculate performance:  98%|█████████▊| 983/1000 [00:04<00:00, 220.81it/s]\n",
      " Mean performance on train set: 11.059074\n",
      "With standard deviation: 1.323635\n",
      "\n",
      " Mean performance on test set: 10.964175\n",
      "With standard deviation: 3.358726\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:04<00:00, 242.35it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 100000000.0 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 9.749823872976888e-19 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 5.347208969568296e-18 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 1.9981807042259307e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 3.298232968104139e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 4.391381083354749e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n",
      "/home/ljia/.local/lib/python3.5/site-packages/scipy/linalg/basic.py:40: RuntimeWarning: scipy.linalg.solve\n",
      "Ill-conditioned matrix detected. Result is not guaranteed to be accurate.\n",
      "Reciprocal condition number/precision: 7.328796748008544e-17 / 1.1102230246251565e-16\n",
      "  RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 37.91001057624817 seconds ---\n",
      "[[4.00000013e+08 1.40000000e+01 4.00000000e+00 ... 2.00000000e+01\n",
      "  2.00000000e+01 2.00000000e+01]\n",
      " [1.40000000e+01 8.00000020e+08 4.00000000e+00 ... 2.00000000e+01\n",
      "  2.00000000e+01 2.00000000e+01]\n",
      " [4.00000000e+00 4.00000000e+00 4.00000013e+08 ... 4.00000029e+08\n",
      "  4.00000029e+08 4.00000029e+08]\n",
      " ...\n",
      " [2.00000000e+01 2.00000000e+01 4.00000029e+08 ... 7.20000015e+17\n",
      "  7.20000015e+17 1.48000004e+10]\n",
      " [2.00000000e+01 2.00000000e+01 4.00000029e+08 ... 7.20000015e+17\n",
      "  7.20000015e+17 1.48000004e+10]\n",
      " [2.00000000e+01 2.00000000e+01 4.00000029e+08 ... 1.48000004e+10\n",
      "  1.48000004e+10 2.02000004e+10]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  99%|█████████▉| 992/1000 [00:03<00:00, 267.91it/s]\n",
      " Mean performance on train set: 66.147687\n",
      "With standard deviation: 11.979989\n",
      "\n",
      " Mean performance on test set: 63.221208\n",
      "With standard deviation: 13.381090\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:03<00:00, 252.05it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 1000000000.0 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 38.43676400184631 seconds ---\n",
      "[[4.00000001e+09 1.40000000e+01 4.00000000e+00 ... 2.00000000e+01\n",
      "  2.00000000e+01 2.00000000e+01]\n",
      " [1.40000000e+01 8.00000002e+09 4.00000000e+00 ... 2.00000000e+01\n",
      "  2.00000000e+01 2.00000000e+01]\n",
      " [4.00000000e+00 4.00000000e+00 4.00000001e+09 ... 4.00000003e+09\n",
      "  4.00000003e+09 4.00000003e+09]\n",
      " ...\n",
      " [2.00000000e+01 2.00000000e+01 4.00000003e+09 ... 7.20000001e+19\n",
      "  7.20000001e+19 1.48000000e+11]\n",
      " [2.00000000e+01 2.00000000e+01 4.00000003e+09 ... 7.20000001e+19\n",
      "  7.20000001e+19 1.48000000e+11]\n",
      " [2.00000000e+01 2.00000000e+01 4.00000003e+09 ... 1.48000000e+11\n",
      "  1.48000000e+11 2.02000000e+11]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance:  98%|█████████▊| 976/1000 [00:04<00:00, 268.70it/s]\n",
      " Mean performance on train set: 96.664827\n",
      "With standard deviation: 1.871320\n",
      "\n",
      " Mean performance on test set: 100.134704\n",
      "With standard deviation: 13.845906\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:04<00:00, 236.90it/s]\n",
      "\n",
      "\n",
      " #--- calculating kernel matrix when lmda = 10000000000.0 ---#\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 185 built in 37.32151246070862 seconds ---\n",
      "[[4.00e+10 1.40e+01 4.00e+00 ... 2.00e+01 2.00e+01 2.00e+01]\n",
      " [1.40e+01 8.00e+10 4.00e+00 ... 2.00e+01 2.00e+01 2.00e+01]\n",
      " [4.00e+00 4.00e+00 4.00e+10 ... 4.00e+10 4.00e+10 4.00e+10]\n",
      " ...\n",
      " [2.00e+01 2.00e+01 4.00e+10 ... 7.20e+21 7.20e+21 1.48e+12]\n",
      " [2.00e+01 2.00e+01 4.00e+10 ... 7.20e+21 7.20e+21 1.48e+12]\n",
      " [2.00e+01 2.00e+01 4.00e+10 ... 1.48e+12 1.48e+12 2.02e+12]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance: 100%|█████████▉| 997/1000 [00:03<00:00, 268.42it/s]\n",
      " Mean performance on train set: 98.175092\n",
      "With standard deviation: 4.720613\n",
      "\n",
      " Mean performance on test set: 100.144883\n",
      "With standard deviation: 13.958659\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:03<00:00, 260.49it/s]\n",
      "\n",
      "\n",
      "       lmda    rmse_test    std_test    rmse_train    std_train    k_time\n",
      "-----------  -----------  ----------  ------------  -----------  --------\n",
      "     1e-10       7.4649      1.71858       5.99354     0.356922   37.3696\n",
      "     1e-09       7.37511     1.76925       5.96304     0.374107   37.4747\n",
      "     1e-08       7.35069     1.78056       5.96511     0.378249   37.3297\n",
      "     1e-07       7.351       1.78147       5.96698     0.382093   37.7415\n",
      "     1e-06       7.35123     1.78052       5.96976     0.386318   37.5213\n",
      "     1e-05       7.34813     1.78029       5.97056     0.390719   37.671\n",
      "     0.0001      7.34984     1.7811        5.9425      0.331983   36.8013\n",
      "     0.001       7.35775     1.78098       5.9334      0.324965   35.8681\n",
      "     0.01        7.37427     1.79115       5.94069     0.347431   36.0018\n",
      "     0.1         7.3293      1.91363       6.42711     1.29367    36.3715\n",
      "     1           7.27134     2.20774       6.62425     1.2242     37.8397\n",
      "    10           7.49787     2.36815       6.81697     1.50182    37.5082\n",
      "   100           7.42887     2.64789       6.68766     1.34809    37.0946\n",
      "  1000           7.24914     2.65554       6.81906     1.41008    37.5171\n",
      " 10000           7.08183     2.6248        6.93431     1.38441    37.0127\n",
      "100000           8.23763     3.6653        9.395       1.04707    35.8353\n",
      "     1e+06       8.52983     3.58059       9.63525     0.68756    37.7773\n",
      "     1e+07      10.9642      3.35873      11.0591      1.32363    37.461\n",
      "     1e+08      63.2212     13.3811       66.1477     11.98       37.91\n",
      "     1e+09     100.135      13.8459       96.6648      1.87132    38.4368\n",
      "     1e+10     100.145      13.9587       98.1751      4.72061    37.3215\n"
     ]
    }
   ],
   "source": [
    "# tree pattern kernel, dataset acyclic.\n",
    "%load_ext line_profiler\n",
    "\n",
    "import sys\n",
    "sys.path.insert(0, \"../\")\n",
    "from pygraph.utils.utils import kernel_train_test\n",
    "from pygraph.kernels.treePatternKernel import treepatternkernel\n",
    "\n",
    "import numpy as np\n",
    "\n",
    "datafile = '../../../../datasets/acyclic/Acyclic/dataset_bps.ds'\n",
    "kernel_file_path = 'kernelmatrices_path_acyclic/'\n",
    "\n",
    "\n",
    "kernel_para = dict(node_label = 'atom', edge_label = 'bond_type', labeled = True, \\\n",
    "    kernel_type = 'untiln', h = 2)\n",
    "\n",
    "kernel_train_test(datafile, kernel_file_path, treepatternkernel, kernel_para, \\\n",
    "    hyper_name = 'lmda', hyper_range = np.logspace(-10, 10, num = 21, base = 10), \\\n",
    "    normalize = False, model_type = 'regression')\n",
    "\n",
    "# kernel_para['depth'] = 10\n",
    "# %lprun -f untildpathkernel \\\n",
    "#     kernel_train_test(datafile, kernel_file_path, untildpathkernel, kernel_para, normalize = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# results\n",
    "\n",
    "# untiln kernel when h = 2\n",
    "       lmda    rmse_test    std_test    rmse_train    std_train    k_time\n",
    "-----------  -----------  ----------  ------------  -----------  --------\n",
    "     1e-10       7.46524     1.71862       5.99486     0.356634   38.1447\n",
    "     1e-09       7.37326     1.77195       5.96155     0.374395   37.4921\n",
    "     1e-08       7.35105     1.78349       5.96481     0.378047   37.9971\n",
    "     1e-07       7.35213     1.77903       5.96728     0.382251   38.3182\n",
    "     1e-06       7.3524      1.77992       5.9696      0.3863     39.6428\n",
    "     1e-05       7.34958     1.78141       5.97114     0.39017    37.3711\n",
    "     0.0001      7.3513      1.78136       5.94251     0.331843   37.3967\n",
    "     0.001       7.35822     1.78119       5.9326      0.32534    36.7357\n",
    "     0.01        7.37552     1.79037       5.94089     0.34763    36.8864\n",
    "     0.1         7.32951     1.91346       6.42634     1.29405    36.8382\n",
    "     1           7.27134     2.20774       6.62425     1.2242     37.2425\n",
    "    10           7.49787     2.36815       6.81697     1.50182    37.8286\n",
    "   100           7.42887     2.64789       6.68766     1.34809    36.3701\n",
    "  1000           7.24914     2.65554       6.81906     1.41008    36.1695\n",
    " 10000           7.08183     2.6248        6.93431     1.38441    37.5723\n",
    "100000           8.021       3.43694       8.69813     0.909839   37.8158\n",
    "     1e+06       8.49625     3.6332        9.59333     0.96626    38.4688\n",
    "     1e+07      10.9067      3.17593      11.5642      2.07792    36.9926\n",
    "     1e+08      61.1524     10.4355       65.3527     13.9538     37.1321\n",
    "     1e+09      99.943      13.6994       98.8848      5.27014    36.7443\n",
    "     1e+10     100.083      13.8503       97.9168      3.22768    37.096\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The line_profiler extension is already loaded. To reload it, use:\n",
      "  %reload_ext line_profiler\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f6be3d4a8d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[(0, {'atom': 'C', 'label': 'C'}), (1, {'atom': 'C', 'label': 'C'}), (2, {'atom': 'C', 'label': 'C'}), (3, {'atom': 'C', 'label': 'C'}), (4, {'atom': 'S', 'label': 'S'})]\n",
      "\n",
      " --- kernel matrix of tree pattern kernel of size 39 built in 3.5270774364471436 seconds ---\n",
      "(array([[1.99007809e+036, 4.00000000e+000, 4.00000000e+000, ...,\n",
      "        1.00000000e+001, 1.00000000e+001, 1.00000000e+001],\n",
      "       [4.00000000e+000, 6.37886713e+019, 4.34000000e+002, ...,\n",
      "        6.37886713e+019, 6.37886713e+019, 6.37886713e+019],\n",
      "       [4.00000000e+000, 4.34000000e+002, 1.99007809e+036, ...,\n",
      "        4.40000000e+002, 4.40000000e+002, 4.40000000e+002],\n",
      "       ...,\n",
      "       [1.00000000e+001, 6.37886713e+019, 4.40000000e+002, ...,\n",
      "        2.94561201e+119, 1.16903692e+080, 4.42354433e+082],\n",
      "       [1.00000000e+001, 6.37886713e+019, 4.40000000e+002, ...,\n",
      "        1.16903692e+080, 4.21212139e+264, 1.66634383e+080],\n",
      "       [1.00000000e+001, 6.37886713e+019, 4.40000000e+002, ...,\n",
      "        4.42354433e+082, 1.66634383e+080, 5.17763068e+117]]), 3.5270774364471436)\n"
     ]
    }
   ],
   "source": [
    "%load_ext line_profiler\n",
    "\n",
    "import networkx as nx\n",
    "import matplotlib.pyplot as plt\n",
    "import sys\n",
    "sys.path.insert(0, \"../\")\n",
    "from pygraph.utils.graphfiles import loadDataset\n",
    "from pygraph.utils.utils import kernel_train_test\n",
    "\n",
    "from pygraph.kernels.treePatternKernel import treepatternkernel, _treepatternkernel_do\n",
    "\n",
    "import numpy as np\n",
    "\n",
    "datafile = '../../../../datasets/acyclic/Acyclic/dataset_bps.ds'\n",
    "\n",
    "dataset, y = loadDataset(datafile)\n",
    "G1 = dataset[100]\n",
    "G2 = dataset[20]\n",
    "data = [G1, G2]\n",
    "# nx.draw_networkx(G1)\n",
    "# plt.show()\n",
    "# print(G1.nodes(data=True)20\n",
    "nx.draw_networkx(G2)\n",
    "plt.show()\n",
    "print(G2.nodes(data=True))\n",
    "\n",
    "\n",
    "%lprun -f _treepatternkernel_do \\\n",
    "    kernel = treepatternkernel(dataset[1:40], node_label = 'atom', edge_label = 'bond_type', labeled = True, \\\n",
    "    kernel_type = 'untiln', lmda = 1, h = 10)\n",
    "\n",
    "print(kernel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " --- This is a classification problem ---\n",
      "\n",
      "\n",
      " Loading dataset from file...\n",
      "\n",
      " Calculating kernel matrix, this could take a while...\n",
      "retrieve patterns: 100%|██████████| 185/185 [00:00<00:00, 2064.69it/s]\n",
      "calculate kernels: 100%|██████████| 185/185 [00:00<00:00, 11170.00it/s]\n",
      "\n",
      " --- kernel matrix of cyclic pattern kernel of size 185 built in 0.10836505889892578 seconds ---\n",
      "[[0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " ...\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]\n",
      " [0. 0. 0. ... 0. 0. 0.]]\n",
      "\n",
      " Starting calculate accuracy/rmse...\n",
      "calculate performance: 100%|██████████| 1000/1000 [00:24<00:00, 36.41it/s]\n",
      " Mean performance on train set: 0.018072\n",
      "With standard deviation: 0.000000\n",
      "\n",
      " Mean performance on test set: 0.000000\n",
      "With standard deviation: 0.000000\n",
      "\n",
      "\n",
      "  accur_test    std_test    accur_train    std_train    k_time\n",
      "------------  ----------  -------------  -----------  --------\n",
      "           0           0      0.0180723            0  0.108365\n"
     ]
    }
   ],
   "source": [
    "%load_ext line_profiler\n",
    "\n",
    "import sys\n",
    "sys.path.insert(0, \"../\")\n",
    "from pygraph.utils.utils import kernel_train_test\n",
    "from pygraph.kernels.cyclicPatternKernel import cyclicpatternkernel\n",
    "\n",
    "import numpy as np\n",
    "\n",
    "datafile = '../../../../datasets/acyclic/Acyclic/dataset_bps.ds'\n",
    "kernel_file_path = 'kernelmatrices_path_acyclic/'\n",
    "\n",
    "kernel_para = dict(node_label = 'atom', edge_label = 'bond_type', labeled = True, cycle_bound = 200)\n",
    "\n",
    "# kernel_train_test(datafile, kernel_file_path, treeletkernel, kernel_para, normalize = False)\n",
    "\n",
    "kernel_train_test(datafile, kernel_file_path, cyclicpatternkernel, kernel_para, \\\n",
    "    normalize = False , model_type = 'classification')\n",
    "\n",
    "# kernel_para['k_func'] = 'minmax'\n",
    "# kernel_train_test(datafile, kernel_file_path, untildpathkernel, kernel_para, \\\n",
    "#     hyper_name = 'depth', hyper_range = np.linspace(0, 10, 11), normalize = True)\n",
    "# kernel_train_test(datafile, kernel_file_path, untildpathkernel, kernel_para, \\\n",
    "#     hyper_name = 'depth', hyper_range = np.linspace(0, 10, 11), normalize = False)\n",
    "# # kernel_train_test(datafile, kernel_file_path, untildpathkernel, kernel_para, normalize = False)\n",
    "\n",
    "# kernel_para['depth'] = 10\n",
    "# %lprun -f untildpathkernel \\\n",
    "#     kernel_train_test(datafile, kernel_file_path, untildpathkernel, kernel_para, normalize = False)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}