From f3b1e6d72a82f2d3ffe1722cf99d316fd28aa766 Mon Sep 17 00:00:00 2001
From: choocewhatulike <1901722105@qq.com>
Date: Mon, 12 Mar 2018 22:41:38 +0800
Subject: [PATCH] add gpu

---
 .../code/__pycache__/model.cpython-36.pyc          | Bin 2342 -> 0 bytes
 model_inplement/code/train.py                      |  91 +++++++++++----------
 2 files changed, 50 insertions(+), 41 deletions(-)
 delete mode 100644 model_inplement/code/__pycache__/model.cpython-36.pyc

diff --git a/model_inplement/code/__pycache__/model.cpython-36.pyc b/model_inplement/code/__pycache__/model.cpython-36.pyc
deleted file mode 100644
index 167f6ff9d4401d8077c23d66909e6c96b6049a11..0000000000000000000000000000000000000000
GIT binary patch
literal 0
KcmV+b0RR6000031

literal 2342
zcmV+>3EB2D4Gjt_J*HYv3jhG)0000000000000001ONa4KmY&$b6NlZ0Av7U0c-$T
z0Av7U0c-(U0b~GV0c-+l16l)Q0AvDe1Z@Rc1pxqL0AvAd25JLp2U-Y60AvGX1cU%&
z1Z4wi34;S#3P%8B1!M+<0AvPb18NC_16m7Y0aE}e2k8I+002%Y0on*wVRC6<Vr*q&
z0000000000000000{{R300000b0h!&0A&DL0b~GL0*3%*0b~JW0)z-!17rha1cU%u
z1cv|w09pl804W9914uzmV+jBN0000000{s900jU500RI30CQFV0049VbOC$-g92&-
zg8%^lbOU?=d;@$0d<BC9d;nhrbOU4od;=5!d<J|6d<cUCd;nhsbOve%WC45z6aahz
zg93a2UkG#tY6)ZlWCM)>d;nhxWB^kDDFjaG0ssI2=>Y%$04V|517vA!DGS;Kb9HcK
za&iLz007zuUtei%X>?y-+6+N-bY*UIX>V>$WpvsKcW-iJUu<D{WpdgIb7gLHUu<D{
zWpdgAZf@EJOlfXqVRG6GZ*_EVb#z~BVR>b8+6qi>XH#!xbZudH+6QxQW^`>~cqs|m
z1aoC<X4(pAZg6#UUvp`CW!ejGb#!obbYF96dS%)WcW-iJUubD$WMyt&b7^{I+7EYc
za%5j_b!}g4VR>b8bJ`JiZ*pW`V{dMBWq5R7b7^{I+7NSPZggL0X=G$&ZeMe0dS%)V
zb7gLHUv716Uu<D{WpZ=c5p!j3bYEj{ZggdMbYF96dSxjA+6iA@V{Bn_b6;Pn0Qxy)
zI$T3<Y;0v`Wn5-;WMOVjOi)~AVRLj&Oi)~HZ)9a`UukY|Y-Md_ZggB@Z)9a$ZEs{{
zY%XwlatHtb00{s90CN@q000034gv%L0s#U62muHI0s#U62mulSdJITGPA*?xX>MtB
zUtePa000000000A000050000(0001UcmMzZX8?Tycu;%+L;zh#eFJ9zeFS(Md;>%P
zT@-x<d<1F$d;n?zd<BC6g8=~maTr$sbOLGvd<0|xWCD!>eFl60Y6N@+g8_X9d;)3!
zd<TO80RV9dR{(SZY6E-%WC3IXjRJiLd;n?%d;n?&d<cU9g8_XBd<jzkDFaS&0RR91
zDFJc@0000f2igW<aByXAWO51s007zrbZ>HFXxal~VRUi}0001T5C8xGat;6h04WM`
z5dZ)HS_xudbYo~=WN%|y3}kO(Uv_0<Uu<b}blL-CZ(~{pb6;#}b9CASb7gLHS_N}o
zc4cGQ19oL&S_forV_$Y<W7-C9b#!obbaER20043u0001T8~^|S+6QKDa(7{JWElVe
z0CN@q000031OW;G1OW;G6afwa3IPrQ4gnAWdJ9NFPA+C|a(7{JWGM#P2wz`rVQpn!
zU)l;^Uu|z>b!=r{U)l^`UvYI|Y;Iv~WnW)%2mk;8awGr%0NM>-Ut?@xb8};5Y;0d&
zavJ~u0CF1u001cgavA^t0CF4v0043W000080001U1ONa42muTbasvPW0Am0E00000
z0000000093000000046&0001G09pZL09pcv0A>MX0b~M%2wDSV17rk*09pix00aP9
z1ycYi1#$@h003hJ000000000500007000030001UjsO4vbO3Y#d;o(2Y661*0RVgf
zd;nhqd;)v`Uj%#ud;nhsd<1*|Uj}psY6yG*d;)v}WC3IXWCM)_d;nhwbO&k*WCVNy
z6aah#g93a2Ukh{xY7Bz_d;nh!bO&k<WCdgfjRAZBUk`K#Y7lf0Y7%?|WCw!+g8_U1
zUlV))Y7=S{Y87M&WC()-0RUtGQvfLnPDWHI1#%Jq007zxXlZ0*Wo}<{X?kVa3T}06
zUu<D{WpZ=c3u0k(V`yJyX>xOP+6`i9WNC6`V{~b6ZeeV41ONa4asdDU04V`-1^@s6
zas>bY0B4$+nVFfHxj$!`nVFfHnYq6y7IFsw0043c0001T2mk;8auNUl0NM{{a&=#5
zX=G$&ZeMe0dS%)UXL5C4Zgp*6Y+-q2a&y`cV{dMBWq5R7c4cE<b7^{Iatr_f0NMjb
zQdQamXL5CN4FCWD+5%={+5}W#ZfM#BbYX62+6Pl_W^`>~cybT`007zvP+@XmZDn+2
za&jI3007zsRAp{+Z*tlTV{dMBWq5R7c4cGQ1Y}`!VcH0FZfRz3a&2EJ1#%Gp0043l
z0001TEdT%jaxMS>0CFz?001cgavA^t0CF1u0043v0001T2mk;8DF6Tfa~=Qy0003F
z0tNvF0R{mE0|o&C0RjO60SW>V0SW;Q0u%vy6+v`#Wo~q7Z*ERybS__CX>MtBUtePa
z0000000007000040000(0001UVgLXDbO34rd;w$ug93d4d;n?!d;x<2TmpRpeFJm=
zY6E-%WC4Q$eFA&{Y6N@$Y6W}(g8_p9eFS^}Y6f%wY6pA-d;n?)g93vAeFby?Y6pA*
zY6*h?d<BC7Y6*h?eFl66QvfLfPI3hR001cpavlHx0NM$4ZgX*UWo3G0axnk^0NMw0
zadl;7dS!An0001TG5`Poau5Ij0NMg=ZE`pO007znbSVdN5dZ)H+6HNEaCLNZS_5cb
zblL`JX=G$&ZrTBL+67^3aA;w2BLDyZavJ~u0CF1u0043v0001TBme*aKL7v#a}EFi
z0006E0S*BS0T2Nc0Tcpy6hU-!Wo~q7Z*ERybS`FZa(7{JWGM!6B>(^bawY%(0CFb)
z0043b0001TBme*aawq@*0CF1u0043u0000f0dg7u0043v0001T2><{9C;$Kea|8eY
z00;pL7IFyy001cr+68WPZE$&V9smFU+6Z)Sa${&NZf<f60002m4s>sFV`wg6b#!lM
za$#h00ssI2S`&0{a${&NZf-7Sb#7yHX>V>}Y}yKDb#7yHX>V>}Y}x@v+6GN;WOZz1
zasvPW0CEWc0043u0001T8vp<RavJ~u0CF4v007zuJZ*1eb!=rm0RR91a|{3g00;sI
M0SW;O0Sp2XAV6b4y#N3J

diff --git a/model_inplement/code/train.py b/model_inplement/code/train.py
index ae7ee925..4b22f69a 100644
--- a/model_inplement/code/train.py
+++ b/model_inplement/code/train.py
@@ -1,8 +1,14 @@
-import gensim
-from gensim import models
-
 import os
 import pickle
+
+import matplotlib.pyplot as plt
+import matplotlib.ticker as ticker
+
+import nltk
+import numpy as np
+import torch
+
+from model import *
             
 class SampleIter:
     def __init__(self, dirname):
@@ -32,35 +38,6 @@ def train_word_vec():
     model = models.Word2Vec(sentences=sents, size=200, sg=0, workers=4, min_count=5)
     model.save('yelp.word2vec')
 
-
-'''
-Train process
-'''
-import math
-import os
-import copy
-import pickle
-
-import matplotlib.pyplot as plt
-import matplotlib.ticker as ticker
-import numpy as np
-import json
-import nltk
-from gensim.models import Word2Vec
-import torch
-from torch.utils.data import DataLoader, Dataset
-
-from model import *
-
-net = HAN(input_size=200, output_size=5, 
-        word_hidden_size=50, word_num_layers=1, word_context_size=100,
-        sent_hidden_size=50, sent_num_layers=1, sent_context_size=100)
-
-optimizer = torch.optim.SGD(net.parameters(), lr=0.01)
-criterion = nn.NLLLoss()
-num_epoch = 1
-batch_size = 64
-
 class Embedding_layer:
     def __init__(self, wv, vector_size):
         self.wv = wv
@@ -73,10 +50,8 @@ class Embedding_layer:
             v = np.zeros(self.vector_size)
         return v
 
-embed_model = Word2Vec.load('yelp.word2vec')
-embedding = Embedding_layer(embed_model.wv, embed_model.wv.vector_size)
-del embed_model
 
+from torch.utils.data import DataLoader, Dataset
 class YelpDocSet(Dataset):
     def __init__(self, dirname, num_files, embedding):
         self.dirname = dirname
@@ -103,12 +78,28 @@ def collate(iterable):
         x_list.append(x)
     return x_list, torch.LongTensor(y_list)
 
-if __name__ == '__main__':
+def train(net, num_epoch, batch_size, print_size=10, use_cuda=False):
+    from gensim.models import Word2Vec
+    import torch
+    import gensim
+    from gensim import models
+
+    embed_model = Word2Vec.load('yelp.word2vec')
+    embedding = Embedding_layer(embed_model.wv, embed_model.wv.vector_size)
+    del embed_model
+
+    optimizer = torch.optim.SGD(net.parameters(), lr=0.01)
+    criterion = nn.NLLLoss()
+
     dirname = 'reviews'
-    dataloader = DataLoader(YelpDocSet(dirname, 238, embedding), batch_size=batch_size, collate_fn=collate)
+    dataloader = DataLoader(YelpDocSet(dirname, 238, embedding), 
+                            batch_size=batch_size, 
+                            collate_fn=collate,
+                            num_workers=4)
     running_loss = 0.0
-    print_size = 10
 
+    if use_cuda:
+        net.cuda()
     for epoch in range(num_epoch):
         for i, batch_samples in enumerate(dataloader):
             x, y = batch_samples
@@ -119,11 +110,16 @@ if __name__ == '__main__':
                     sent_vec = []
                     for word in sent:
                         vec = embedding.get_vec(word)
-                        sent_vec.append(torch.Tensor(vec.reshape((1, -1))))
+                        vec = torch.Tensor(vec.reshape((1, -1)))
+                        if use_cuda:
+                            vec = vec.cuda()
+                        sent_vec.append(vec)
                     sent_vec = torch.cat(sent_vec, dim=0)
                     # print(sent_vec.size())
                     doc.append(Variable(sent_vec))
                 doc_list.append(doc)
+            if use_cuda:
+                y = y.cuda()
             y = Variable(y)
             predict = net(doc_list)
             loss = criterion(predict, y)
@@ -131,8 +127,21 @@ if __name__ == '__main__':
             loss.backward()
             optimizer.step()
             running_loss += loss.data[0]
-            print(loss.data[0])
             if i % print_size == print_size-1:
                 print(running_loss/print_size)
                 running_loss = 0.0
-        
+                torch.save(net.state_dict(), 'model.dict')
+    torch.save(net.state_dict(), 'model.dict')
+    
+
+if __name__ == '__main__':
+    '''
+    Train process
+    '''
+    
+
+    net = HAN(input_size=200, output_size=5, 
+            word_hidden_size=50, word_num_layers=1, word_context_size=100,
+            sent_hidden_size=50, sent_num_layers=1, sent_context_size=100)
+    
+    train(net, num_epoch=1, batch_size=64, use_cuda=True)