OpenI
/
graphkit-learn
Not watched
1
0
Fork 0
Code
Releases 2 Wiki Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain
You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
69 Commits
4 Branches
345 MB
58 Download
Tree: 22a1f1e8d8
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from '22a1f1e8d8'
${ noResults }
graphkit-learn/notebooks/run_randomwalkkernel.ipynb

run_randomwalkkernel.ipynb 545 kB
Raw Normal View History

Plus de choses sont changées...
7 years ago
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
Plus de choses sont changées...
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
Plus de choses sont changées...
7 years ago
​
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
Plus de choses sont changées...
7 years ago
​
​
​
​
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
Plus de choses sont changées...
7 years ago
​
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
​
​
​
​
​
​
​
​
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
1. add parallel computing scheme to spkernel and model_selection_precomputed. 2. modify model_selection_precomputed so that all results are written into memory and then to a file at last section of code, in case that on cpu/disk seperated systems the IO takes too much time. 3. correct utils.floyd_warshall_numpy function. DONNOT use the last version.
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
Plus de choses sont changées...
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
Plus de choses sont changées...
7 years ago
​
​
​
​
​
​
​
​
​
​
1. add Sylvester Equation Methods for the generalized random walk kernel. 2. correct an error in the common walk kernel. DON NOT use the old one. 3. improve the method to construct fully-labeled direct product graphs, much faster for sparse graphs.
7 years ago
​
Plus de choses sont changées...
7 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
 12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364656667686970717273747576777879808182838485868788899091929394959697989910010110210310410510610710810911011111211311411511611711811912012112212312412512612712812913013113213313413513613713813914014114214314414514614714814915015115215315415515615715815916016116216316416516616716816917017117217317417517617717817918018118218318418518618718818919019119219319419519619719819920020120220320420520620720820921021121221321421521621721821922022122222322422522622722822923023123223323423523623723823924024124224324424524624724824925025125225325425525625725825926026126226326426526626726826927027127227327427527627727827928028128228328428528628728828929029129229329429529629729829930030130230330430530630730830931031131231331431531631731831932032132232332432532632732832933033133233333433533633733833934034134234334434534634734834935035135235335435535635735835936036136236336436536636736836937037137237337437537637737837938038138238338438538638738838939039139239339439539639739839940040140240340440540640740840941041141241341441541641741841942042142242342442542642742842943043143243343443543643743843944044144244344444544644744844945045145245345445545645745845946046146246346446546646746846947047147247347447547647747847948048148248348448548648748848949049149249349449549649749849950050150250350450550650750850951051151251351451551651751851952052152252352452552652752852953053153253353453553653753853954054154254354454554654754854955055155255355455555655755855956056156256356456556656756856957057157257357457557657757857958058158258358458558658758858959059159259359459559659759859960060160260360460560660760860961061161261361461561661761861962062162262362462562662762862963063163263363463563663763863964064164264364464564664764864965065165265365465565665765865966066166266366466566666766866967067167267367467567667767867968068168268368468568668768868969069169269369469569669769869970070170270370470570670770870971071171271371471571671771871972072172272372472572672772872973073173273373473573673773873974074174274374474574674774874975075175275375475575675775875976076176276376476576676776876977077177277377477577677777877978078178278378478578678778878979079179279379479579679779879980080180280380480580680780880981081181281381481581681781881982082182282382482582682782882983083183283383483583683783883984084184284384484584684784884985085185285385485585685785885986086186286386486586686786886987087187287387487587687787887988088188288388488588688788888989089189289389489589689789889990090190290390490590690790890991091191291391491591691791891992092192292392492592692792892993093193293393493593693793893994094194294394494594694794894995095195295395495595695795895996096196296396496596696796896997097197297397497597697797897998098198298398498598698798898999099199299399499599699799899910001001100210031004100510061007100810091010101110121013101410151016101710181019102010211022102310241025102610271028102910301031103210331034103510361037103810391040104110421043104410451046104710481049105010511052105310541055105610571058105910601061106210631064106510661067106810691070107110721073107410751076107710781079108010811082108310841085108610871088108910901091109210931094109510961097109810991100110111021103110411051106110711081109111011111112111311141115111611171118111911201121112211231124112511261127112811291130113111321133113411351136113711381139114011411142114311441145114611471148114911501151115211531154115511561157115811591160116111621163116411651166116711681169117011711172117311741175117611771178117911801181118211831184118511861187118811891190119111921193119411951196119711981199120012011202120312041205120612071208120912101211121212131214121512161217121812191220122112221223122412251226122712281229123012311232123312341235123612371238123912401241124212431244124512461247124812491250125112521253125412551256125712581259126012611262126312641265126612671268126912701271127212731274127512761277127812791280128112821283128412851286128712881289129012911292129312941295129612971298129913001301130213031304130513061307130813091310131113121313131413151316131713181319132013211322132313241325132613271328132913301331133213331334133513361337133813391340134113421343134413451346134713481349135013511352135313541355135613571358135913601361136213631364136513661367136813691370137113721373137413751376137713781379138013811382138313841385138613871388138913901391139213931394139513961397139813991400140114021403140414051406140714081409141014111412141314141415141614171418141914201421142214231424142514261427142814291430143114321433143414351436143714381439144014411442144314441445144614471448144914501451145214531454145514561457145814591460146114621463146414651466146714681469147014711472147314741475147614771478147914801481148214831484148514861487148814891490149114921493149414951496149714981499150015011502150315041505150615071508150915101511151215131514151515161517151815191520152115221523152415251526152715281529153015311532153315341535153615371538153915401541154215431544154515461547154815491550155115521553155415551556155715581559156015611562156315641565156615671568156915701571157215731574157515761577157815791580158115821583158415851586158715881589159015911592159315941595159615971598159916001601160216031604160516061607160816091610161116121613161416151616161716181619162016211622162316241625162616271628162916301631163216331634163516361637163816391640164116421643164416451646164716481649165016511652165316541655165616571658165916601661166216631664166516661667166816691670167116721673167416751676167716781679168016811682168316841685168616871688168916901691169216931694169516961697169816991700170117021703170417051706170717081709171017111712171317141715171617171718171917201721172217231724172517261727172817291730173117321733173417351736173717381739174017411742 
  1. {

  2.  "cells": [

  3.   {

  4.    "cell_type": "code",

  5.    "execution_count": 1,

  6.    "metadata": {

  7.     "scrolled": true

  8.    },

  9.    "outputs": [

  10.     {

  11.      "name": "stdout",

  12.      "output_type": "stream",

  13.      "text": [

  14.       "\n",

  15.       "Acyclic\n",

  16.       "\n",

  17.       "--- This is a regression problem ---\n",

  18.       "\n",

  19.       "\n",

  20.       "I. Loading dataset from file...\n",

  21.       "\n",

  22.       "2. Calculating gram matrices. This could take a while...\n",

  23.       "\n",

  24.       " None edge weight specified. Set all weight to 1.\n",

  25.       "\n",

  26.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 4963.99it/s]\n",

  27.       "calculating kernels:   6%|▌         | 1048/16836.0 [00:00<00:01, 10475.79it/s]"

  28.      ]

  29.     },

  30.     {

  31.      "name": "stderr",

  32.      "output_type": "stream",

  33.      "text": [

  34.       "../pygraph/kernels/randomWalkKernel.py:99: UserWarning: The Sylvester equation (rather than generalized Sylvester equation) is used; edge label number has to smaller than 3.\n",

  35.       "  'The Sylvester equation (rather than generalized Sylvester equation) is used; edge label number has to smaller than 3.'\n"

  36.      ]

  37.     },

  38.     {

  39.      "name": "stdout",

  40.      "output_type": "stream",

  41.      "text": [

  42.       "calculating kernels:  95%|█████████▍| 15982/16836.0 [00:00<00:00, 17223.87it/s]\n",

  43.       " --- kernel matrix of random walk kernel of size 183 built in 1.029677391052246 seconds ---\n",

  44.       "\n",

  45.       "the gram matrix with parameters {'weight': 1.0, 'compute_method': 'sylvester'} is: \n",

  46.       "ignored, as it contains elements that are not numbers.\n",

  47.       "\n",

  48.       " None edge weight specified. Set all weight to 1.\n",

  49.       "\n",

  50.       "\n",

  51.       "compute adjacency matrices:   0%|          | 0/183 [00:00<?, ?it/s]\u001b[A\n",

  52.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 5079.19it/s]\u001b[A\n",

  53.       "calculating kernels:   0%|          | 0/16836.0 [00:00<?, ?it/s]\u001b[A\n",

  54.       "calculating kernels:  12%|█▏        | 2099/16836.0 [00:00<00:00, 20980.72it/s]\u001b[A"

  55.      ]

  56.     },

  57.     {

  58.      "name": "stderr",

  59.      "output_type": "stream",

  60.      "text": [

  61.       "../pygraph/utils/model_selection_precomputed.py:126: RuntimeWarning: invalid value encountered in sqrt\n",

  62.       "  Kmatrix[i][j] /= np.sqrt(Kmatrix_diag[i] * Kmatrix_diag[j])\n"

  63.      ]

  64.     },

  65.     {

  66.      "name": "stdout",

  67.      "output_type": "stream",

  68.      "text": [

  69.       "\n",

  70.       "calculating kernels:  25%|██▍       | 4148/16836.0 [00:00<00:00, 20830.01it/s]\u001b[A\n",

  71.       "calculating kernels:  37%|███▋      | 6178/16836.0 [00:00<00:00, 20666.10it/s]\u001b[A\n",

  72.       "calculating kernels:  49%|████▊     | 8176/16836.0 [00:00<00:00, 20455.36it/s]\u001b[A\n",

  73.       "calculating kernels:  60%|█████▉    | 10061/16836.0 [00:00<00:00, 19945.49it/s]\u001b[A\n",

  74.       "calculating kernels:  71%|███████   | 11957/16836.0 [00:00<00:00, 19635.96it/s]\u001b[A\n",

  75.       "calculating kernels:  82%|████████▏ | 13777/16836.0 [00:00<00:00, 19179.89it/s]\u001b[A\n",

  76.       "calculating kernels:  92%|█████████▏| 15527/16836.0 [00:00<00:00, 18453.92it/s]\u001b[A\n",

  77.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 19014.40it/s]\u001b[A\n",

  78.       " --- kernel matrix of random walk kernel of size 183 built in 0.925408124923706 seconds ---\n",

  79.       "\n",

  80.       "the gram matrix with parameters {'weight': 0.31622776601683794, 'compute_method': 'sylvester'} is: \n",

  81.       "ignored, as it contains elements that are not numbers.\n",

  82.       "\n",

  83.       " None edge weight specified. Set all weight to 1.\n",

  84.       "\n",

  85.       "\n",

  86.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:02<00:00, 8414.86it/s] \n",

  87.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 4778.99it/s]\n",

  88.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 19706.88it/s]\n",

  89.       "\n",

  90.       " --- kernel matrix of random walk kernel of size 183 built in 0.8956427574157715 seconds ---\n",

  91.       "\n",

  92.       "the gram matrix with parameters {'weight': 0.1, 'compute_method': 'sylvester'} is: \n",

  93.       "[[1.         0.99734465 1.         ... 0.97126366 0.97126366 0.97892691]\n",

  94.       " [0.99734465 1.         0.99734465 ... 0.9857487  0.9857487  0.99113366]\n",

  95.       " [1.         0.99734465 1.         ... 0.97126366 0.97126366 0.97892691]\n",

  96.       " ...\n",

  97.       " [0.97126366 0.9857487  0.97126366 ... 1.         1.         0.99911683]\n",

  98.       " [0.97126366 0.9857487  0.97126366 ... 1.         1.         0.99911683]\n",

  99.       " [0.97892691 0.99113366 0.97892691 ... 0.99911683 0.99911683 1.        ]]\n"

  100.      ]

  101.     },

  102.     {

  103.      "data": {

  104.       "image/png": "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\n",

  105.       "text/plain": [

  106.        "<Figure size 288x288 with 2 Axes>"

  107.       ]

  108.      },

  109.      "metadata": {},

  110.      "output_type": "display_data"

  111.     },

  112.     {

  113.      "name": "stdout",

  114.      "output_type": "stream",

  115.      "text": [

  116.       "\n",

  117.       " None edge weight specified. Set all weight to 1.\n",

  118.       "\n",

  119.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 4703.72it/s]\n",

  120.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 19269.04it/s]\n",

  121.       "\n",

  122.       " --- kernel matrix of random walk kernel of size 183 built in 0.9144845008850098 seconds ---\n",

  123.       "\n",

  124.       "the gram matrix with parameters {'weight': 0.03162277660168379, 'compute_method': 'sylvester'} is: \n",

  125.       "[[1.         0.99947611 1.         ... 0.9950873  0.9950873  0.9957061 ]\n",

  126.       " [0.99947611 1.         0.99947611 ... 0.99775202 0.99775202 0.99818019]\n",

  127.       " [1.         0.99947611 1.         ... 0.9950873  0.9950873  0.9957061 ]\n",

  128.       " ...\n",

  129.       " [0.9950873  0.99775202 0.9950873  ... 1.         1.         0.99993104]\n",

  130.       " [0.9950873  0.99775202 0.9950873  ... 1.         1.         0.99993104]\n",

  131.       " [0.9957061  0.99818019 0.9957061  ... 0.99993104 0.99993104 1.        ]]\n"

  132.      ]

  133.     },

  134.     {

  135.      "data": {

  136.       "image/png": "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\n",

  137.       "text/plain": [

  138.        "<Figure size 288x288 with 2 Axes>"

  139.       ]

  140.      },

  141.      "metadata": {},

  142.      "output_type": "display_data"

  143.     },

  144.     {

  145.      "name": "stdout",

  146.      "output_type": "stream",

  147.      "text": [

  148.       "\n",

  149.       " None edge weight specified. Set all weight to 1.\n",

  150.       "\n",

  151.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 5034.78it/s]\n",

  152.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 18679.61it/s]\n",

  153.       "\n",

  154.       " --- kernel matrix of random walk kernel of size 183 built in 0.9394266605377197 seconds ---\n",

  155.       "\n",

  156.       "the gram matrix with parameters {'weight': 0.01, 'compute_method': 'sylvester'} is: \n",

  157.       "[[1.         0.99985372 1.         ... 0.99871463 0.99871463 0.99877465]\n",

  158.       " [0.99985372 1.         0.99985372 ... 0.99943339 0.99943339 0.999475  ]\n",

  159.       " [1.         0.99985372 1.         ... 0.99871463 0.99871463 0.99877465]\n",

  160.       " ...\n",

  161.       " [0.99871463 0.99943339 0.99871463 ... 1.         1.         0.99999333]\n",

  162.       " [0.99871463 0.99943339 0.99871463 ... 1.         1.         0.99999333]\n",

  163.       " [0.99877465 0.999475   0.99877465 ... 0.99999333 0.99999333 1.        ]]\n"

  164.      ]

  165.     },

  166.     {

  167.      "data": {

  168.       "image/png": "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\n",

  169.       "text/plain": [

  170.        "<Figure size 288x288 with 2 Axes>"

  171.       ]

  172.      },

  173.      "metadata": {},

  174.      "output_type": "display_data"

  175.     },

  176.     {

  177.      "name": "stdout",

  178.      "output_type": "stream",

  179.      "text": [

  180.       "\n",

  181.       " None edge weight specified. Set all weight to 1.\n",

  182.       "\n",

  183.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 4192.15it/s]\n",

  184.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 19200.08it/s]\n",

  185.       "\n",

  186.       " --- kernel matrix of random walk kernel of size 183 built in 0.9225409030914307 seconds ---\n",

  187.       "\n",

  188.       "the gram matrix with parameters {'weight': 0.0031622776601683794, 'compute_method': 'sylvester'} is: \n",

  189.       "[[1.         0.99995537 1.         ... 0.99961757 0.99961757 0.99962354]\n",

  190.       " [0.99995537 1.         0.99995537 ... 0.99983399 0.99983399 0.99983813]\n",

  191.       " [1.         0.99995537 1.         ... 0.99961757 0.99961757 0.99962354]\n",

  192.       " ...\n",

  193.       " [0.99961757 0.99983399 0.99961757 ... 1.         1.         0.99999934]\n",

  194.       " [0.99961757 0.99983399 0.99961757 ... 1.         1.         0.99999934]\n",

  195.       " [0.99962354 0.99983813 0.99962354 ... 0.99999934 0.99999934 1.        ]]\n"

  196.      ]

  197.     },

  198.     {

  199.      "data": {

  200.       "image/png": "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\n",

  201.       "text/plain": [

  202.        "<Figure size 288x288 with 2 Axes>"

  203.       ]

  204.      },

  205.      "metadata": {},

  206.      "output_type": "display_data"

  207.     },

  208.     {

  209.      "name": "stdout",

  210.      "output_type": "stream",

  211.      "text": [

  212.       "\n",

  213.       " None edge weight specified. Set all weight to 1.\n",

  214.       "\n",

  215.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 5060.54it/s]\n",

  216.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 18723.67it/s]\n",

  217.       "\n",

  218.       " --- kernel matrix of random walk kernel of size 183 built in 0.9375 seconds ---\n",

  219.       "\n",

  220.       "the gram matrix with parameters {'weight': 0.001, 'compute_method': 'sylvester'} is: \n",

  221.       "[[1.         0.99998604 1.         ... 0.9998814  0.9998814  0.99988199]\n",

  222.       " [0.99998604 1.         0.99998604 ... 0.99994879 0.99994879 0.9999492 ]\n",

  223.       " [1.         0.99998604 1.         ... 0.9998814  0.9998814  0.99988199]\n",

  224.       " ...\n",

  225.       " [0.9998814  0.99994879 0.9998814  ... 1.         1.         0.99999993]\n",

  226.       " [0.9998814  0.99994879 0.9998814  ... 1.         1.         0.99999993]\n",

  227.       " [0.99988199 0.9999492  0.99988199 ... 0.99999993 0.99999993 1.        ]]\n"

  228.      ]

  229.     },

  230.     {

  231.      "data": {

  232.       "image/png": "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\n",

  233.       "text/plain": [

  234.        "<Figure size 288x288 with 2 Axes>"

  235.       ]

  236.      },

  237.      "metadata": {},

  238.      "output_type": "display_data"

  239.     },

  240.     {

  241.      "name": "stdout",

  242.      "output_type": "stream",

  243.      "text": [

  244.       "\n",

  245.       " None edge weight specified. Set all weight to 1.\n",

  246.       "\n",

  247.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 2867.18it/s]\n",

  248.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 19141.80it/s]\n",

  249.       "\n",

  250.       " --- kernel matrix of random walk kernel of size 183 built in 0.9451649188995361 seconds ---\n",

  251.       "\n",

  252.       "the gram matrix with parameters {'weight': 0.00031622776601683794, 'compute_method': 'sylvester'} is: \n",

  253.       "[[1.         0.9999956  1.         ... 0.99996272 0.99996272 0.99996278]\n",

  254.       " [0.9999956  1.         0.9999956  ... 0.99998393 0.99998393 0.99998398]\n",

  255.       " [1.         0.9999956  1.         ... 0.99996272 0.99996272 0.99996278]\n",

  256.       " ...\n",

  257.       " [0.99996272 0.99998393 0.99996272 ... 1.         1.         0.99999999]\n",

  258.       " [0.99996272 0.99998393 0.99996272 ... 1.         1.         0.99999999]\n",

  259.       " [0.99996278 0.99998398 0.99996278 ... 0.99999999 0.99999999 1.        ]]\n"

  260.      ]

  261.     },

  262.     {

  263.      "data": {

  264.       "image/png": "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\n",

  265.       "text/plain": [

  266.        "<Figure size 288x288 with 2 Axes>"

  267.       ]

  268.      },

  269.      "metadata": {},

  270.      "output_type": "display_data"

  271.     },

  272.     {

  273.      "name": "stdout",

  274.      "output_type": "stream",

  275.      "text": [

  276.       "\n",

  277.       " None edge weight specified. Set all weight to 1.\n",

  278.       "\n",

  279.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 4782.41it/s]\n",

  280.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 19126.90it/s]\n",

  281.       "\n",

  282.       " --- kernel matrix of random walk kernel of size 183 built in 0.9203457832336426 seconds ---\n",

  283.       "\n",

  284.       "the gram matrix with parameters {'weight': 0.0001, 'compute_method': 'sylvester'} is: \n",

  285.       "[[1.         0.99999861 1.         ... 0.99998824 0.99998824 0.99998824]\n",

  286.       " [0.99999861 1.         0.99999861 ... 0.99999493 0.99999493 0.99999494]\n",

  287.       " [1.         0.99999861 1.         ... 0.99998824 0.99998824 0.99998824]\n",

  288.       " ...\n",

  289.       " [0.99998824 0.99999493 0.99998824 ... 1.         1.         1.        ]\n",

  290.       " [0.99998824 0.99999493 0.99998824 ... 1.         1.         1.        ]\n",

  291.       " [0.99998824 0.99999494 0.99998824 ... 1.         1.         1.        ]]\n"

  292.      ]

  293.     },

  294.     {

  295.      "data": {

  296.       "image/png": "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\n",

  297.       "text/plain": [

  298.        "<Figure size 288x288 with 2 Axes>"

  299.       ]

  300.      },

  301.      "metadata": {},

  302.      "output_type": "display_data"

  303.     },

  304.     {

  305.      "name": "stdout",

  306.      "output_type": "stream",

  307.      "text": [

  308.       "\n",

  309.       " None edge weight specified. Set all weight to 1.\n",

  310.       "\n",

  311.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 4012.01it/s]\n",

  312.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 18759.77it/s]\n",

  313.       "\n",

  314.       " --- kernel matrix of random walk kernel of size 183 built in 0.945075273513794 seconds ---\n",

  315.       "\n",

  316.       "the gram matrix with parameters {'weight': 3.1622776601683795e-05, 'compute_method': 'sylvester'} is: \n",

  317.       "[[1.         0.99999956 1.         ... 0.99999628 0.99999628 0.99999628]\n",

  318.       " [0.99999956 1.         0.99999956 ... 0.9999984  0.9999984  0.9999984 ]\n",

  319.       " [1.         0.99999956 1.         ... 0.99999628 0.99999628 0.99999628]\n",

  320.       " ...\n",

  321.       " [0.99999628 0.9999984  0.99999628 ... 1.         1.         1.        ]\n",

  322.       " [0.99999628 0.9999984  0.99999628 ... 1.         1.         1.        ]\n",

  323.       " [0.99999628 0.9999984  0.99999628 ... 1.         1.         1.        ]]\n"

  324.      ]

  325.     },

  326.     {

  327.      "data": {

  328.       "image/png": "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\n",

  329.       "text/plain": [

  330.        "<Figure size 288x288 with 2 Axes>"

  331.       ]

  332.      },

  333.      "metadata": {},

  334.      "output_type": "display_data"

  335.     },

  336.     {

  337.      "name": "stdout",

  338.      "output_type": "stream",

  339.      "text": [

  340.       "\n",

  341.       " None edge weight specified. Set all weight to 1.\n",

  342.       "\n",

  343.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 4240.77it/s]\n",

  344.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 19121.56it/s]\n",

  345.       "\n",

  346.       " --- kernel matrix of random walk kernel of size 183 built in 0.9256701469421387 seconds ---\n",

  347.       "\n",

  348.       "the gram matrix with parameters {'weight': 1e-05, 'compute_method': 'sylvester'} is: \n",

  349.       "[[1.         0.99999986 1.         ... 0.99999882 0.99999882 0.99999882]\n",

  350.       " [0.99999986 1.         0.99999986 ... 0.99999949 0.99999949 0.99999949]\n",

  351.       " [1.         0.99999986 1.         ... 0.99999882 0.99999882 0.99999882]\n",

  352.       " ...\n",

  353.       " [0.99999882 0.99999949 0.99999882 ... 1.         1.         1.        ]\n",

  354.       " [0.99999882 0.99999949 0.99999882 ... 1.         1.         1.        ]\n",

  355.       " [0.99999882 0.99999949 0.99999882 ... 1.         1.         1.        ]]\n"

  356.      ]

  357.     },

  358.     {

  359.      "data": {

  360.       "image/png": "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\n",

  361.       "text/plain": [

  362.        "<Figure size 288x288 with 2 Axes>"

  363.       ]

  364.      },

  365.      "metadata": {},

  366.      "output_type": "display_data"

  367.     },

  368.     {

  369.      "name": "stdout",

  370.      "output_type": "stream",

  371.      "text": [

  372.       "\n",

  373.       " None edge weight specified. Set all weight to 1.\n",

  374.       "\n",

  375.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 4764.24it/s]\n",

  376.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 19201.92it/s]\n",

  377.       "\n",

  378.       " --- kernel matrix of random walk kernel of size 183 built in 0.9170398712158203 seconds ---\n",

  379.       "\n",

  380.       "the gram matrix with parameters {'weight': 3.162277660168379e-06, 'compute_method': 'sylvester'} is: \n",

  381.       "[[1.         0.99999996 1.         ... 0.99999963 0.99999963 0.99999963]\n",

  382.       " [0.99999996 1.         0.99999996 ... 0.99999984 0.99999984 0.99999984]\n",

  383.       " [1.         0.99999996 1.         ... 0.99999963 0.99999963 0.99999963]\n",

  384.       " ...\n",

  385.       " [0.99999963 0.99999984 0.99999963 ... 1.         1.         1.        ]\n",

  386.       " [0.99999963 0.99999984 0.99999963 ... 1.         1.         1.        ]\n",

  387.       " [0.99999963 0.99999984 0.99999963 ... 1.         1.         1.        ]]\n"

  388.      ]

  389.     },

  390.     {

  391.      "data": {

  392.       "image/png": "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\n",

  393.       "text/plain": [

  394.        "<Figure size 288x288 with 2 Axes>"

  395.       ]

  396.      },

  397.      "metadata": {},

  398.      "output_type": "display_data"

  399.     },

  400.     {

  401.      "name": "stdout",

  402.      "output_type": "stream",

  403.      "text": [

  404.       "\n",

  405.       " None edge weight specified. Set all weight to 1.\n",

  406.       "\n",

  407.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 4370.83it/s]\n",

  408.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 19434.31it/s]\n",

  409.       "\n",

  410.       " --- kernel matrix of random walk kernel of size 183 built in 0.9101333618164062 seconds ---\n",

  411.       "\n",

  412.       "the gram matrix with parameters {'weight': 1e-06, 'compute_method': 'sylvester'} is: \n",

  413.       "[[1.         0.99999999 1.         ... 0.99999988 0.99999988 0.99999988]\n",

  414.       " [0.99999999 1.         0.99999999 ... 0.99999995 0.99999995 0.99999995]\n",

  415.       " [1.         0.99999999 1.         ... 0.99999988 0.99999988 0.99999988]\n",

  416.       " ...\n",

  417.       " [0.99999988 0.99999995 0.99999988 ... 1.         1.         1.        ]\n",

  418.       " [0.99999988 0.99999995 0.99999988 ... 1.         1.         1.        ]\n",

  419.       " [0.99999988 0.99999995 0.99999988 ... 1.         1.         1.        ]]\n"

  420.      ]

  421.     },

  422.     {

  423.      "data": {

  424.       "image/png": "iVBORw0KGgoAAAANSUhEUgAAARAAAAD8CAYAAAC/+/tYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAGsJJREFUeJzt3X2MXfV95/H359rjcWwPNq7BNmAICl7y1MQbHNokTUsW0xCkLlA1FNpVoEpEV4H8s4q0RKw2VVfRos1GldJNojoJAlo1hGZD8SoUAt7Vpn8kqY2WJDyEYsKTjR9iwGTA9Xg897t/3HOHO3fOvffcc+/cc87M5yUd+Tz8zu/+rj3z9e/p/I4iAjOzPGpFF8DMqssBxMxycwAxs9wcQMwsNwcQM8vNAcTMcnMAsSVP0u2Sjkh6bAh5fUTSoy3bCUlXZbz37ZJ+KGlK0mcHLcsoyPNAbKmT9NvA68BdEfHuIea7HtgHnBMRx9uuPRcRb207dyZwHnAV8GpE/PdhlWWhuAZiS15E/AB4pfWcpLdJekDSI5L+UdLbc2T9B8A/tAePLuU4EhF7gOkcn1UIBxCzdDuBz0TERcBnga/myONa4FtDLVXJLC+6AGZlI2kN8EHg7yQ1T48n134f+POU2w5ExEdb8tgM/DrwYMu5rwAfSg7PkvRosv93EfGFoX6JEXEAMZuvBhyLiG3tFyLiu8B3M+RxDXBvRMw2RyLipuZ+0gcyL/+qcRPGrE1E/Ap4VtLHAdTw3j6zuY5F3nwBBxAzJH0L+CFwoaT9kj4J/DHwSUk/AR4Hruwjv7cCW4D/22c5NknaD/wH4D8lZTmtj/svl/SUpH2Sbkm5Pi7p28n1HyflHIiHcc0WAUnLgH8GLgP2A3uA6yLiiZY0nwbeExH/XtK1wNUR8YeDfK5rIGaLw8XAvoj4RUScBO5mfq3pSuDOZP87wKVq6SXOwwHEbHE4G3ix5Xh/ci41TUScAl4Dfm2QD/UojFlBPvqR1fHyKzOZ0j7y06nHgRMtp3ZGxM4FKVgfHEDMCnL0lRl+/OA5mdKObX7mRERs75LkAI2O26ZzknNpafZLWg6sBV7OXuL53IQxK0wwE/VMWwZ7gK2Szpe0gsYs2F1taXYB1yf7fwD87xhwFMU1ELOCBFBnOKOgEXFK0s00Zr4uA26PiMcl/TmwNyJ2Ad8E/lrSPhrP/lw76Oc6gJgVqE6m2kUmEXE/cH/buf/csn8C+PjQPhAHELPCBMFMxedhFd4H0mv2XFlJek7Sz5JFY/Ym59ZLekjS08mfpxddzlZpC+d0KnMyffvLyb/LTyW9r7iSv6nDd/gzSQdaFvG5ouXa55Lv8JSkj6bnWpw6kWkrq0IDSDJ77ivAx4B3AtdJemeRZerTRyJiW0vv+C3A7ojYCuxOjsvkDuDytnOdyvwxYGuy3Qh8bURl7OUO5n8HgL9I/i22JVV5kp+la4F3Jfd8NfmZK4UAZohMW1kVXQPJMnuuSlpn+t1JY2Wp0khbOIfOZb6SxgpdERE/AtYlj6gXqsN36ORK4O6ImIqIZ2msDnbxghWuTwFMRz3TVlZFB5Ass+fKKoDvJytW3Zic2xgRB5P9Q8DGYorWl05lrtq/zc1JU+v2lqZj6b9DPeNWVkUHkCr7rYh4H42q/k3JupqzkvH18tY9U1SxzImvAW8DtgEHgS8VW5xsImPzxU2YzrLMniuliDiQ/HkEuJdG1fhws5qf/HmkuBJm1qnMlfm3iYjDETETEXXg67zZTCn3dwiYybiVVdEBJMvsudKRtFrSRHMf+F3gMebO9LseuK+YEvalU5l3AZ9IRmN+E3itpalTKm19M1fT+LeAxne4NlkH43waHcL/NOryddKYSFbtJkyh80A6zZ4rskwZbQTuTZ6EXg78bUQ8IGkPcE+yIM3zNJa1K41k4ZxLgA3JwjWfB24jvcz3A1fQ6Hg8DvzJyAucosN3uETSNhq/k88BfwqQzMS8B3gCOAXcFBHZnl4bCTHDQE/TF84LCpkV5N3vWRH/83sbMqV9+7kHH+nxMF0hPBPVrCABnCy8F2EwDiBmBapHtZswDiBmBWnMRHUAMbMcAjFT8SbMgpW+34fkWmZzVlLVyw/+DkWohzJtZbUgASTnQ3KV+odPUfXyg7/DSDWbMFm2slqoJszsQ3IAkpoPyT3R9S6zJUXMRLWbMAsVQNIeYvqNTok3rF8WZ5w1xvb3roxj9cZf6JGTEwCcuWJydr/V1ImxOcfjK6c7XptP1Kbmn62Pt8+JSU+XZmzidN6yacucDJad7H7PzIpseffKp1+1qfS5jStXrOW01WcHQH289w92p3y60fR06vkY6/5v1um+diuXrWHtijNTJzfF9Kn0vMeWd73ej0lePRoRZ2RJG8A0pVldIJfCOlGTtuqNAGecNcahR85j1xur+LerjwNw8NTrAGxefvbsfqs9U3NfZ/H+8SMdr6V55Pj5885dtOrZTOmyenJyU9fr75g4NJR8+vXi5LqeabZMHBtKPu1enVyVev70ieO57uvH1BvpEXt89cmu1/sxecPnns+aNsI1kE56PsSUvNNiJ8D2966MZvDY9UbjB+UvX/g9AD5z7u7Z/Vb7npq7NMUFFx7seK2dQqx5Zn7kv+uCD2VKl9Vpz3efNf3z8y4cSj79WvtC919WgJfO7T1DMks+7dbv/2Xq+Zlzuv+n3em+fpw6mB6wl2/e1PV6P17oM329xP0bWSxUAJl9SI5G4LgW+KNOiY/Va7PBo1kDef/WxovNNy9fM7s/5wPOPXPO8ZwaSNu1NGWogVxa6hrIS0PJp92rk+mBoncNJFOroKupN9KXAnmzBjKEpUJu+E7mpI1OVNdA5qnwQ3JmI+QmTEdpS8x3cuTkBAdPvc5fvvB7c2oe0OgLae632lGb20ZfVVvT8Vqad6348bxzm5fNbwOnpcvqwMT8crc6e9n8vp08+fTrpeneaz2fNfbqUPJpd+jU2tTzm5a/luu+fhydnt8ZD7BhbLLr9X781z7SNh7ndwAxsxwCcTI8CjOwM1dMsnn52Xzm3N1zah7QqImkj8J06QOZqkYfiEdh3rRYRmEai9NlV3cTZnBuwgyWT7/chJmryCaMO1HNLJdAzJT4OZcsHEDMCuROVDPLJYLKD+NWu/RmlSbqGbeBPiXDO5slbZP0Q0mPJy/o+sMseZeiBjJ1Yow9U7/Gvqc2z84ibXaE7pk6M7VTdFVtbofn8frJjtfSnJvyitSxlHNp6bJaX+s+srCm9pah5NOvLRk6b9fVev/fkiWfdq/V00ee1ta6PyzX6b5+HKs3/r5rqlOP2uyf62r/AsCvYnzO+VY1zX9wsD1tTfW+O1FPxkh+BZvvP74tWZvnFuA/tqU5DnwiIp6WdBbwiKQHI6LriIRrILYkNQNCe2BoPa6pPrv1yqtXmjRBtsWEhrCgUM93NkfEP0fE08n+SzReMNbz+YFS1EDMlqoRDeP29c5mSRcDK4BnemXsAGJL0pzmh+afb+43dathtDZh+hH0NZFsg6S9Lcc7kyfaAZD0MJA24/DWOZ8ZEZI6vgwqecvfXwPXJ68K7aoUAWR85TTvHz/CBRcenJ1R2pwY9v7xI3MmiTW19nk00q/oeC3NwZn5adImkqWly+rAzJAmkvXIp1+eSDbXMCeSNV7kl1VfyxUe7fZiqYjY0fFTpMOSNkfEwW7vbJZ0GvA94NaI+FGWQrkPxKwgzRpIlm1APd/ZnLyb+l7grojIvCaBA4hZgUa0qPJtwGWSngZ2JMdI2i7pG0maa4DfBm6Q9GiybeuVcSmaMB7GHSyffnkYtyTDuCGm6wv/KxgRLwOXppzfC3wq2f8b4G/6zbsUAcRs1LIO4/aTV78a64H4WRizyinDKIxXJDOz3BqdqK6BmFlOXg/ErILK0QdS7vfeZuEAYlYgrwdiZrlEwHTdAWQIGtU4Vbw6Z+XWT1MjS9q8TZemRhPGAWRgtanG6udrnlk2uwp6czHjR46fn7qwcfsEr9ZnVrJM/kqbNDYd818hmZYuq9NqK7teX6ZsPzy98unXuHq/RHpcvX80suTTbqLDhLFV6j75r9N9/TgR6XmslLpeX0hDmGVaqFIEELOlyMO4ZjYAN2HMbACeym5muTRWZXcAGVh9PLho1bPcdcGHZl8v2Vzc56JVz6Yu9NPeudmaJkvHZ9YO07R0Wb1en+p6fU1tfCj59OtYvffowbpa7w7SLPm0e60+lnp+ba37wk2d7utH69O4rU5T4+/3V/Hmv0frMzGdZEnTTSBO1f1uXLPKKcfDdG7CmFVSOaayexTGrJJKUwPxKMwweCaqLby02kWnGkg/M1Fz94MM550vhSpFAPFM1MHy6Zdnos5V1ExUr0hmZgNxDcTMcgnglJ/GNbM8vKCQmQ3EfSBmlk+4D8TMclryE8kkPQdMAjPAqYjYLmk98G3grcBzwDUR0ftNzWZLUNUDyDC6gD8SEdta3hx+C7A7IrYCu5NjM2sTiJl6LdNWVgtRsiuBO5P9O4GrFuAzzBaFOsq0ldWgASSA70t6RNKNybmNEXEw2T8EbBzwM8wWpUg6UbNsZTVoJ+pvRcQBSWcCD0n6eevFiAhJkXZjEnBuBBibOH3AYphVU4wgOPTTLynpNOAJ4O8j4uZeeQ9UA4mIA8mfR4B7gYuBw5I2J4XZDBzpcO/OiNgeEduXrVo9SDHMKipb7WMINZB++iX/C/CDrBnnDiCSVkuaaO4Dvws8BuwCrk+SXQ/cl/czzBa7CGXaBpSpX1LSRTS6HL6fNeNBmjAbgXvVeJJxOfC3EfGApD3APZI+CTwPXDPAZ5gtWiOcB9KzX1JSDfgS8O+AHVkzzh1AIuIXwHtTzr8MXNpPXstOwpOTmzjt+RmenNwEwIGJNUDjfHO/1fra8TnHB2bWdLyWJu0R+bS1Rwd5lL7X4/ozkW0diayP/WfV69H5hcxnLNKXR+i1bEKn+/qxssPj+uMa63p9wfS3qPIGSXtbjndGxM7mgaSHgU0p99065yM790t+Grg/IvYnlYJMPBPVrCBBX52oR1vmWs3PK6JjrUHSYUmbI+Jgl37JDwAflvRpYA2wQtLrEdF1HpcDiFlhRjZE2+yXvI0O/ZIR8cezpZJuALb3Ch6wMBPJzCyjiGzbgG4DLpP0NI3+jdsAJG2X9I1BMnYNxKxAo5gH0qlfMiL2Ap9KOX8HcEeWvB1AzArSqF2Ud5ZpFg4gZgUq8zT1LBxAzApUrzuADGxmBbxj4hA/P+9CLp04BMDZy14HGueb+63W1N4y57g1Tfu1NGlzK9LeVTvIHIxe8zyy5p11vkhWUzGc1zpkyafd8Q5zLVap+7tvO93XjxMdeiNXaqbr9YUSDGWWaaFKEUDMlqrRhqzhcwAxK4o7Uc1sIBWvgjiAmBXINRAzy23E/bZD5wBiVpAIiBIvmJyFA4hZgVwDMbP8HEDMLB9PJBsKr0g2WD798opkcxW2Ihm4BmJmOXkimZkNxDUQM8vNNRAzy801EDPLJXANxMzy80QyM8vPAcTMcnMTxsxyCdBwV6scOQcQs8LINRAzG4D7QMwsNwcQM8vNAcTMcvFEMjMbhCpeA6n2goxmVRcZtwFIWi/pIUlPJ3+e3iHduZK+L+lJSU9IemuvvB1AzAqkyLYN6BZgd0RsBXYnx2nuAr4YEe8ALgaO9MrYAcSsSKFs22CuBO5M9u8ErmpPIOmdwPKIeAggIl6PiJ5L+zmAmBUla/Nl8BrIxog4mOwfAjampPlXwDFJ35X0/yR9UeqxziQl6UStTdV5cXIda184zouT6wB4abrRTHtxct3sfqsty16fc9yapv1amnHNf7P8sfr8ecVp6bLqtWZo1rfbD2sN0zLotfbpsO9rVSP9f/LmmrOdri+o7MFhg6S9Lcc7I2Jn80DSw8CmlPtunfNxESGlNoqWAx8G/jXwAvBt4Abgm90KVYoAYrZU9dG/cTQitne6GBE7On6GdFjS5og4KGkz6X0b+4FHI+IXyT1/D/wmPQKImzBmRRpNE2YXcH2yfz1wX0qaPcA6SWckx/8GeKJXxj0DiKTbJR2R9FjLudRhITV8WdI+ST+V9L5e+ZstVUqexs2yDeg24DJJTwM7kmMkbZf0DYCImAE+C+yW9DNAwNd7ZZylBnIHcHnbuU7DQh8DtibbjcDXMuRvtnSNYBQmIl6OiEsjYmtE7IiIV5LzeyPiUy3pHoqI90TEr0fEDRFxslfePQNIRPwAeKXtdKdhoSuBu6LhRzSqRJt7f0WzJWo0TZgFk7cPpNOw0NnAiy3p9ifn5pF0o6S9kvZOn3ojZzHMqm1EE8kWzMCdqBGRK0ZGxM6I2B4R28eWrx60GGbVtERrIIebTZO2YaEDwJaWdOck58ysXcbaR5lrIHnngTSHhW5j7rDQLuBmSXcDvwG81tLU6ag+XmPLxDFeOncDWyZeAuCssVcB2DJxbHa/1bra3NjXmqb9Wppxzf/q62rzJ3alpRuWhcy7rKZjJvV8r4line7rx1Snl2sz1vX6gipxcMii50+wpG8Bl9CYCbcf+DyNwHGPpE8CzwPXJMnvB64A9gHHgT9ZgDKbLRqLflHliLiuw6VLU9IGcNOghTKzalh6dWizMlnsTRgzWyAl7yDNwgHErEgOIGaWmwOImeUh3IQxs7z8btzh8Ipkg+VTJV6RrI1rIGaWmwOImeXlPhAzy88BxMxyKfmj+lk4gJgVyKMwZpab+0DMLD8HEDPLxX0gZpaXkq3KHEDMiuQaiJnl5U5UM8vPw7iD0/Q0r06uYv3+X/LqZOPdvodOrQXg1clVs/utXqsfmnPcmqb9WpqJ2vwVuF+rj2VKl9VYdH8A7HjGVcB75dOvLA+mZVkFfRgPuI1S86G5vNeHziuSmdlAHEDMLK+q10BGXGczszlG8GpLSeslPSTp6eTP+QvsNNL9N0mPS3pS0pcl9RxldgAxK9CIXm15C7A7IrYCu5PjueWQPgh8CHgP8G7g/cDv9MrYAcSsKFlrH4MHkCuBO5P9O4GrOpRmJbACGAfGgMO9Mi5FH0iMjXH6xHFmzjmD0yeOA7Bp+WsAnD5xfHa/1dq20ZHWNO3X0qQtE7i2djJTuqx6jVKs0vxRnzz5LISqjbBkMRPpY6bN0ZdO1xeKGNnTuBtb3lF9CNjYniAifijp/wAHk6L9j4h4slfGpQggZktW9trFBkl7W453RsTO5oGkh4FNKffdOufjIkKa3yiSdAHwDuCc5NRDkj4cEf/YrVAOIGYFUmSOIEcjYnunixGxo+NnSIclbY6Ig5I2A0dSkl0N/CgiXk/u+QfgA0DXAOI+ELOijK4PZBdwfbJ/PXBfSpoXgN+RtFzSGI0O1J5NGAcQswKNaBTmNuAySU8DO5JjJG2X9I0kzXeAZ4CfAT8BfhIR/6tXxm7CmBVpBBPJIuJl4NKU83uBTyX7M8Cf9pu3A4hZgao+E9UBxKwofrXlcPhp3MHy6Zefxs13fUG4BmJmeQg3YcxsENnngZSSA4hZgVwDMbN8/FoHMxtE1UdhenY7S7pd0hFJj7Wc+zNJByQ9mmxXtFz7nKR9kp6S9NGFKrjZYqB6tq2ssoxb3QFcnnL+LyJiW7LdDyDpncC1wLuSe74qVWysz2xUgkYnapatpHoGkIj4AfBKxvyuBO6OiKmIeBbYB1w8QPnMFrURPQuzYAaZOXOzpJ8mTZzmGotnAy+2pNmfnDOzNKN5GnfB5A0gXwPeBmyjsYLRl/rNQNKNkvZK2nuy/i85i2FWXc2JZEuuBhIRhyNiJiLqwNd5s5lyANjSkvSc5FxaHjsjYntEbF9Re0ueYphVW9b+jyr3gaRJVjVquhpojtDsAq6VNC7pfGAr8E+DFdFs8ar6KEzPeSCSvgVcQmNNxv3A54FLJG2j0Tp7jmQdgYh4XNI9wBPAKeCmZJ0BM0tR5uZJFj0DSERcl3L6m13SfwH4wiCFMlsSAqhXO4J4JqpZkaodPxxAzIq06JswoxDTp5h6YwWnDh5i6o3GtJGj0xMATL2xYna/1bH63JGb1jTt19rVVOdEymI+7fd1SpfVyh73nsjYu94rn37V6PnKU6YyfGaWfNp1WrSn10udClnsZxRKPMKSRSkCiNlS5RpIQWpdxra6Xcubp9mwKUDuRC1GPWqzv/D1qKVeaz2fNTi03uuAYguu4j9ilQ0grb/c7b/ozeM8AWCQe8361cerLUupsj1TrbWLtBpI88/m1m++/dxjlsvoXm25YEpRA9HYcsZXn2T55k2Mrz4JwIaxSQDGV5+c3W+1rtbyAJ6Yk2b2WpdBgpWaf/E0Tb15jzqny2pc818TMbcM2Sbp9sqnX1lGNMbp/ZnDHBlZtKMsXZX7OZcsShFAzJYqj8IMQZ55IL+K8TnHrWnar6VJm9+Rdp/ngQyWTzvPA2njGkgxPApjlRegmWoHkMqG9ayjMM2t33wdPGwkRtCJKunjkh6XVJe0vUu6y5PF0PdJuiVL3pWugeS5Bp2DQ/t9DiK20EY0jPsY8PvAX3UsR2Px868Al9FYinSPpF0R8US3jCsbQICOTZjmtX6bMK1NFw/j2kiMIIBExJMA6j6ieDGwLyJ+kaS9m8Yi6V0DiH9LzIoSNGaiZtkWXq4F0StdAzGrMhH9NGE2SNrbcrwzInbO5iU9DGxKue/WiLhvgGJ2VdkA4qnstihkDyBHI6JjB2hE7BiwJJkXRG9V2QBiVnkBlGcYdw+wNVkM/QCNN0z+Ua+b3AdiViBFZNoG+gzp6mRB9A8A35P0YHL+LEn3A0TEKeBm4EHgSeCeiHi8V96ugZgVaTSjMPcC96acfwm4ouX4fuD+fvJ2ADErjB+mK4wnklnlBQ4gRfGShrYoVPxHrrIBxA/T2WJQ9RXJKhtAzCovgJlq/ydV2QDiiWRWfdXvRFWU4AtI+iXwBnC06LIMYAPVLj/4OwzDeRFxRpaEa1duig9u+USmTB/Y98VHus1ELUopaiARcYakvWX8C8qq6uUHf4dClOA/8EGUIoCYLUkB+MVSZpZPQI+1YMuuTAFkZ+8kpVb18oO/w2h5FGZ4Wtc2qKKqlx/8HQrhPhAzy80BxMzyqf48EAcQs6IEUHcfiJnl5RqImeXmAGJmuUQQMzNFl2IgDiBmRfJMVDPLzU0YM8slwqMwZjYA10DMLK9wDcTM8vFMVDPLKwAP45pZHgGEh3HNLJfwgkJmNoCq10BKsSq72VIk6QEaq8hncTQiLl/I8uThAGJmuXV/C7WZWRcOIGaWmwOImeXmAGJmuTmAmFluDiBmlpsDiJnl5gBiZrk5gJhZbv8frOJmkbqHpCsAAAAASUVORK5CYII=\n",

  425.       "text/plain": [

  426.        "<Figure size 288x288 with 2 Axes>"

  427.       ]

  428.      },

  429.      "metadata": {},

  430.      "output_type": "display_data"

  431.     },

  432.     {

  433.      "name": "stdout",

  434.      "output_type": "stream",

  435.      "text": [

  436.       "\n",

  437.       " None edge weight specified. Set all weight to 1.\n",

  438.       "\n",

  439.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 5057.64it/s]\n",

  440.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 19104.93it/s]\n",

  441.       "\n",

  442.       " --- kernel matrix of random walk kernel of size 183 built in 0.919196367263794 seconds ---\n",

  443.       "\n",

  444.       "the gram matrix with parameters {'weight': 3.162277660168379e-07, 'compute_method': 'sylvester'} is: \n",

  445.       "[[1.         1.         1.         ... 0.99999996 0.99999996 0.99999996]\n",

  446.       " [1.         1.         1.         ... 0.99999998 0.99999998 0.99999998]\n",

  447.       " [1.         1.         1.         ... 0.99999996 0.99999996 0.99999996]\n",

  448.       " ...\n",

  449.       " [0.99999996 0.99999998 0.99999996 ... 1.         1.         1.        ]\n",

  450.       " [0.99999996 0.99999998 0.99999996 ... 1.         1.         1.        ]\n",

  451.       " [0.99999996 0.99999998 0.99999996 ... 1.         1.         1.        ]]\n"

  452.      ]

  453.     },

  454.     {

  455.      "data": {

  456.       "image/png": "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\n",

  457.       "text/plain": [

  458.        "<Figure size 288x288 with 2 Axes>"

  459.       ]

  460.      },

  461.      "metadata": {},

  462.      "output_type": "display_data"

  463.     },

  464.     {

  465.      "name": "stdout",

  466.      "output_type": "stream",

  467.      "text": [

  468.       "\n",

  469.       " None edge weight specified. Set all weight to 1.\n",

  470.       "\n",

  471.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 5008.70it/s]\n",

  472.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 18795.77it/s]\n",

  473.       "\n",

  474.       " --- kernel matrix of random walk kernel of size 183 built in 0.934323787689209 seconds ---\n",

  475.       "\n",

  476.       "the gram matrix with parameters {'weight': 1e-07, 'compute_method': 'sylvester'} is: \n",

  477.       "[[1.         1.         1.         ... 0.99999999 0.99999999 0.99999999]\n",

  478.       " [1.         1.         1.         ... 0.99999999 0.99999999 0.99999999]\n",

  479.       " [1.         1.         1.         ... 0.99999999 0.99999999 0.99999999]\n",

  480.       " ...\n",

  481.       " [0.99999999 0.99999999 0.99999999 ... 1.         1.         1.        ]\n",

  482.       " [0.99999999 0.99999999 0.99999999 ... 1.         1.         1.        ]\n",

  483.       " [0.99999999 0.99999999 0.99999999 ... 1.         1.         1.        ]]\n"

  484.      ]

  485.     },

  486.     {

  487.      "data": {

  488.       "image/png": "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\n",

  489.       "text/plain": [

  490.        "<Figure size 288x288 with 2 Axes>"

  491.       ]

  492.      },

  493.      "metadata": {},

  494.      "output_type": "display_data"

  495.     },

  496.     {

  497.      "name": "stdout",

  498.      "output_type": "stream",

  499.      "text": [

  500.       "\n",

  501.       " None edge weight specified. Set all weight to 1.\n",

  502.       "\n",

  503.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 4952.37it/s]\n",

  504.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 19015.32it/s]\n",

  505.       "\n",

  506.       " --- kernel matrix of random walk kernel of size 183 built in 0.92435622215271 seconds ---\n",

  507.       "\n",

  508.       "the gram matrix with parameters {'weight': 3.162277660168379e-08, 'compute_method': 'sylvester'} is: \n",

  509.       "[[1. 1. 1. ... 1. 1. 1.]\n",

  510.       " [1. 1. 1. ... 1. 1. 1.]\n",

  511.       " [1. 1. 1. ... 1. 1. 1.]\n",

  512.       " ...\n",

  513.       " [1. 1. 1. ... 1. 1. 1.]\n",

  514.       " [1. 1. 1. ... 1. 1. 1.]\n",

  515.       " [1. 1. 1. ... 1. 1. 1.]]\n"

  516.      ]

  517.     },

  518.     {

  519.      "data": {

  520.       "image/png": "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\n",

  521.       "text/plain": [

  522.        "<Figure size 288x288 with 2 Axes>"

  523.       ]

  524.      },

  525.      "metadata": {},

  526.      "output_type": "display_data"

  527.     },

  528.     {

  529.      "name": "stdout",

  530.      "output_type": "stream",

  531.      "text": [

  532.       "\n",

  533.       " None edge weight specified. Set all weight to 1.\n",

  534.       "\n",

  535.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 4936.63it/s]\n",

  536.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 18942.83it/s]\n",

  537.       "\n",

  538.       " --- kernel matrix of random walk kernel of size 183 built in 0.9278466701507568 seconds ---\n",

  539.       "\n",

  540.       "the gram matrix with parameters {'weight': 1e-08, 'compute_method': 'sylvester'} is: \n",

  541.       "[[1. 1. 1. ... 1. 1. 1.]\n",

  542.       " [1. 1. 1. ... 1. 1. 1.]\n",

  543.       " [1. 1. 1. ... 1. 1. 1.]\n",

  544.       " ...\n",

  545.       " [1. 1. 1. ... 1. 1. 1.]\n",

  546.       " [1. 1. 1. ... 1. 1. 1.]\n",

  547.       " [1. 1. 1. ... 1. 1. 1.]]\n"

  548.      ]

  549.     },

  550.     {

  551.      "data": {

  552.       "image/png": "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\n",

  553.       "text/plain": [

  554.        "<Figure size 288x288 with 2 Axes>"

  555.       ]

  556.      },

  557.      "metadata": {},

  558.      "output_type": "display_data"

  559.     },

  560.     {

  561.      "name": "stdout",

  562.      "output_type": "stream",

  563.      "text": [

  564.       "\n",

  565.       " None edge weight specified. Set all weight to 1.\n",

  566.       "\n",

  567.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 4757.39it/s]\n",

  568.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 19270.71it/s]\n",

  569.       "\n",

  570.       " --- kernel matrix of random walk kernel of size 183 built in 0.914006233215332 seconds ---\n",

  571.       "\n",

  572.       "the gram matrix with parameters {'weight': 3.1622776601683795e-09, 'compute_method': 'sylvester'} is: \n",

  573.       "[[1. 1. 1. ... 1. 1. 1.]\n",

  574.       " [1. 1. 1. ... 1. 1. 1.]\n",

  575.       " [1. 1. 1. ... 1. 1. 1.]\n",

  576.       " ...\n",

  577.       " [1. 1. 1. ... 1. 1. 1.]\n",

  578.       " [1. 1. 1. ... 1. 1. 1.]\n",

  579.       " [1. 1. 1. ... 1. 1. 1.]]\n"

  580.      ]

  581.     },

  582.     {

  583.      "data": {

  584.       "image/png": "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\n",

  585.       "text/plain": [

  586.        "<Figure size 288x288 with 2 Axes>"

  587.       ]

  588.      },

  589.      "metadata": {},

  590.      "output_type": "display_data"

  591.     },

  592.     {

  593.      "name": "stdout",

  594.      "output_type": "stream",

  595.      "text": [

  596.       "\n",

  597.       " None edge weight specified. Set all weight to 1.\n",

  598.       "\n",

  599.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 2577.69it/s]\n",

  600.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 19000.57it/s]\n",

  601.       "\n",

  602.       " --- kernel matrix of random walk kernel of size 183 built in 0.9592876434326172 seconds ---\n",

  603.       "\n",

  604.       "the gram matrix with parameters {'weight': 1e-09, 'compute_method': 'sylvester'} is: \n",

  605.       "[[1. 1. 1. ... 1. 1. 1.]\n",

  606.       " [1. 1. 1. ... 1. 1. 1.]\n",

  607.       " [1. 1. 1. ... 1. 1. 1.]\n",

  608.       " ...\n",

  609.       " [1. 1. 1. ... 1. 1. 1.]\n",

  610.       " [1. 1. 1. ... 1. 1. 1.]\n",

  611.       " [1. 1. 1. ... 1. 1. 1.]]\n"

  612.      ]

  613.     },

  614.     {

  615.      "data": {

  616.       "image/png": "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\n",

  617.       "text/plain": [

  618.        "<Figure size 288x288 with 2 Axes>"

  619.       ]

  620.      },

  621.      "metadata": {},

  622.      "output_type": "display_data"

  623.     },

  624.     {

  625.      "name": "stdout",

  626.      "output_type": "stream",

  627.      "text": [

  628.       "\n",

  629.       " None edge weight specified. Set all weight to 1.\n",

  630.       "\n",

  631.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 4863.53it/s]\n",

  632.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 18756.36it/s]\n",

  633.       "\n",

  634.       " --- kernel matrix of random walk kernel of size 183 built in 0.937312126159668 seconds ---\n",

  635.       "\n",

  636.       "the gram matrix with parameters {'weight': 3.1622776601683795e-10, 'compute_method': 'sylvester'} is: \n",

  637.       "[[1. 1. 1. ... 1. 1. 1.]\n",

  638.       " [1. 1. 1. ... 1. 1. 1.]\n",

  639.       " [1. 1. 1. ... 1. 1. 1.]\n",

  640.       " ...\n",

  641.       " [1. 1. 1. ... 1. 1. 1.]\n",

  642.       " [1. 1. 1. ... 1. 1. 1.]\n",

  643.       " [1. 1. 1. ... 1. 1. 1.]]\n"

  644.      ]

  645.     },

  646.     {

  647.      "data": {

  648.       "image/png": "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\n",

  649.       "text/plain": [

  650.        "<Figure size 288x288 with 2 Axes>"

  651.       ]

  652.      },

  653.      "metadata": {},

  654.      "output_type": "display_data"

  655.     },

  656.     {

  657.      "name": "stdout",

  658.      "output_type": "stream",

  659.      "text": [

  660.       "\n",

  661.       " None edge weight specified. Set all weight to 1.\n",

  662.       "\n",

  663.       "compute adjacency matrices: 100%|██████████| 183/183 [00:00<00:00, 5068.23it/s]\n",

  664.       "calculating kernels: 100%|██████████| 16836/16836.0 [00:00<00:00, 18680.54it/s]\n",

  665.       "\n",

  666.       " --- kernel matrix of random walk kernel of size 183 built in 0.9394876956939697 seconds ---\n",

  667.       "\n",

  668.       "the gram matrix with parameters {'weight': 1e-10, 'compute_method': 'sylvester'} is: \n",

  669.       "[[1. 1. 1. ... 1. 1. 1.]\n",

  670.       " [1. 1. 1. ... 1. 1. 1.]\n",

  671.       " [1. 1. 1. ... 1. 1. 1.]\n",

  672.       " ...\n",

  673.       " [1. 1. 1. ... 1. 1. 1.]\n",

  674.       " [1. 1. 1. ... 1. 1. 1.]\n",

  675.       " [1. 1. 1. ... 1. 1. 1.]]\n"

  676.      ]

  677.     },

  678.     {

  679.      "data": {

  680.       "image/png": "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\n",

  681.       "text/plain": [

  682.        "<Figure size 288x288 with 2 Axes>"

  683.       ]

  684.      },

  685.      "metadata": {},

  686.      "output_type": "display_data"

  687.     },

  688.     {

  689.      "name": "stdout",

  690.      "output_type": "stream",

  691.      "text": [

  692.       "\n",

  693.       "21 gram matrices are calculated, 2 of which are ignored.\n",

  694.       "\n",

  695.       "3. Fitting and predicting using nested cross validation. This could really take a while...\n",

  696.       "calculate performance:   0%|          | 3/23370 [00:00<17:25, 22.34it/s]"

  697.      ]

  698.     },

  699.     {

  700.      "name": "stderr",

  701.      "output_type": "stream",

  702.      "text": [

  703.       "/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",

  704.       "  warnings.warn(\"Singular matrix in solving dual problem. Using \"\n"

  705.      ]

  706.     },

  707.     {

  708.      "name": "stdout",

  709.      "output_type": "stream",

  710.      "text": [

  711.       "                                                                            \n",

  712.       "4. Getting final performance...\n",

  713.       "best_params_out:  [{'weight': 0.03162277660168379, 'compute_method': 'sylvester'}]\n",

  714.       "best_params_in:  [{'alpha': 3.162277660168379e-07}]\n",

  715.       "\n",

  716.       "best_val_perf:  32.02923522615009\n",

  717.       "best_val_std:  0.5850856589049531\n",

  718.       "final_performance:  [32.99307681717582]\n",

  719.       "final_confidence:  [4.695257115013143]\n",

  720.       "train_performance: [31.036971104287186]\n",

  721.       "train_std:  [0.47758411875469825]\n",

  722.       "\n",

  723.       "time to calculate gram matrix with different hyper-params: 0.93±0.02s\n",

  724.       "time to calculate best gram matrix: 0.91±nans\n",

  725.       "total training time with all hyper-param choices: 485.03s\n",

  726.       "\n"

  727.      ]

  728.     },

  729.     {

  730.      "name": "stderr",

  731.      "output_type": "stream",

  732.      "text": [

  733.       "/usr/local/lib/python3.5/dist-packages/numpy/core/_methods.py:135: RuntimeWarning: Degrees of freedom <= 0 for slice\n",

  734.       "  keepdims=keepdims)\n",

  735.       "/usr/local/lib/python3.5/dist-packages/numpy/core/_methods.py:127: RuntimeWarning: invalid value encountered in double_scalars\n",

  736.       "  ret = ret.dtype.type(ret / rcount)\n"

  737.      ]

  738.     },

  739.     {

  740.      "name": "stdout",

  741.      "output_type": "stream",

  742.      "text": [

  743.       "params                                                                                  train_perf      valid_perf      test_perf         gram_matrix_time\n",

  744.       "--------------------------------------------------------------------------------------  --------------  --------------  --------------  ------------------\n",

  745.       "{'weight': 0.1, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}                     31.50±0.70      34.85±0.95      35.09±5.04                    0.9\n",

  746.       "{'weight': 0.03162277660168379, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}     31.60±0.52      33.37±0.62      34.32±4.75                    0.91\n",

  747.       "{'weight': 0.01, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}                    31.56±0.57      32.57±0.62      32.62±4.84                    0.94\n",

  748.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}   31.81±0.49      32.90±0.48      33.24±4.21                    0.92\n",

  749.       "{'weight': 0.001, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}                   32.05±0.55      32.96±0.62      33.66±5.29                    0.94\n",

  750.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}  32.05±0.50      32.74±0.60      33.40±4.28                    0.95\n",

  751.       "{'weight': 0.0001, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}                  32.36±0.45      32.99±0.51      33.04±4.17                    0.92\n",

  752.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}  32.37±0.46      33.00±0.60      33.37±3.87                    0.95\n",

  753.       "{'weight': 1e-05, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}                   32.39±0.41      33.12±0.45      33.36±3.67                    0.93\n",

  754.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}   32.88±0.57      33.29±0.66      33.15±5.28                    0.92\n",

  755.       "{'weight': 1e-06, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}                   33.96±0.51      34.37±0.64      35.29±4.83                    0.91\n",

  756.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}   34.45±0.61      34.72±0.73      33.96±5.77                    0.92\n",

  757.       "{'weight': 1e-07, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}                   34.46±0.47      34.86±0.54      34.34±4.39                    0.93\n",

  758.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}   34.21±0.48      34.47±0.56      36.58±4.29                    0.92\n",

  759.       "{'weight': 1e-08, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}                   34.40±0.52      34.71±0.65      34.78±4.60                    0.93\n",

  760.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}  34.29±0.56      34.59±0.59      35.58±4.82                    0.91\n",

  761.       "{'weight': 1e-09, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}                   47.18±29.16     50.31±39.36     45.47±28.59                   0.96\n",

  762.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}  35.47±0.55      35.56±0.58      36.47±4.75                    0.94\n",

  763.       "{'weight': 1e-10, 'alpha': '1.00e-10', 'compute_method': 'sylvester'}                   38.63±0.38      38.63±0.44      38.96±4.99                    0.94\n",

  764.       "{'weight': 0.1, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}                     31.49±0.70      34.63±0.87      34.81±4.88                    0.9\n",

  765.       "{'weight': 0.03162277660168379, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}     31.60±0.52      33.21±0.61      34.17±4.68                    0.91\n",

  766.       "{'weight': 0.01, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}                    31.30±0.54      32.31±0.61      32.22±4.74                    0.94\n",

  767.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}   31.82±0.48      32.87±0.48      33.10±4.17                    0.92\n",

  768.       "{'weight': 0.001, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}                   32.07±0.56      32.96±0.63      33.67±5.31                    0.94\n",

  769.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}  32.14±0.50      32.76±0.59      33.42±4.26                    0.95\n",

  770.       "{'weight': 0.0001, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}                  32.38±0.45      33.00±0.51      33.07±4.16                    0.92\n",

  771.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}  32.37±0.46      33.00±0.60      33.36±3.86                    0.95\n",

  772.       "{'weight': 1e-05, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}                   32.48±0.41      33.15±0.45      33.51±3.65                    0.93\n",

  773.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}   33.53±0.58      33.87±0.66      33.83±5.42                    0.92\n",

  774.       "{'weight': 1e-06, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}                   34.19±0.53      34.58±0.66      35.50±4.86                    0.91\n",

  775.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}   34.47±0.61      34.75±0.73      33.99±5.76                    0.92\n",

  776.       "{'weight': 1e-07, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}                   34.46±0.47      34.86±0.54      34.34±4.39                    0.93\n",

  777.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}   34.21±0.48      34.46±0.56      36.55±4.24                    0.92\n",

  778.       "{'weight': 1e-08, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}                   34.40±0.52      34.68±0.65      34.76±4.54                    0.93\n",

  779.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}  43.58±18.79     42.67±16.88     42.30±12.24                   0.91\n",

  780.       "{'weight': 1e-09, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}                   35.70±0.66      35.76±0.61      33.73±5.07                    0.96\n",

  781.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}  38.55±0.53      38.57±0.58      39.36±5.70                    0.94\n",

  782.       "{'weight': 1e-10, 'alpha': '3.16e-10', 'compute_method': 'sylvester'}                   43.08±0.55      42.92±0.61      43.37±6.66                    0.94\n",

  783.       "{'weight': 0.1, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}                     31.51±0.71      34.48±0.81      34.64±4.87                    0.9\n",

  784.       "{'weight': 0.03162277660168379, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}     31.60±0.52      33.04±0.60      34.01±4.61                    0.91\n",

  785.       "{'weight': 0.01, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}                    31.20±0.54      32.28±0.62      31.93±4.63                    0.94\n",

  786.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}   31.88±0.47      32.86±0.48      32.89±4.12                    0.92\n",

  787.       "{'weight': 0.001, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}                   32.03±0.57      32.86±0.64      33.62±5.30                    0.94\n",

  788.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}  32.25±0.50      32.83±0.58      33.51±4.24                    0.95\n",

  789.       "{'weight': 0.0001, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}                  32.39±0.45      33.00±0.51      33.08±4.16                    0.92\n",

  790.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}  32.38±0.46      32.99±0.59      33.35±3.86                    0.95\n",

  791.       "{'weight': 1e-05, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}                   32.84±0.42      33.43±0.44      33.98±3.68                    0.93\n",

  792.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}   34.04±0.60      34.34±0.67      34.37±5.50                    0.92\n",

  793.       "{'weight': 1e-06, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}                   34.28±0.54      34.66±0.67      35.58±4.87                    0.91\n",

  794.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}   34.48±0.61      34.75±0.73      33.98±5.75                    0.92\n",

  795.       "{'weight': 1e-07, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}                   34.45±0.47      34.85±0.54      34.32±4.36                    0.93\n",

  796.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}   34.20±0.48      34.43±0.56      36.47±4.11                    0.92\n",

  797.       "{'weight': 1e-08, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}                   56.91±48.98     54.91±41.64     60.40±70.20                   0.93\n",

  798.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}  35.43±0.47      35.51±0.44      35.75±4.52                    0.91\n",

  799.       "{'weight': 1e-09, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}                   38.81±0.53      38.74±0.48      36.92±5.75                    0.96\n",

  800.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}  43.03±0.65      42.94±0.70      43.66±7.02                    0.94\n",

  801.       "{'weight': 1e-10, 'alpha': '1.00e-09', 'compute_method': 'sylvester'}                   46.03±0.78      45.72±0.84      46.25±7.59                    0.94\n",

  802.       "{'weight': 0.1, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}                     31.58±0.71      34.41±0.78      34.59±4.96                    0.9\n",

  803.       "{'weight': 0.03162277660168379, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}     31.58±0.51      32.86±0.60      33.82±4.55                    0.91\n",

  804.       "{'weight': 0.01, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}                    31.48±0.54      32.61±0.63      32.01±4.62                    0.94\n",

  805.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}   32.04±0.45      32.95±0.49      32.79±4.08                    0.92\n",

  806.       "{'weight': 0.001, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}                   32.00±0.58      32.76±0.65      33.57±5.26                    0.94\n",

  807.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}  32.31±0.50      32.87±0.58      33.56±4.24                    0.95\n",

  808.       "{'weight': 0.0001, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}                  32.39±0.45      33.00±0.51      33.07±4.17                    0.92\n",

  809.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}  32.47±0.47      33.04±0.58      33.37±3.86                    0.95\n",

  810.       "{'weight': 1e-05, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}                   33.47±0.42      33.97±0.45      34.75±3.74                    0.93\n",

  811.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}   34.27±0.61      34.56±0.68      34.62±5.53                    0.92\n",

  812.       "{'weight': 1e-06, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}                   34.30±0.54      34.68±0.67      35.60±4.86                    0.91\n",

  813.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}   34.48±0.61      34.74±0.73      33.94±5.70                    0.92\n",

  814.       "{'weight': 1e-07, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}                   34.45±0.47      34.81±0.54      34.29±4.30                    0.93\n",

  815.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}   52.94±77.69     60.07±112.17    55.88±80.66                   0.92\n",

  816.       "{'weight': 1e-08, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}                   35.56±0.57      35.64±0.69      35.50±4.76                    0.93\n",

  817.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}  38.59±0.44      38.56±0.45      38.34±5.67                    0.91\n",

  818.       "{'weight': 1e-09, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}                   43.26±0.59      43.01±0.56      41.74±6.83                    0.96\n",

  819.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}  46.00±0.83      45.77±0.87      46.49±7.83                    0.94\n",

  820.       "{'weight': 1e-10, 'alpha': '3.16e-09', 'compute_method': 'sylvester'}                   47.28±0.90      46.90±0.96      47.46±7.95                    0.94\n",

  821.       "{'weight': 0.1, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}                     31.70±0.71      34.39±0.78      34.61±5.09                    0.9\n",

  822.       "{'weight': 0.03162277660168379, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}     31.50±0.50      32.64±0.59      33.58±4.50                    0.91\n",

  823.       "{'weight': 0.01, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}                    31.73±0.54      32.87±0.65      32.11±4.65                    0.94\n",

  824.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}   32.15±0.44      33.02±0.50      32.72±4.05                    0.92\n",

  825.       "{'weight': 0.001, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}                   32.09±0.57      32.77±0.66      33.64±5.22                    0.94\n",

  826.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}  32.33±0.50      32.89±0.58      33.57±4.23                    0.95\n",

  827.       "{'weight': 0.0001, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}                  32.40±0.45      32.99±0.50      33.07±4.18                    0.92\n",

  828.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}  32.84±0.47      33.35±0.56      33.63±3.94                    0.95\n",

  829.       "{'weight': 1e-05, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}                   33.96±0.44      34.42±0.46      35.32±3.78                    0.93\n",

  830.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}   34.35±0.61      34.63±0.68      34.70±5.53                    0.92\n",

  831.       "{'weight': 1e-06, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}                   34.30±0.54      34.68±0.67      35.59±4.80                    0.91\n",

  832.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}   34.48±0.61      34.72±0.72      33.84±5.57                    0.92\n",

  833.       "{'weight': 1e-07, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}                   43.56±18.15     43.80±16.96     46.55±25.51                   0.93\n",

  834.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}   35.34±0.44      35.37±0.49      36.87±4.29                    0.92\n",

  835.       "{'weight': 1e-08, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}                   38.64±0.54      38.64±0.68      38.09±6.02                    0.93\n",

  836.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}  43.12±0.59      42.97±0.67      42.60±7.30                    0.91\n",

  837.       "{'weight': 1e-09, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}                   46.19±0.75      45.79±0.74      44.90±7.51                    0.96\n",

  838.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}  47.25±0.93      46.95±0.96      47.68±8.15                    0.94\n",

  839.       "{'weight': 1e-10, 'alpha': '1.00e-08', 'compute_method': 'sylvester'}                   47.71±0.94      47.31±1.00      47.89±8.07                    0.94\n",

  840.       "{'weight': 0.1, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}                     31.85±0.70      34.38±0.79      34.59±5.17                    0.9\n",

  841.       "{'weight': 0.03162277660168379, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}     31.34±0.47      32.35±0.58      33.27±4.49                    0.91\n",

  842.       "{'weight': 0.01, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}                    31.82±0.54      32.91±0.65      32.07±4.68                    0.94\n",

  843.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}   32.15±0.44      32.97±0.49      32.59±3.99                    0.92\n",

  844.       "{'weight': 0.001, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}                   32.20±0.57      32.84±0.67      33.74±5.20                    0.94\n",

  845.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}  32.33±0.50      32.89±0.58      33.55±4.23                    0.95\n",

  846.       "{'weight': 0.0001, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}                  32.48±0.46      33.04±0.50      33.12±4.24                    0.92\n",

  847.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}  33.51±0.47      33.94±0.55      34.18±4.06                    0.95\n",

  848.       "{'weight': 1e-05, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}                   34.19±0.45      34.61±0.47      35.57±3.79                    0.93\n",

  849.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}   34.37±0.61      34.64±0.68      34.72±5.50                    0.92\n",

  850.       "{'weight': 1e-06, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}                   34.30±0.54      34.65±0.66      35.54±4.65                    0.91\n",

  851.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}   58.75±72.06     56.18±60.23     55.00±68.59                   0.92\n",

  852.       "{'weight': 1e-07, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}                   35.60±0.43      35.71±0.46      35.69±4.27                    0.93\n",

  853.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}   38.48±0.48      38.45±0.57      39.54±5.08                    0.92\n",

  854.       "{'weight': 1e-08, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}                   43.16±0.66      43.01±0.79      42.04±7.25                    0.93\n",

  855.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}  46.10±0.81      45.83±0.90      45.46±8.26                    0.91\n",

  856.       "{'weight': 1e-09, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}                   47.43±0.85      46.95±0.85      46.22±7.79                    0.96\n",

  857.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}  47.69±0.97      47.37±0.99      48.10±8.26                    0.94\n",

  858.       "{'weight': 1e-10, 'alpha': '3.16e-08', 'compute_method': 'sylvester'}                   47.86±0.96      47.45±1.01      48.02±8.11                    0.94\n",

  859.       "{'weight': 0.1, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}                     31.95±0.68      34.26±0.77      34.41±5.11                    0.9\n",

  860.       "{'weight': 0.03162277660168379, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}     31.09±0.46      32.05±0.56      32.99±4.54                    0.91\n",

  861.       "{'weight': 0.01, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}                    31.91±0.53      32.91±0.66      31.97±4.72                    0.94\n",

  862.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}   32.15±0.43      32.90±0.48      32.43±3.92                    0.92\n",

  863.       "{'weight': 0.001, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}                   32.26±0.56      32.88±0.67      33.79±5.19                    0.94\n",

  864.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}  32.34±0.50      32.88±0.58      33.48±4.21                    0.95\n",

  865.       "{'weight': 0.0001, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}                  32.85±0.47      33.35±0.49      33.42±4.37                    0.92\n",

  866.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}  34.03±0.48      34.42±0.56      34.63±4.13                    0.95\n",

  867.       "{'weight': 1e-05, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}                   34.26±0.46      34.67±0.48      35.64±3.77                    0.93\n",

  868.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}   34.38±0.61      34.62±0.67      34.70±5.39                    0.92\n",

  869.       "{'weight': 1e-06, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}                   61.55±51.61     65.59±63.72     63.36±49.34                   0.91\n",

  870.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}   35.64±0.54      35.72±0.58      33.86±4.72                    0.92\n",

  871.       "{'weight': 1e-07, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}                   38.66±0.43      38.68±0.44      39.42±5.34                    0.93\n",

  872.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}   43.01±0.57      42.87±0.71      43.70±5.98                    0.92\n",

  873.       "{'weight': 1e-08, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}                   46.18±0.81      45.89±0.91      44.71±7.87                    0.93\n",

  874.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}  47.36±0.94      47.02±1.03      46.67±8.63                    0.91\n",

  875.       "{'weight': 1e-09, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}                   47.86±0.89      47.36±0.89      46.68±7.89                    0.96\n",

  876.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}  47.83±0.98      47.50±1.00      48.24±8.29                    0.94\n",

  877.       "{'weight': 1e-10, 'alpha': '1.00e-07', 'compute_method': 'sylvester'}                   47.90±0.96      47.49±1.02      48.07±8.12                    0.94\n",

  878.       "{'weight': 0.1, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}                     31.95±0.65      34.03±0.72      34.06±4.96                    0.9\n",

  879.       "{'weight': 0.03162277660168379, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}     31.04±0.48      32.03±0.59      32.99±4.70                    0.91\n",

  880.       "{'weight': 0.01, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}                    32.06±0.53      32.97±0.66      31.94±4.77                    0.94\n",

  881.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}   32.24±0.43      32.93±0.46      32.37±3.89                    0.92\n",

  882.       "{'weight': 0.001, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}                   32.28±0.56      32.89±0.67      33.79±5.18                    0.94\n",

  883.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}  32.43±0.51      32.92±0.57      33.39±4.21                    0.95\n",

  884.       "{'weight': 0.0001, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}                  33.51±0.49      33.94±0.51      34.00±4.54                    0.92\n",

  885.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}  34.27±0.49      34.62±0.56      34.81±4.15                    0.95\n",

  886.       "{'weight': 1e-05, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}                   34.29±0.46      34.66±0.47      35.62±3.70                    0.93\n",

  887.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}   42.80±16.32     41.61±13.81     42.52±19.98                   0.92\n",

  888.       "{'weight': 1e-06, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}                   35.48±0.49      35.63±0.55      36.80±4.42                    0.91\n",

  889.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}   38.79±0.45      38.75±0.49      35.87±5.03                    0.92\n",

  890.       "{'weight': 1e-07, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}                   43.02±0.58      42.93±0.61      44.57±6.74                    0.93\n",

  891.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}   46.01±0.70      45.75±0.85      46.45±6.64                    0.92\n",

  892.       "{'weight': 1e-08, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}                   47.45±0.89      47.10±0.97      45.84±8.10                    0.93\n",

  893.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}  47.79±0.98      47.44±1.07      47.09±8.76                    0.91\n",

  894.       "{'weight': 1e-09, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}                   48.00±0.90      47.49±0.90      46.83±7.92                    0.96\n",

  895.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}  47.88±0.98      47.54±1.01      48.28±8.31                    0.94\n",

  896.       "{'weight': 1e-10, 'alpha': '3.16e-07', 'compute_method': 'sylvester'}                   47.92±0.96      47.50±1.02      48.08±8.13                    0.94\n",

  897.       "{'weight': 0.1, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}                     31.89±0.63      33.74±0.68      33.67±4.88                    0.9\n",

  898.       "{'weight': 0.03162277660168379, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}     31.29±0.50      32.32±0.64      33.30±4.91                    0.91\n",

  899.       "{'weight': 0.01, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}                    32.16±0.53      32.99±0.67      31.90±4.77                    0.94\n",

  900.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}   32.36±0.43      33.01±0.45      32.40±3.89                    0.92\n",

  901.       "{'weight': 0.001, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}                   32.30±0.56      32.89±0.66      33.75±5.17                    0.94\n",

  902.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}  32.82±0.50      33.23±0.55      33.48±4.29                    0.95\n",

  903.       "{'weight': 0.0001, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}                  34.03±0.52      34.40±0.53      34.47±4.65                    0.92\n",

  904.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}  34.35±0.49      34.67±0.55      34.82±4.16                    0.95\n",

  905.       "{'weight': 1e-05, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}                   42.85±36.62     43.96±42.30     47.92±62.98                   0.93\n",

  906.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}   35.54±0.58      35.59±0.54      35.96±4.73                    0.92\n",

  907.       "{'weight': 1e-06, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}                   38.53±0.50      38.55±0.58      40.75±4.95                    0.91\n",

  908.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}   43.40±0.50      43.18±0.60      39.41±6.02                    0.92\n",

  909.       "{'weight': 1e-07, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}                   45.88±0.78      45.68±0.80      47.86±7.56                    0.93\n",

  910.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}   47.28±0.78      46.96±0.93      47.61±6.92                    0.92\n",

  911.       "{'weight': 1e-08, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}                   47.90±0.92      47.52±1.00      46.24±8.18                    0.93\n",

  912.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}  47.94±1.00      47.58±1.09      47.23±8.80                    0.91\n",

  913.       "{'weight': 1e-09, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}                   48.04±0.91      47.53±0.91      46.88±7.93                    0.96\n",

  914.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}  47.89±0.99      47.56±1.01      48.29±8.31                    0.94\n",

  915.       "{'weight': 1e-10, 'alpha': '1.00e-06', 'compute_method': 'sylvester'}                   47.92±0.96      47.51±1.02      48.09±8.13                    0.94\n",

  916.       "{'weight': 0.1, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}                     31.80±0.59      33.45±0.66      33.32±4.89                    0.9\n",

  917.       "{'weight': 0.03162277660168379, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}     31.50±0.52      32.51±0.67      33.56±5.18                    0.91\n",

  918.       "{'weight': 0.01, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}                    32.21±0.53      32.94±0.67      31.83±4.74                    0.94\n",

  919.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}   32.42±0.43      33.05±0.44      32.43±3.90                    0.92\n",

  920.       "{'weight': 0.001, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}                   32.40±0.56      32.94±0.65      33.73±5.16                    0.94\n",

  921.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}  33.52±0.50      33.84±0.54      33.92±4.42                    0.95\n",

  922.       "{'weight': 0.0001, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}                  34.27±0.54      34.59±0.55      34.71±4.69                    0.92\n",

  923.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}  45.75±20.38     47.08±30.43     45.36±20.45                   0.95\n",

  924.       "{'weight': 1e-05, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}                   35.42±0.48      35.54±0.47      36.30±4.12                    0.93\n",

  925.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}   38.62±0.49      38.57±0.52      39.20±5.04                    0.92\n",

  926.       "{'weight': 1e-06, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}                   42.87±0.56      42.74±0.65      46.00±6.18                    0.91\n",

  927.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}   46.47±0.63      46.10±0.78      41.89±6.74                    0.92\n",

  928.       "{'weight': 1e-07, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}                   47.09±0.89      46.83±0.90      49.22±7.88                    0.93\n",

  929.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}   47.73±0.82      47.38±0.96      48.01±7.02                    0.92\n",

  930.       "{'weight': 1e-08, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}                   48.05±0.93      47.66±1.00      46.37±8.20                    0.93\n",

  931.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}  47.98±1.00      47.62±1.10      47.27±8.81                    0.91\n",

  932.       "{'weight': 1e-09, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}                   48.06±0.91      47.55±0.91      46.90±7.93                    0.96\n",

  933.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}  47.90±0.99      47.56±1.01      48.30±8.31                    0.94\n",

  934.       "{'weight': 1e-10, 'alpha': '3.16e-06', 'compute_method': 'sylvester'}                   47.92±0.96      47.51±1.02      48.09±8.13                    0.94\n",

  935.       "{'weight': 0.1, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}                     31.61±0.53      33.10±0.63      32.94±4.68                    0.9\n",

  936.       "{'weight': 0.03162277660168379, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}     31.65±0.53      32.59±0.69      33.78±5.54                    0.91\n",

  937.       "{'weight': 0.01, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}                    32.31±0.52      32.96±0.66      31.83±4.70                    0.94\n",

  938.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}   32.46±0.43      33.06±0.43      32.44±3.93                    0.92\n",

  939.       "{'weight': 0.001, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}                   32.81±0.55      33.26±0.63      33.95±5.15                    0.94\n",

  940.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}  34.07±0.51      34.31±0.54      34.32±4.50                    0.95\n",

  941.       "{'weight': 0.0001, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}                  598.78±3017.08  459.73±2268.44  564.55±2838.36                0.92\n",

  942.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}  35.52±0.44      35.63±0.49      35.45±4.75                    0.95\n",

  943.       "{'weight': 1e-05, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}                   38.53±0.44      38.54±0.47      38.96±4.50                    0.93\n",

  944.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}   43.06±0.54      42.88±0.71      43.83±6.11                    0.92\n",

  945.       "{'weight': 1e-06, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}                   45.73±0.74      45.46±0.80      49.26±7.06                    0.91\n",

  946.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}   47.78±0.71      47.33±0.88      42.96±7.03                    0.92\n",

  947.       "{'weight': 1e-07, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}                   47.51±0.94      47.23±0.94      49.70±7.98                    0.93\n",

  948.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}   47.87±0.83      47.52±0.97      48.14±7.05                    0.92\n",

  949.       "{'weight': 1e-08, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}                   48.09±0.93      47.70±1.01      46.41±8.21                    0.93\n",

  950.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}  48.00±1.00      47.63±1.10      47.28±8.82                    0.91\n",

  951.       "{'weight': 1e-09, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}                   48.06±0.91      47.55±0.91      46.90±7.93                    0.96\n",

  952.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}  47.90±0.99      47.56±1.01      48.30±8.31                    0.94\n",

  953.       "{'weight': 1e-10, 'alpha': '1.00e-05', 'compute_method': 'sylvester'}                   47.92±0.96      47.51±1.02      48.09±8.13                    0.94\n",

  954.       "{'weight': 0.1, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}                     31.31±0.48      32.65±0.60      32.49±4.10                    0.9\n",

  955.       "{'weight': 0.03162277660168379, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}     31.85±0.56      32.71±0.71      34.05±5.77                    0.91\n",

  956.       "{'weight': 0.01, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}                    32.42±0.52      33.02±0.65      31.88±4.68                    0.94\n",

  957.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}   32.58±0.43      33.13±0.42      32.54±4.05                    0.92\n",

  958.       "{'weight': 0.001, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}                   33.53±0.56      33.88±0.62      34.52±5.10                    0.94\n",

  959.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}  48.28±27.74     47.93±25.95     47.43±35.24                   0.95\n",

  960.       "{'weight': 0.0001, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}                  35.50±0.53      35.62±0.54      36.13±5.15                    0.92\n",

  961.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}  38.64±0.50      38.63±0.59      38.29±5.36                    0.95\n",

  962.       "{'weight': 1e-05, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}                   43.08±0.47      42.95±0.53      42.78±5.65                    0.93\n",

  963.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}   45.99±0.70      45.69±0.90      46.82±6.87                    0.92\n",

  964.       "{'weight': 1e-06, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}                   46.93±0.85      46.60±0.89      50.61±7.42                    0.91\n",

  965.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}   48.23±0.74      47.76±0.91      43.33±7.14                    0.92\n",

  966.       "{'weight': 1e-07, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}                   47.65±0.95      47.36±0.95      49.85±8.02                    0.93\n",

  967.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}   47.92±0.83      47.56±0.97      48.18±7.06                    0.92\n",

  968.       "{'weight': 1e-08, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}                   48.11±0.93      47.72±1.01      46.42±8.21                    0.93\n",

  969.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}  48.00±1.01      47.64±1.10      47.29±8.82                    0.91\n",

  970.       "{'weight': 1e-09, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}                   48.06±0.91      47.55±0.91      46.90±7.94                    0.96\n",

  971.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}  47.90±0.99      47.56±1.01      48.30±8.31                    0.94\n",

  972.       "{'weight': 1e-10, 'alpha': '3.16e-05', 'compute_method': 'sylvester'}                   47.92±0.96      47.51±1.02      48.09±8.13                    0.94\n",

  973.       "{'weight': 0.1, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}                     31.21±0.44      32.41±0.60      32.32±3.62                    0.9\n",

  974.       "{'weight': 0.03162277660168379, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}     32.02±0.59      32.77±0.73      34.24±5.76                    0.91\n",

  975.       "{'weight': 0.01, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}                    32.50±0.52      33.05±0.65      31.94±4.65                    0.94\n",

  976.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}   33.07±0.44      33.52±0.41      33.02±4.38                    0.92\n",

  977.       "{'weight': 0.001, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}                   50.42±36.16     52.79±46.99     48.27±28.35                   0.94\n",

  978.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}  35.59±0.53      35.62±0.58      35.82±4.85                    0.95\n",

  979.       "{'weight': 0.0001, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}                  38.55±0.58      38.63±0.62      39.05±6.06                    0.92\n",

  980.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}  43.14±0.55      42.92±0.63      42.65±5.77                    0.95\n",

  981.       "{'weight': 1e-05, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}                   46.13±0.65      45.85±0.69      45.26±6.62                    0.93\n",

  982.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}   47.24±0.81      46.86±0.99      48.07±7.19                    0.92\n",

  983.       "{'weight': 1e-06, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}                   47.35±0.90      47.00±0.93      51.07±7.55                    0.91\n",

  984.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}   48.38±0.75      47.90±0.93      43.45±7.17                    0.92\n",

  985.       "{'weight': 1e-07, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}                   47.69±0.96      47.41±0.95      49.90±8.03                    0.93\n",

  986.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}   47.93±0.83      47.58±0.98      48.20±7.07                    0.92\n",

  987.       "{'weight': 1e-08, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}                   48.11±0.93      47.72±1.01      46.43±8.21                    0.93\n",

  988.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}  48.00±1.01      47.64±1.10      47.29±8.82                    0.91\n",

  989.       "{'weight': 1e-09, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}                   48.07±0.91      47.55±0.91      46.90±7.94                    0.96\n",

  990.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}  47.90±0.99      47.56±1.01      48.30±8.31                    0.94\n",

  991.       "{'weight': 1e-10, 'alpha': '1.00e-04', 'compute_method': 'sylvester'}                   47.92±0.96      47.51±1.02      48.09±8.13                    0.94\n",

  992.       "{'weight': 0.1, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}                     31.32±0.41      32.41±0.57      32.45±3.63                    0.9\n",

  993.       "{'weight': 0.03162277660168379, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}     32.16±0.60      32.79±0.72      34.35±5.59                    0.91\n",

  994.       "{'weight': 0.01, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}                    32.73±0.52      33.20±0.63      32.16±4.62                    0.94\n",

  995.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}   41.00±10.54     40.65±10.66     38.51±9.40                    0.92\n",

  996.       "{'weight': 0.001, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}                   35.44±0.63      35.56±0.61      36.11±4.90                    0.94\n",

  997.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}  38.62±0.52      38.60±0.61      39.12±6.10                    0.95\n",

  998.       "{'weight': 0.0001, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}                  43.05±0.70      43.04±0.74      43.09±6.91                    0.92\n",

  999.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}  46.13±0.63      45.72±0.67      45.54±6.13                    0.95\n",

  1000.       "{'weight': 1e-05, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}                   47.42±0.77      47.07±0.79      46.29±7.02                    0.93\n",

  1001.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}   47.67±0.85      47.27±1.03      48.51±7.29                    0.92\n",

  1002.       "{'weight': 1e-06, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}                   47.49±0.91      47.13±0.94      51.22±7.59                    0.91\n",

  1003.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}   48.43±0.75      47.94±0.93      43.49±7.18                    0.92\n",

  1004.       "{'weight': 1e-07, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}                   47.71±0.96      47.42±0.96      49.92±8.03                    0.93\n",

  1005.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}   47.94±0.83      47.58±0.98      48.20±7.07                    0.92\n",

  1006.       "{'weight': 1e-08, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}                   48.12±0.93      47.72±1.01      46.43±8.21                    0.93\n",

  1007.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}  48.00±1.01      47.64±1.10      47.29±8.82                    0.91\n",

  1008.       "{'weight': 1e-09, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}                   48.07±0.91      47.55±0.91      46.90±7.94                    0.96\n",

  1009.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}  47.90±0.99      47.56±1.01      48.30±8.31                    0.94\n",

  1010.       "{'weight': 1e-10, 'alpha': '3.16e-04', 'compute_method': 'sylvester'}                   47.92±0.96      47.51±1.02      48.09±8.13                    0.94\n",

  1011.       "{'weight': 0.1, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}                     31.72±0.42      32.67±0.55      32.93±4.01                    0.9\n",

  1012.       "{'weight': 0.03162277660168379, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}     32.36±0.58      32.88±0.71      34.48±5.32                    0.91\n",

  1013.       "{'weight': 0.01, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}                    41.37±20.18     40.32±15.94     39.17±17.58                   0.94\n",

  1014.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}   35.50±0.50      35.62±0.49      35.71±5.17                    0.92\n",

  1015.       "{'weight': 0.001, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}                   38.59±0.54      38.60±0.54      38.76±5.95                    0.94\n",

  1016.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}  43.04±0.68      42.89±0.80      43.65±7.66                    0.95\n",

  1017.       "{'weight': 0.0001, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}                  46.07±0.82      45.94±0.86      45.73±7.39                    0.92\n",

  1018.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}  47.39±0.69      46.89±0.70      46.75±6.32                    0.95\n",

  1019.       "{'weight': 1e-05, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}                   47.87±0.82      47.50±0.83      46.66±7.16                    0.93\n",

  1020.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}   47.81±0.86      47.41±1.04      48.65±7.33                    0.92\n",

  1021.       "{'weight': 1e-06, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}                   47.53±0.92      47.17±0.95      51.27±7.60                    0.91\n",

  1022.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}   48.44±0.75      47.96±0.93      43.50±7.18                    0.92\n",

  1023.       "{'weight': 1e-07, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}                   47.71±0.96      47.42±0.96      49.92±8.03                    0.93\n",

  1024.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}   47.94±0.83      47.58±0.98      48.20±7.07                    0.92\n",

  1025.       "{'weight': 1e-08, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}                   48.12±0.93      47.72±1.01      46.43±8.21                    0.93\n",

  1026.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}  48.00±1.01      47.64±1.10      47.29±8.82                    0.91\n",

  1027.       "{'weight': 1e-09, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}                   48.07±0.91      47.55±0.91      46.90±7.94                    0.96\n",

  1028.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}  47.90±0.99      47.56±1.01      48.30±8.31                    0.94\n",

  1029.       "{'weight': 1e-10, 'alpha': '1.00e-03', 'compute_method': 'sylvester'}                   47.92±0.96      47.51±1.02      48.09±8.13                    0.94\n",

  1030.       "{'weight': 0.1, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}                     32.58±0.46      33.36±0.55      34.02±4.39                    0.9\n",

  1031.       "{'weight': 0.03162277660168379, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}     33.18±0.55      33.53±0.67      35.10±5.15                    0.91\n",

  1032.       "{'weight': 0.01, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}                    35.63±0.56      35.74±0.62      34.78±4.81                    0.94\n",

  1033.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}   38.64±0.53      38.64±0.54      38.69±5.76                    0.92\n",

  1034.       "{'weight': 0.001, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}                   43.09±0.64      42.94±0.72      42.67±7.39                    0.94\n",

  1035.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}  45.98±0.93      45.69±1.04      46.56±8.53                    0.95\n",

  1036.       "{'weight': 0.0001, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}                  47.36±0.89      47.16±0.93      46.84±7.58                    0.92\n",

  1037.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}  47.83±0.71      47.30±0.72      47.17±6.38                    0.95\n",

  1038.       "{'weight': 1e-05, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}                   48.02±0.84      47.64±0.84      46.77±7.21                    0.93\n",

  1039.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}   47.86±0.87      47.45±1.05      48.69±7.34                    0.92\n",

  1040.       "{'weight': 1e-06, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}                   47.55±0.92      47.19±0.95      51.29±7.61                    0.91\n",

  1041.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}   48.45±0.75      47.96±0.93      43.51±7.19                    0.92\n",

  1042.       "{'weight': 1e-07, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}                   47.71±0.96      47.42±0.96      49.92±8.03                    0.93\n",

  1043.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}   47.94±0.83      47.58±0.98      48.20±7.07                    0.92\n",

  1044.       "{'weight': 1e-08, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}                   48.12±0.93      47.72±1.01      46.43±8.21                    0.93\n",

  1045.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}  48.00±1.01      47.64±1.10      47.29±8.82                    0.91\n",

  1046.       "{'weight': 1e-09, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}                   48.07±0.91      47.55±0.91      46.90±7.94                    0.96\n",

  1047.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}  47.90±0.99      47.56±1.01      48.30±8.31                    0.94\n",

  1048.       "{'weight': 1e-10, 'alpha': '3.16e-03', 'compute_method': 'sylvester'}                   47.92±0.96      47.51±1.02      48.09±8.13                    0.94\n",

  1049.       "{'weight': 0.1, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}                     33.97±0.49      34.50±0.52      35.90±4.67                    0.9\n",

  1050.       "{'weight': 0.03162277660168379, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}     59.15±45.48     55.11±36.72     60.71±55.90                   0.91\n",

  1051.       "{'weight': 0.01, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}                    38.95±0.51      38.97±0.57      37.74±5.47                    0.94\n",

  1052.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}   43.10±0.61      42.93±0.63      42.47±6.63                    0.92\n",

  1053.       "{'weight': 0.001, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}                   46.10±0.84      45.78±0.93      45.31±8.21                    0.94\n",

  1054.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}  47.23±1.06      46.86±1.17      47.78±8.87                    0.95\n",

  1055.       "{'weight': 0.0001, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}                  47.81±0.92      47.58±0.96      47.22±7.65                    0.92\n",

  1056.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}  47.98±0.72      47.44±0.73      47.30±6.40                    0.95\n",

  1057.       "{'weight': 1e-05, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}                   48.07±0.84      47.68±0.85      46.81±7.22                    0.93\n",

  1058.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}   47.87±0.87      47.46±1.05      48.71±7.34                    0.92\n",

  1059.       "{'weight': 1e-06, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}                   47.55±0.92      47.19±0.95      51.29±7.61                    0.91\n",

  1060.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}   48.45±0.75      47.96±0.93      43.51±7.19                    0.92\n",

  1061.       "{'weight': 1e-07, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}                   47.71±0.96      47.42±0.96      49.92±8.03                    0.93\n",

  1062.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}   47.94±0.83      47.58±0.98      48.20±7.07                    0.92\n",

  1063.       "{'weight': 1e-08, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}                   48.12±0.93      47.72±1.01      46.43±8.21                    0.93\n",

  1064.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}  48.00±1.01      47.64±1.10      47.29±8.82                    0.91\n",

  1065.       "{'weight': 1e-09, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}                   48.07±0.91      47.55±0.91      46.90±7.94                    0.96\n",

  1066.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}  47.90±0.99      47.56±1.01      48.30±8.31                    0.94\n",

  1067.       "{'weight': 1e-10, 'alpha': '1.00e-02', 'compute_method': 'sylvester'}                   47.92±0.96      47.51±1.02      48.09±8.13                    0.94\n",

  1068.       "{'weight': 0.1, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}                     36.87±0.51      37.17±0.50      39.60±5.10                    0.9\n",

  1069.       "{'weight': 0.03162277660168379, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}     39.29±0.65      39.39±0.73      40.68±6.83                    0.91\n",

  1070.       "{'weight': 0.01, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}                    43.27±0.56      43.17±0.62      41.49±6.30                    0.94\n",

  1071.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}   46.14±0.73      45.80±0.75      44.95±7.20                    0.92\n",

  1072.       "{'weight': 0.001, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}                   47.37±0.95      46.98±1.04      46.43±8.53                    0.94\n",

  1073.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}  47.67±1.11      47.27±1.22      48.20±8.99                    0.95\n",

  1074.       "{'weight': 0.0001, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}                  47.95±0.93      47.72±0.96      47.35±7.67                    0.92\n",

  1075.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}  48.02±0.73      47.48±0.73      47.35±6.41                    0.95\n",

  1076.       "{'weight': 1e-05, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}                   48.08±0.84      47.70±0.85      46.82±7.23                    0.93\n",

  1077.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}   47.87±0.87      47.47±1.05      48.71±7.34                    0.92\n",

  1078.       "{'weight': 1e-06, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}                   47.55±0.92      47.19±0.95      51.29±7.61                    0.91\n",

  1079.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}   48.45±0.75      47.96±0.93      43.51±7.19                    0.92\n",

  1080.       "{'weight': 1e-07, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}                   47.71±0.96      47.42±0.96      49.92±8.03                    0.93\n",

  1081.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}   47.94±0.83      47.58±0.98      48.20±7.07                    0.92\n",

  1082.       "{'weight': 1e-08, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}                   48.12±0.93      47.72±1.01      46.43±8.21                    0.93\n",

  1083.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}  48.00±1.01      47.64±1.10      47.29±8.82                    0.91\n",

  1084.       "{'weight': 1e-09, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}                   48.07±0.91      47.55±0.91      46.90±7.93                    0.96\n",

  1085.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}  47.90±0.99      47.56±1.01      48.30±8.31                    0.94\n",

  1086.       "{'weight': 1e-10, 'alpha': '3.16e-02', 'compute_method': 'sylvester'}                   47.92±0.96      47.51±1.02      48.09±8.13                    0.94\n",

  1087.       "{'weight': 0.1, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}                     48.43±18.13     48.22±16.40     52.59±23.61                   0.9\n",

  1088.       "{'weight': 0.03162277660168379, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}     43.13±0.78      43.09±0.82      44.57±7.59                    0.91\n",

  1089.       "{'weight': 0.01, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}                    46.24±0.67      46.00±0.74      44.05±6.81                    0.94\n",

  1090.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}   47.44±0.80      47.01±0.81      46.00±7.43                    0.92\n",

  1091.       "{'weight': 0.001, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}                   47.82±0.99      47.40±1.08      46.83±8.63                    0.94\n",

  1092.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}  47.81±1.13      47.41±1.23      48.34±9.02                    0.95\n",

  1093.       "{'weight': 0.0001, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}                  48.00±0.93      47.77±0.97      47.39±7.68                    0.92\n",

  1094.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}  48.04±0.73      47.49±0.73      47.36±6.41                    0.95\n",

  1095.       "{'weight': 1e-05, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}                   48.09±0.84      47.70±0.85      46.83±7.22                    0.93\n",

  1096.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}   47.88±0.87      47.47±1.05      48.71±7.34                    0.92\n",

  1097.       "{'weight': 1e-06, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}                   47.55±0.92      47.19±0.95      51.29±7.60                    0.91\n",

  1098.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}   48.45±0.75      47.96±0.93      43.51±7.18                    0.92\n",

  1099.       "{'weight': 1e-07, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}                   47.71±0.96      47.43±0.96      49.92±8.03                    0.93\n",

  1100.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}   47.94±0.83      47.58±0.98      48.20±7.07                    0.92\n",

  1101.       "{'weight': 1e-08, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}                   48.12±0.93      47.72±1.01      46.43±8.21                    0.93\n",

  1102.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}  48.00±1.01      47.64±1.10      47.29±8.82                    0.91\n",

  1103.       "{'weight': 1e-09, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}                   48.07±0.91      47.55±0.91      46.90±7.93                    0.96\n",

  1104.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}  47.90±0.99      47.57±1.01      48.30±8.31                    0.94\n",

  1105.       "{'weight': 1e-10, 'alpha': '1.00e-01', 'compute_method': 'sylvester'}                   47.92±0.96      47.51±1.02      48.08±8.13                    0.94\n",

  1106.       "{'weight': 0.1, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}                     43.46±0.67      43.51±0.69      46.93±6.52                    0.9\n",

  1107.       "{'weight': 0.03162277660168379, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}     45.84±0.90      45.61±0.93      47.32±8.05                    0.91\n",

  1108.       "{'weight': 0.01, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}                    47.53±0.73      47.22±0.81      45.16±7.01                    0.94\n",

  1109.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}   47.90±0.83      47.44±0.84      46.37±7.50                    0.92\n",

  1110.       "{'weight': 0.001, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}                   47.97±1.00      47.54±1.09      46.96±8.65                    0.94\n",

  1111.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}  47.85±1.13      47.45±1.24      48.38±9.02                    0.95\n",

  1112.       "{'weight': 0.0001, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}                  48.02±0.93      47.78±0.97      47.40±7.69                    0.92\n",

  1113.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}  48.04±0.73      47.50±0.73      47.35±6.39                    0.95\n",

  1114.       "{'weight': 1e-05, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}                   48.09±0.84      47.71±0.85      46.83±7.21                    0.93\n",

  1115.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}   47.88±0.87      47.47±1.05      48.70±7.33                    0.92\n",

  1116.       "{'weight': 1e-06, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}                   47.55±0.92      47.19±0.95      51.29±7.59                    0.91\n",

  1117.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}   48.45±0.75      47.96±0.93      43.51±7.17                    0.92\n",

  1118.       "{'weight': 1e-07, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}                   47.71±0.96      47.43±0.96      49.91±8.03                    0.93\n",

  1119.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}   47.94±0.83      47.58±0.97      48.21±7.07                    0.92\n",

  1120.       "{'weight': 1e-08, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}                   48.12±0.93      47.73±1.01      46.44±8.20                    0.93\n",

  1121.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}  48.00±1.01      47.64±1.10      47.29±8.81                    0.91\n",

  1122.       "{'weight': 1e-09, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}                   48.07±0.91      47.55±0.91      46.91±7.93                    0.96\n",

  1123.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}  47.90±0.99      47.57±1.01      48.31±8.30                    0.94\n",

  1124.       "{'weight': 1e-10, 'alpha': '3.16e-01', 'compute_method': 'sylvester'}                   47.92±0.96      47.51±1.02      48.08±8.12                    0.94\n",

  1125.       "{'weight': 0.1, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}                     45.43±0.79      45.31±0.80      48.63±7.34                    0.9\n",

  1126.       "{'weight': 0.03162277660168379, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}     47.10±0.97      46.78±1.01      48.62±8.26                    0.91\n",

  1127.       "{'weight': 0.01, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}                    47.99±0.76      47.66±0.83      45.58±7.05                    0.94\n",

  1128.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}   48.06±0.84      47.59±0.85      46.51±7.52                    0.92\n",

  1129.       "{'weight': 0.001, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}                   48.02±1.01      47.59±1.09      47.03±8.60                    0.94\n",

  1130.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}  47.88±1.13      47.48±1.24      48.41±8.99                    0.95\n",

  1131.       "{'weight': 0.0001, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}                  48.03±0.93      47.80±0.97      47.40±7.72                    0.92\n",

  1132.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}  48.05±0.73      47.51±0.72      47.32±6.35                    0.95\n",

  1133.       "{'weight': 1e-05, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}                   48.10±0.84      47.72±0.85      46.83±7.17                    0.93\n",

  1134.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}   47.88±0.87      47.48±1.05      48.67±7.30                    0.92\n",

  1135.       "{'weight': 1e-06, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}                   47.56±0.92      47.20±0.95      51.30±7.56                    0.91\n",

  1136.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}   48.46±0.75      47.97±0.93      43.52±7.15                    0.92\n",

  1137.       "{'weight': 1e-07, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}                   47.72±0.96      47.44±0.95      49.87±8.03                    0.93\n",

  1138.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}   47.95±0.83      47.60±0.97      48.22±7.07                    0.92\n",

  1139.       "{'weight': 1e-08, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}                   48.12±0.93      47.74±1.01      46.46±8.18                    0.93\n",

  1140.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}  48.01±1.01      47.65±1.10      47.28±8.79                    0.91\n",

  1141.       "{'weight': 1e-09, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}                   48.07±0.91      47.56±0.91      46.91±7.91                    0.96\n",

  1142.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}  47.91±0.99      47.58±1.01      48.33±8.29                    0.94\n",

  1143.       "{'weight': 1e-10, 'alpha': '1.00e+00', 'compute_method': 'sylvester'}                   47.93±0.96      47.52±1.02      48.06±8.11                    0.94\n",

  1144.       "{'weight': 0.1, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}                     46.74±0.90      46.49±0.89      49.73±7.73                    0.9\n",

  1145.       "{'weight': 0.03162277660168379, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}     47.63±1.01      47.28±1.04      49.28±8.31                    0.91\n",

  1146.       "{'weight': 0.01, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}                    48.22±0.77      47.88±0.84      45.82±7.00                    0.94\n",

  1147.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}   48.18±0.84      47.71±0.85      46.66±7.53                    0.92\n",

  1148.       "{'weight': 0.001, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}                   48.11±1.01      47.69±1.09      47.17±8.44                    0.94\n",

  1149.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}  47.95±1.13      47.57±1.23      48.50±8.90                    0.95\n",

  1150.       "{'weight': 0.0001, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}                  48.10±0.93      47.88±0.96      47.43±7.82                    0.92\n",

  1151.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}  48.13±0.72      47.60±0.71      47.28±6.23                    0.95\n",

  1152.       "{'weight': 1e-05, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}                   48.17±0.84      47.80±0.86      46.90±7.05                    0.93\n",

  1153.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}   47.96±0.87      47.56±1.06      48.63±7.21                    0.92\n",

  1154.       "{'weight': 1e-06, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}                   47.63±0.92      47.28±0.95      51.35±7.46                    0.91\n",

  1155.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}   48.53±0.75      48.05±0.94      43.61±7.08                    0.92\n",

  1156.       "{'weight': 1e-07, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}                   47.79±0.96      47.51±0.95      49.81±8.03                    0.93\n",

  1157.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}   48.02±0.83      47.68±0.96      48.32±7.09                    0.92\n",

  1158.       "{'weight': 1e-08, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}                   48.20±0.93      47.82±1.02      46.60±8.13                    0.93\n",

  1159.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}  48.09±1.00      47.72±1.11      47.31±8.72                    0.91\n",

  1160.       "{'weight': 1e-09, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}                   48.15±0.91      47.64±0.91      46.99±7.85                    0.96\n",

  1161.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}  47.98±0.98      47.66±1.01      48.46±8.26                    0.94\n",

  1162.       "{'weight': 1e-10, 'alpha': '3.16e+00', 'compute_method': 'sylvester'}                   48.00±0.96      47.60±1.00      48.04±8.08                    0.94\n",

  1163.       "{'weight': 0.1, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}                     47.97±0.94      47.66±0.94      50.88±7.60                    0.9\n",

  1164.       "{'weight': 0.03162277660168379, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}     48.44±1.00      48.11±1.00      50.44±8.33                    0.91\n",

  1165.       "{'weight': 0.01, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}                    48.92±0.76      48.58±0.82      46.63±6.90                    0.94\n",

  1166.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}   48.84±0.83      48.38±0.85      47.42±7.59                    0.92\n",

  1167.       "{'weight': 0.001, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}                   48.77±0.99      48.36±1.07      47.99±7.99                    0.94\n",

  1168.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}  48.61±1.11      48.25±1.20      49.16±8.74                    0.95\n",

  1169.       "{'weight': 0.0001, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}                  48.76±0.92      48.53±0.95      47.95±8.20                    0.92\n",

  1170.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}  48.78±0.71      48.29±0.67      47.59±6.02                    0.95\n",

  1171.       "{'weight': 1e-05, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}                   48.82±0.83      48.46±0.86      47.53±6.80                    0.93\n",

  1172.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}   48.62±0.85      48.22±1.07      48.92±7.08                    0.92\n",

  1173.       "{'weight': 1e-06, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}                   48.30±0.90      47.95±0.95      51.92±7.23                    0.91\n",

  1174.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}   49.18±0.74      48.70±0.93      44.32±6.97                    0.92\n",

  1175.       "{'weight': 1e-07, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}                   48.46±0.94      48.17±0.94      50.05±8.06                    0.93\n",

  1176.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}   48.67±0.82      48.35±0.93      49.03±7.24                    0.92\n",

  1177.       "{'weight': 1e-08, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}                   48.85±0.92      48.48±1.04      47.42±8.05                    0.93\n",

  1178.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}  48.74±0.99      48.36±1.11      47.85±8.51                    0.91\n",

  1179.       "{'weight': 1e-09, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}                   48.80±0.89      48.31±0.91      47.66±7.75                    0.96\n",

  1180.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}  48.64±0.97      48.34±0.99      49.24±8.24                    0.94\n",

  1181.       "{'weight': 1e-10, 'alpha': '1.00e+01', 'compute_method': 'sylvester'}                   48.66±0.95      48.27±0.96      48.45±8.00                    0.94\n",

  1182.       "{'weight': 0.1, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}                     52.94±0.85      52.52±0.88      55.63±7.44                    0.9\n",

  1183.       "{'weight': 0.03162277660168379, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}     53.17±0.92      52.83±0.87      55.85±8.70                    0.91\n",

  1184.       "{'weight': 0.01, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}                    53.60±0.69      53.18±0.71      51.64±7.27                    0.94\n",

  1185.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}   53.52±0.77      53.03±0.83      52.36±8.18                    0.92\n",

  1186.       "{'weight': 0.001, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}                   53.45±0.88      53.01±0.94      53.03±7.23                    0.94\n",

  1187.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}  53.31±1.00      52.93±1.05      53.73±8.95                    0.95\n",

  1188.       "{'weight': 0.0001, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}                  53.45±0.86      53.12±0.89      52.29±9.55                    0.92\n",

  1189.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}  53.49±0.63      53.02±0.57      51.39±6.52                    0.95\n",

  1190.       "{'weight': 1e-05, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}                   53.51±0.73      53.08±0.81      52.14±6.81                    0.93\n",

  1191.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}   53.34±0.76      52.89±0.96      52.62±7.62                    0.92\n",

  1192.       "{'weight': 1e-06, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}                   53.03±0.81      52.60±0.87      56.17±7.09                    0.91\n",

  1193.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}   53.83±0.66      53.31±0.82      49.34±7.28                    0.92\n",

  1194.       "{'weight': 1e-07, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}                   53.19±0.86      52.83±0.86      53.65±8.28                    0.93\n",

  1195.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}   53.36±0.76      53.02±0.81      53.76±8.09                    0.92\n",

  1196.       "{'weight': 1e-08, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}                   53.52±0.84      53.10±0.99      52.52±8.26                    0.93\n",

  1197.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}  53.44±0.89      52.99±1.03      52.30±8.04                    0.91\n",

  1198.       "{'weight': 1e-09, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}                   53.48±0.81      52.99±0.86      52.39±7.86                    0.96\n",

  1199.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}  53.32±0.88      52.99±0.88      54.16±8.51                    0.94\n",

  1200.       "{'weight': 1e-10, 'alpha': '3.16e+01', 'compute_method': 'sylvester'}                   53.37±0.85      52.94±0.82      52.53±7.93                    0.94\n",

  1201.       "{'weight': 0.1, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}                     71.76±0.63      71.23±0.72      73.72±8.44                    0.9\n",

  1202.       "{'weight': 0.03162277660168379, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}     71.75±0.75      71.30±0.73      75.00±9.66                    0.91\n",

  1203.       "{'weight': 0.01, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}                    72.12±0.57      71.58±0.53      70.82±8.28                    0.94\n",

  1204.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}   72.06±0.68      71.55±0.82      71.37±9.47                    0.92\n",

  1205.       "{'weight': 0.001, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}                   72.01±0.64      71.48±0.67      71.95±7.36                    0.94\n",

  1206.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}  71.93±0.77      71.43±0.77      72.02±10.11                   0.95\n",

  1207.       "{'weight': 0.0001, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}                  72.04±0.80      71.50±0.84      70.63±11.35                   0.92\n",

  1208.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}  72.14±0.52      71.64±0.48      68.93±8.76                    0.95\n",

  1209.       "{'weight': 1e-05, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}                   72.08±0.55      71.52±0.64      70.69±7.93                    0.93\n",

  1210.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}   72.02±0.61      71.48±0.68      69.87±9.52                    0.92\n",

  1211.       "{'weight': 1e-06, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}                   71.73±0.62      71.17±0.69      73.70±7.93                    0.91\n",

  1212.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}   72.31±0.55      71.74±0.61      68.82±8.49                    0.92\n",

  1213.       "{'weight': 1e-07, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}                   71.92±0.69      71.42±0.72      70.73±8.86                    0.93\n",

  1214.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}   71.94±0.67      71.51±0.70      72.34±9.58                    0.92\n",

  1215.       "{'weight': 1e-08, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}                   72.04±0.69      71.50±0.82      71.70±9.10                    0.93\n",

  1216.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}  72.05±0.67      71.54±0.74      70.64±7.81                    0.91\n",

  1217.       "{'weight': 1e-09, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}                   72.05±0.65      71.54±0.71      71.12±8.42                    0.96\n",

  1218.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}  71.90±0.71      71.44±0.67      72.92±9.28                    0.94\n",

  1219.       "{'weight': 1e-10, 'alpha': '1.00e+02', 'compute_method': 'sylvester'}                   72.02±0.65      71.47±0.60      70.40±8.05                    0.94\n",

  1220.       "{'weight': 0.1, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}                     102.40±0.76     101.93±0.83     103.76±9.39                   0.9\n",

  1221.       "{'weight': 0.03162277660168379, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}     102.22±0.87     101.78±0.86     105.65±10.18                  0.91\n",

  1222.       "{'weight': 0.01, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}                    102.58±0.69     102.08±0.67     101.85±8.68                   0.94\n",

  1223.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}   102.53±0.83     102.10±0.98     102.29±9.97                   0.92\n",

  1224.       "{'weight': 0.001, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}                   102.50±0.65     102.00±0.70     102.66±7.86                   0.94\n",

  1225.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}  102.47±0.87     102.00±0.88     102.43±10.76                  0.95\n",

  1226.       "{'weight': 0.0001, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}                  102.57±1.00     102.05±1.02     101.28±11.86                  0.92\n",

  1227.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}  102.74±0.73     102.30±0.69     99.21±9.92                    0.95\n",

  1228.       "{'weight': 1e-05, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}                   102.60±0.68     102.07±0.70     101.33±8.76                   0.93\n",

  1229.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}   102.66±0.81     102.17±0.80     99.87±10.54                   0.92\n",

  1230.       "{'weight': 1e-06, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}                   102.36±0.73     101.85±0.76     103.44±8.74                   0.91\n",

  1231.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}   102.74±0.70     102.24±0.74     100.21±9.08                   0.92\n",

  1232.       "{'weight': 1e-07, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}                   102.60±0.75     102.14±0.79     100.49±9.12                   0.93\n",

  1233.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}   102.45±0.81     102.06±0.84     102.93±10.05                  0.92\n",

  1234.       "{'weight': 1e-08, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}                   102.50±0.81     102.00±0.91     102.65±9.55                   0.93\n",

  1235.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}  102.61±0.66     102.17±0.70     101.12±7.89                   0.91\n",

  1236.       "{'weight': 1e-09, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}                   102.56±0.72     102.10±0.73     101.85±8.64                   0.96\n",

  1237.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}  102.40±0.80     101.96±0.75     103.51±9.68                   0.94\n",

  1238.       "{'weight': 1e-10, 'alpha': '3.16e+02', 'compute_method': 'sylvester'}                   102.63±0.67     102.12±0.66     100.66±8.11                   0.94\n",

  1239.       "{'weight': 0.1, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}                     125.22±0.98     124.81±1.04     126.34±9.67                   0.9\n",

  1240.       "{'weight': 0.03162277660168379, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}     124.98±1.06     124.58±1.06     128.45±10.25                  0.91\n",

  1241.       "{'weight': 0.01, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}                    125.37±0.86     124.92±0.86     124.89±8.70                   0.94\n",

  1242.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}   125.31±1.02     124.95±1.16     125.28±9.98                   0.92\n",

  1243.       "{'weight': 0.001, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}                   125.29±0.80     124.85±0.85     125.55±8.00                   0.94\n",

  1244.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}  125.30±1.08     124.88±1.09     125.23±10.85                  0.95\n",

  1245.       "{'weight': 0.0001, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}                  125.39±1.22     124.92±1.24     124.21±11.83                  0.92\n",

  1246.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}  125.62±0.97     125.23±0.92     122.06±10.12                  0.95\n",

  1247.       "{'weight': 1e-05, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}                   125.42±0.88     124.94±0.89     124.25±8.97                   0.93\n",

  1248.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}   125.55±1.06     125.12±1.04     122.60±10.73                  0.92\n",

  1249.       "{'weight': 1e-06, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}                   125.22±0.92     124.78±0.94     125.95±8.97                   0.91\n",

  1250.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}   125.52±0.90     125.09±0.93     123.42±9.16                   0.92\n",

  1251.       "{'weight': 1e-07, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}                   125.51±0.91     125.10±0.94     123.09±9.09                   0.93\n",

  1252.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}   125.25±0.99     124.92±1.02     125.79±10.02                  0.92\n",

  1253.       "{'weight': 1e-08, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}                   125.28±0.99     124.84±1.07     125.64±9.59                   0.93\n",

  1254.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}  125.45±0.79     125.07±0.83     123.96±7.87                   0.91\n",

  1255.       "{'weight': 1e-09, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}                   125.37±0.87     124.96±0.87     124.77±8.61                   0.96\n",

  1256.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}  125.20±0.99     124.80±0.95     126.34±9.72                   0.94\n",

  1257.       "{'weight': 1e-10, 'alpha': '1.00e+03', 'compute_method': 'sylvester'}                   125.50±0.80     125.05±0.81     123.43±8.06                   0.94\n",

  1258.       "{'weight': 0.1, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}                     135.59±1.09     135.20±1.14     136.63±9.73                   0.9\n",

  1259.       "{'weight': 0.03162277660168379, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}     135.33±1.16     134.95±1.16     138.81±10.24                  0.91\n",

  1260.       "{'weight': 0.01, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}                    135.73±0.95     135.31±0.95     135.35±8.68                   0.94\n",

  1261.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}   135.68±1.12     135.34±1.24     135.71±9.95                   0.92\n",

  1262.       "{'weight': 0.001, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}                   135.67±0.88     135.24±0.93     135.95±8.02                   0.94\n",

  1263.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}  135.68±1.19     135.29±1.20     135.61±10.84                  0.95\n",

  1264.       "{'weight': 0.0001, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}                  135.78±1.33     135.33±1.34     134.63±11.78                  0.92\n",

  1265.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}  136.03±1.07     135.65±1.03     132.47±10.14                  0.95\n",

  1266.       "{'weight': 1e-05, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}                   135.80±0.98     135.34±0.98     134.66±9.01                   0.93\n",

  1267.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}   135.97±1.17     135.55±1.15     132.97±10.76                  0.92\n",

  1268.       "{'weight': 1e-06, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}                   135.62±1.02     135.20±1.03     136.23±9.01                   0.91\n",

  1269.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}   135.89±0.99     135.48±1.02     133.94±9.15                   0.92\n",

  1270.       "{'weight': 1e-07, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}                   135.93±0.99     135.54±1.02     133.41±9.05                   0.93\n",

  1271.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}   135.63±1.08     135.31±1.11     136.19±9.97                   0.92\n",

  1272.       "{'weight': 1e-08, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}                   135.64±1.08     135.23±1.15     136.07±9.57                   0.93\n",

  1273.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}  135.84±0.86     135.48±0.90     134.35±7.84                   0.91\n",

  1274.       "{'weight': 1e-09, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}                   135.75±0.95     135.36±0.94     135.18±8.58                   0.96\n",

  1275.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}  135.57±1.08     135.19±1.04     136.72±9.71                   0.94\n",

  1276.       "{'weight': 1e-10, 'alpha': '3.16e+03', 'compute_method': 'sylvester'}                   135.90±0.87     135.47±0.88     133.80±8.02                   0.94\n",

  1277.       "{'weight': 0.1, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}                     139.32±1.13     138.94±1.18     140.34±9.75                   0.9\n",

  1278.       "{'weight': 0.03162277660168379, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}     139.06±1.20     138.69±1.19     142.55±10.23                  0.91\n",

  1279.       "{'weight': 0.01, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}                    139.47±0.98     139.06±0.98     139.11±8.67                   0.94\n",

  1280.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}   139.42±1.15     139.08±1.28     139.48±9.94                   0.92\n",

  1281.       "{'weight': 0.001, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}                   139.41±0.91     138.98±0.96     139.70±8.03                   0.94\n",

  1282.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}  139.42±1.23     139.04±1.24     139.35±10.83                  0.95\n",

  1283.       "{'weight': 0.0001, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}                  139.52±1.37     139.08±1.38     138.39±11.76                  0.92\n",

  1284.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}  139.78±1.11     139.41±1.07     136.23±10.14                  0.95\n",

  1285.       "{'weight': 1e-05, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}                   139.54±1.02     139.09±1.02     138.42±9.01                   0.93\n",

  1286.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}   139.72±1.22     139.32±1.19     136.71±10.76                  0.92\n",

  1287.       "{'weight': 1e-06, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}                   139.37±1.06     138.96±1.06     139.94±9.02                   0.91\n",

  1288.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}   139.62±1.02     139.23±1.06     137.73±9.14                   0.92\n",

  1289.       "{'weight': 1e-07, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}                   139.69±1.02     139.30±1.05     137.13±9.03                   0.93\n",

  1290.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}   139.36±1.12     139.06±1.14     139.93±9.95                   0.92\n",

  1291.       "{'weight': 1e-08, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}                   139.38±1.11     138.97±1.18     139.83±9.56                   0.93\n",

  1292.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}  139.59±0.88     139.24±0.93     138.10±7.83                   0.91\n",

  1293.       "{'weight': 1e-09, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}                   139.49±0.98     139.11±0.97     138.94±8.57                   0.96\n",

  1294.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}  139.30±1.12     138.94±1.08     140.47±9.70                   0.94\n",

  1295.       "{'weight': 1e-10, 'alpha': '1.00e+04', 'compute_method': 'sylvester'}                   139.65±0.90     139.23±0.91     137.54±8.00                   0.94\n",

  1296.       "{'weight': 0.1, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}                     140.55±1.15     140.18±1.19     141.57±9.75                   0.9\n",

  1297.       "{'weight': 0.03162277660168379, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}     140.29±1.21     139.93±1.20     143.78±10.23                  0.91\n",

  1298.       "{'weight': 0.01, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}                    140.70±0.99     140.30±0.99     140.36±8.67                   0.94\n",

  1299.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}   140.65±1.16     140.32±1.29     140.72±9.93                   0.92\n",

  1300.       "{'weight': 0.001, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}                   140.64±0.92     140.22±0.97     140.94±8.03                   0.94\n",

  1301.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}  140.66±1.24     140.27±1.25     140.58±10.83                  0.95\n",

  1302.       "{'weight': 0.0001, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}                  140.76±1.38     140.32±1.39     139.63±11.75                  0.92\n",

  1303.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}  141.02±1.12     140.65±1.09     137.47±10.14                  0.95\n",

  1304.       "{'weight': 1e-05, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}                   140.78±1.03     140.33±1.03     139.66±9.02                   0.93\n",

  1305.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}   140.96±1.23     140.56±1.21     137.95±10.76                  0.92\n",

  1306.       "{'weight': 1e-06, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}                   140.60±1.07     140.20±1.08     141.17±9.02                   0.91\n",

  1307.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}   140.86±1.04     140.46±1.07     138.98±9.14                   0.92\n",

  1308.       "{'weight': 1e-07, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}                   140.93±1.03     140.55±1.06     138.37±9.03                   0.93\n",

  1309.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}   140.60±1.13     140.29±1.15     141.17±9.94                   0.92\n",

  1310.       "{'weight': 1e-08, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}                   140.61±1.12     140.21±1.20     141.07±9.56                   0.93\n",

  1311.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}  140.83±0.89     140.48±0.93     139.34±7.83                   0.91\n",

  1312.       "{'weight': 1e-09, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}                   140.72±0.99     140.35±0.98     140.18±8.56                   0.96\n",

  1313.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}  140.54±1.13     140.17±1.09     141.70±9.70                   0.94\n",

  1314.       "{'weight': 1e-10, 'alpha': '3.16e+04', 'compute_method': 'sylvester'}                   140.89±0.90     140.47±0.92     138.78±8.00                   0.94\n",

  1315.       "{'weight': 0.1, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}                     140.95±1.15     140.57±1.20     141.96±9.76                   0.9\n",

  1316.       "{'weight': 0.03162277660168379, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}     140.69±1.21     140.32±1.21     144.18±10.23                  0.91\n",

  1317.       "{'weight': 0.01, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}                    141.10±1.00     140.69±0.99     140.75±8.66                   0.94\n",

  1318.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}   141.05±1.17     140.72±1.29     141.11±9.93                   0.92\n",

  1319.       "{'weight': 0.001, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}                   141.04±0.92     140.62±0.97     141.34±8.03                   0.94\n",

  1320.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}  141.05±1.25     140.67±1.26     140.98±10.83                  0.95\n",

  1321.       "{'weight': 0.0001, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}                  141.15±1.39     140.72±1.40     140.03±11.75                  0.92\n",

  1322.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}  141.42±1.13     141.05±1.09     137.86±10.14                  0.95\n",

  1323.       "{'weight': 1e-05, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}                   141.18±1.03     140.73±1.03     140.06±9.02                   0.93\n",

  1324.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}   141.36±1.23     140.96±1.21     138.34±10.76                  0.92\n",

  1325.       "{'weight': 1e-06, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}                   141.00±1.07     140.60±1.08     141.56±9.02                   0.91\n",

  1326.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}   141.25±1.04     140.86±1.07     139.38±9.14                   0.92\n",

  1327.       "{'weight': 1e-07, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}                   141.32±1.03     140.95±1.06     138.76±9.03                   0.93\n",

  1328.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}   141.00±1.13     140.69±1.15     141.57±9.94                   0.92\n",

  1329.       "{'weight': 1e-08, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}                   141.01±1.12     140.61±1.20     141.47±9.56                   0.93\n",

  1330.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}  141.22±0.89     140.87±0.94     139.74±7.82                   0.91\n",

  1331.       "{'weight': 1e-09, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}                   141.12±0.99     140.74±0.98     140.57±8.56                   0.96\n",

  1332.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}  140.93±1.13     140.57±1.09     142.10±9.69                   0.94\n",

  1333.       "{'weight': 1e-10, 'alpha': '1.00e+05', 'compute_method': 'sylvester'}                   141.28±0.91     140.87±0.92     139.18±7.99                   0.94\n",

  1334.       "{'weight': 0.1, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}                     141.08±1.15     140.70±1.20     142.09±9.76                   0.9\n",

  1335.       "{'weight': 0.03162277660168379, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}     140.81±1.21     140.45±1.21     144.30±10.23                  0.91\n",

  1336.       "{'weight': 0.01, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}                    141.22±1.00     140.82±0.99     140.88±8.66                   0.94\n",

  1337.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}   141.17±1.17     140.84±1.29     141.24±9.93                   0.92\n",

  1338.       "{'weight': 0.001, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}                   141.16±0.92     140.74±0.97     141.46±8.03                   0.94\n",

  1339.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}  141.18±1.25     140.80±1.26     141.11±10.83                  0.95\n",

  1340.       "{'weight': 0.0001, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}                  141.28±1.39     140.84±1.40     140.15±11.74                  0.92\n",

  1341.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}  141.54±1.13     141.18±1.09     137.99±10.13                  0.95\n",

  1342.       "{'weight': 1e-05, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}                   141.30±1.03     140.85±1.03     140.18±9.02                   0.93\n",

  1343.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}   141.48±1.24     141.08±1.21     138.47±10.76                  0.92\n",

  1344.       "{'weight': 1e-06, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}                   141.13±1.07     140.72±1.08     141.69±9.03                   0.91\n",

  1345.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}   141.38±1.04     140.99±1.07     139.51±9.14                   0.92\n",

  1346.       "{'weight': 1e-07, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}                   141.45±1.03     141.07±1.07     138.89±9.03                   0.93\n",

  1347.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}   141.12±1.13     140.82±1.15     141.69±9.94                   0.92\n",

  1348.       "{'weight': 1e-08, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}                   141.13±1.13     140.73±1.20     141.60±9.56                   0.93\n",

  1349.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}  141.35±0.90     141.00±0.94     139.86±7.82                   0.91\n",

  1350.       "{'weight': 1e-09, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}                   141.25±0.99     140.87±0.98     140.70±8.56                   0.96\n",

  1351.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}  141.06±1.13     140.69±1.09     142.22±9.69                   0.94\n",

  1352.       "{'weight': 1e-10, 'alpha': '3.16e+05', 'compute_method': 'sylvester'}                   141.41±0.91     141.00±0.92     139.30±7.99                   0.94\n",

  1353.       "{'weight': 0.1, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}                     141.12±1.15     140.74±1.20     142.13±9.76                   0.9\n",

  1354.       "{'weight': 0.03162277660168379, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}     140.85±1.21     140.49±1.21     144.34±10.23                  0.91\n",

  1355.       "{'weight': 0.01, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}                    141.26±1.00     140.86±0.99     140.92±8.66                   0.94\n",

  1356.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}   141.21±1.17     140.88±1.29     141.28±9.93                   0.92\n",

  1357.       "{'weight': 0.001, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}                   141.20±0.93     140.78±0.97     141.50±8.03                   0.94\n",

  1358.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}  141.22±1.25     140.84±1.26     141.15±10.83                  0.95\n",

  1359.       "{'weight': 0.0001, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}                  141.32±1.39     140.88±1.40     140.19±11.74                  0.92\n",

  1360.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}  141.58±1.13     141.22±1.09     138.03±10.13                  0.95\n",

  1361.       "{'weight': 1e-05, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}                   141.34±1.03     140.89±1.03     140.22±9.02                   0.93\n",

  1362.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}   141.52±1.24     141.12±1.22     138.51±10.76                  0.92\n",

  1363.       "{'weight': 1e-06, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}                   141.17±1.07     140.76±1.08     141.73±9.03                   0.91\n",

  1364.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}   141.42±1.04     141.02±1.07     139.55±9.14                   0.92\n",

  1365.       "{'weight': 1e-07, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}                   141.49±1.03     141.11±1.07     138.93±9.03                   0.93\n",

  1366.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}   141.16±1.13     140.86±1.15     141.73±9.94                   0.92\n",

  1367.       "{'weight': 1e-08, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}                   141.17±1.13     140.77±1.20     141.64±9.56                   0.93\n",

  1368.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}  141.39±0.90     141.04±0.94     139.90±7.82                   0.91\n",

  1369.       "{'weight': 1e-09, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}                   141.29±0.99     140.91±0.98     140.74±8.56                   0.96\n",

  1370.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}  141.10±1.13     140.73±1.09     142.26±9.69                   0.94\n",

  1371.       "{'weight': 1e-10, 'alpha': '1.00e+06', 'compute_method': 'sylvester'}                   141.45±0.91     141.04±0.93     139.34±7.99                   0.94\n",

  1372.       "{'weight': 0.1, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}                     141.13±1.15     140.75±1.20     142.14±9.76                   0.9\n",

  1373.       "{'weight': 0.03162277660168379, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}     140.86±1.21     140.50±1.21     144.35±10.23                  0.91\n",

  1374.       "{'weight': 0.01, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}                    141.28±1.00     140.87±0.99     140.93±8.66                   0.94\n",

  1375.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}   141.23±1.17     140.90±1.29     141.29±9.93                   0.92\n",

  1376.       "{'weight': 0.001, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}                   141.21±0.93     140.80±0.97     141.52±8.03                   0.94\n",

  1377.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}  141.23±1.25     140.85±1.26     141.16±10.83                  0.95\n",

  1378.       "{'weight': 0.0001, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}                  141.33±1.39     140.90±1.40     140.20±11.74                  0.92\n",

  1379.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}  141.60±1.13     141.23±1.09     138.04±10.13                  0.95\n",

  1380.       "{'weight': 1e-05, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}                   141.35±1.03     140.91±1.03     140.24±9.02                   0.93\n",

  1381.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}   141.54±1.24     141.14±1.22     138.52±10.76                  0.92\n",

  1382.       "{'weight': 1e-06, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}                   141.18±1.07     140.78±1.08     141.74±9.03                   0.91\n",

  1383.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}   141.43±1.04     141.04±1.08     139.56±9.14                   0.92\n",

  1384.       "{'weight': 1e-07, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}                   141.50±1.03     141.13±1.07     138.94±9.03                   0.93\n",

  1385.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}   141.17±1.13     140.87±1.15     141.75±9.94                   0.92\n",

  1386.       "{'weight': 1e-08, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}                   141.19±1.13     140.79±1.20     141.65±9.56                   0.93\n",

  1387.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}  141.40±0.90     141.05±0.94     139.91±7.82                   0.91\n",

  1388.       "{'weight': 1e-09, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}                   141.30±0.99     140.92±0.98     140.75±8.56                   0.96\n",

  1389.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}  141.11±1.13     140.75±1.09     142.28±9.69                   0.94\n",

  1390.       "{'weight': 1e-10, 'alpha': '3.16e+06', 'compute_method': 'sylvester'}                   141.46±0.91     141.05±0.93     139.35±7.99                   0.94\n",

  1391.       "{'weight': 0.1, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}                     141.13±1.15     140.75±1.20     142.14±9.76                   0.9\n",

  1392.       "{'weight': 0.03162277660168379, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}     140.87±1.21     140.50±1.21     144.36±10.23                  0.91\n",

  1393.       "{'weight': 0.01, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}                    141.28±1.00     140.88±0.99     140.94±8.66                   0.94\n",

  1394.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}   141.23±1.17     140.90±1.29     141.30±9.93                   0.92\n",

  1395.       "{'weight': 0.001, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}                   141.22±0.93     140.80±0.97     141.52±8.03                   0.94\n",

  1396.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}  141.24±1.25     140.85±1.26     141.16±10.83                  0.95\n",

  1397.       "{'weight': 0.0001, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}                  141.34±1.39     140.90±1.40     140.21±11.74                  0.92\n",

  1398.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}  141.60±1.13     141.24±1.09     138.05±10.13                  0.95\n",

  1399.       "{'weight': 1e-05, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}                   141.36±1.03     140.91±1.03     140.24±9.02                   0.93\n",

  1400.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}   141.54±1.24     141.14±1.22     138.53±10.76                  0.92\n",

  1401.       "{'weight': 1e-06, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}                   141.18±1.07     140.78±1.08     141.74±9.03                   0.91\n",

  1402.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}   141.44±1.04     141.04±1.08     139.56±9.14                   0.92\n",

  1403.       "{'weight': 1e-07, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}                   141.51±1.03     141.13±1.07     138.94±9.03                   0.93\n",

  1404.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}   141.18±1.13     140.87±1.15     141.75±9.94                   0.92\n",

  1405.       "{'weight': 1e-08, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}                   141.19±1.13     140.79±1.20     141.65±9.56                   0.93\n",

  1406.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}  141.41±0.90     141.06±0.94     139.92±7.82                   0.91\n",

  1407.       "{'weight': 1e-09, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}                   141.30±0.99     140.93±0.98     140.76±8.56                   0.96\n",

  1408.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}  141.12±1.13     140.75±1.09     142.28±9.69                   0.94\n",

  1409.       "{'weight': 1e-10, 'alpha': '1.00e+07', 'compute_method': 'sylvester'}                   141.47±0.91     141.05±0.93     139.36±7.99                   0.94\n",

  1410.       "{'weight': 0.1, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}                     141.13±1.15     140.75±1.20     142.14±9.76                   0.9\n",

  1411.       "{'weight': 0.03162277660168379, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}     140.87±1.21     140.51±1.21     144.36±10.23                  0.91\n",

  1412.       "{'weight': 0.01, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}                    141.28±1.00     140.88±0.99     140.94±8.66                   0.94\n",

  1413.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}   141.23±1.17     140.90±1.29     141.30±9.93                   0.92\n",

  1414.       "{'weight': 0.001, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}                   141.22±0.93     140.80±0.97     141.52±8.03                   0.94\n",

  1415.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}  141.24±1.25     140.86±1.26     141.16±10.83                  0.95\n",

  1416.       "{'weight': 0.0001, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}                  141.34±1.39     140.90±1.40     140.21±11.74                  0.92\n",

  1417.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}  141.60±1.13     141.24±1.09     138.05±10.13                  0.95\n",

  1418.       "{'weight': 1e-05, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}                   141.36±1.03     140.91±1.03     140.24±9.02                   0.93\n",

  1419.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}   141.54±1.24     141.14±1.22     138.53±10.76                  0.92\n",

  1420.       "{'weight': 1e-06, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}                   141.19±1.07     140.78±1.08     141.75±9.03                   0.91\n",

  1421.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}   141.44±1.04     141.04±1.08     139.57±9.14                   0.92\n",

  1422.       "{'weight': 1e-07, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}                   141.51±1.03     141.13±1.07     138.94±9.03                   0.93\n",

  1423.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}   141.18±1.13     140.87±1.15     141.75±9.94                   0.92\n",

  1424.       "{'weight': 1e-08, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}                   141.19±1.13     140.79±1.20     141.65±9.56                   0.93\n",

  1425.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}  141.41±0.90     141.06±0.94     139.92±7.82                   0.91\n",

  1426.       "{'weight': 1e-09, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}                   141.30±0.99     140.93±0.98     140.76±8.56                   0.96\n",

  1427.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}  141.12±1.13     140.75±1.09     142.28±9.69                   0.94\n",

  1428.       "{'weight': 1e-10, 'alpha': '3.16e+07', 'compute_method': 'sylvester'}                   141.47±0.91     141.05±0.93     139.36±7.99                   0.94\n",

  1429.       "{'weight': 0.1, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}                     141.13±1.15     140.76±1.20     142.15±9.76                   0.9\n",

  1430.       "{'weight': 0.03162277660168379, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}     140.87±1.21     140.51±1.21     144.36±10.23                  0.91\n",

  1431.       "{'weight': 0.01, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}                    141.28±1.00     140.88±0.99     140.94±8.66                   0.94\n",

  1432.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}   141.23±1.17     140.90±1.29     141.30±9.93                   0.92\n",

  1433.       "{'weight': 0.001, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}                   141.22±0.93     140.80±0.97     141.52±8.03                   0.94\n",

  1434.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}  141.24±1.25     140.86±1.26     141.16±10.83                  0.95\n",

  1435.       "{'weight': 0.0001, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}                  141.34±1.39     140.90±1.40     140.21±11.74                  0.92\n",

  1436.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}  141.60±1.13     141.24±1.09     138.05±10.13                  0.95\n",

  1437.       "{'weight': 1e-05, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}                   141.36±1.03     140.91±1.03     140.24±9.02                   0.93\n",

  1438.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}   141.54±1.24     141.14±1.22     138.53±10.76                  0.92\n",

  1439.       "{'weight': 1e-06, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}                   141.19±1.07     140.78±1.08     141.75±9.03                   0.91\n",

  1440.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}   141.44±1.04     141.04±1.08     139.57±9.14                   0.92\n",

  1441.       "{'weight': 1e-07, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}                   141.51±1.03     141.13±1.07     138.94±9.03                   0.93\n",

  1442.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}   141.18±1.13     140.87±1.15     141.75±9.94                   0.92\n",

  1443.       "{'weight': 1e-08, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}                   141.19±1.13     140.79±1.20     141.66±9.56                   0.93\n",

  1444.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}  141.41±0.90     141.06±0.94     139.92±7.82                   0.91\n",

  1445.       "{'weight': 1e-09, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}                   141.30±0.99     140.93±0.98     140.76±8.56                   0.96\n",

  1446.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}  141.12±1.13     140.75±1.09     142.28±9.69                   0.94\n",

  1447.       "{'weight': 1e-10, 'alpha': '1.00e+08', 'compute_method': 'sylvester'}                   141.47±0.91     141.05±0.93     139.36±7.99                   0.94\n",

  1448.       "{'weight': 0.1, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}                     141.13±1.15     140.76±1.20     142.15±9.76                   0.9\n",

  1449.       "{'weight': 0.03162277660168379, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}     140.87±1.21     140.51±1.21     144.36±10.23                  0.91\n",

  1450.       "{'weight': 0.01, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}                    141.28±1.00     140.88±0.99     140.94±8.66                   0.94\n",

  1451.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}   141.23±1.17     140.90±1.29     141.30±9.93                   0.92\n",

  1452.       "{'weight': 0.001, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}                   141.22±0.93     140.80±0.97     141.52±8.03                   0.94\n",

  1453.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}  141.24±1.25     140.86±1.26     141.16±10.83                  0.95\n",

  1454.       "{'weight': 0.0001, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}                  141.34±1.39     140.90±1.40     140.21±11.74                  0.92\n",

  1455.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}  141.60±1.13     141.24±1.09     138.05±10.13                  0.95\n",

  1456.       "{'weight': 1e-05, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}                   141.36±1.03     140.91±1.03     140.24±9.02                   0.93\n",

  1457.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}   141.54±1.24     141.14±1.22     138.53±10.76                  0.92\n",

  1458.       "{'weight': 1e-06, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}                   141.19±1.07     140.78±1.08     141.75±9.03                   0.91\n",

  1459.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}   141.44±1.04     141.04±1.08     139.57±9.14                   0.92\n",

  1460.       "{'weight': 1e-07, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}                   141.51±1.03     141.13±1.07     138.94±9.03                   0.93\n",

  1461.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}   141.18±1.13     140.87±1.15     141.75±9.94                   0.92\n",

  1462.       "{'weight': 1e-08, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}                   141.19±1.13     140.79±1.20     141.66±9.56                   0.93\n",

  1463.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}  141.41±0.90     141.06±0.94     139.92±7.82                   0.91\n",

  1464.       "{'weight': 1e-09, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}                   141.30±0.99     140.93±0.98     140.76±8.56                   0.96\n",

  1465.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}  141.12±1.13     140.75±1.09     142.28±9.69                   0.94\n",

  1466.       "{'weight': 1e-10, 'alpha': '3.16e+08', 'compute_method': 'sylvester'}                   141.47±0.91     141.05±0.93     139.36±7.99                   0.94\n",

  1467.       "{'weight': 0.1, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}                     141.13±1.15     140.76±1.20     142.15±9.76                   0.9\n",

  1468.       "{'weight': 0.03162277660168379, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}     140.87±1.21     140.51±1.21     144.36±10.23                  0.91\n",

  1469.       "{'weight': 0.01, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}                    141.28±1.00     140.88±0.99     140.94±8.66                   0.94\n",

  1470.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}   141.23±1.17     140.90±1.29     141.30±9.93                   0.92\n",

  1471.       "{'weight': 0.001, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}                   141.22±0.93     140.80±0.97     141.52±8.03                   0.94\n",

  1472.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}  141.24±1.25     140.86±1.26     141.16±10.83                  0.95\n",

  1473.       "{'weight': 0.0001, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}                  141.34±1.39     140.90±1.40     140.21±11.74                  0.92\n",

  1474.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}  141.60±1.13     141.24±1.09     138.05±10.13                  0.95\n",

  1475.       "{'weight': 1e-05, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}                   141.36±1.03     140.91±1.03     140.24±9.02                   0.93\n",

  1476.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}   141.54±1.24     141.14±1.22     138.53±10.76                  0.92\n",

  1477.       "{'weight': 1e-06, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}                   141.19±1.07     140.78±1.08     141.75±9.03                   0.91\n",

  1478.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}   141.44±1.04     141.04±1.08     139.57±9.14                   0.92\n",

  1479.       "{'weight': 1e-07, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}                   141.51±1.03     141.13±1.07     138.94±9.03                   0.93\n",

  1480.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}   141.18±1.13     140.87±1.15     141.75±9.94                   0.92\n",

  1481.       "{'weight': 1e-08, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}                   141.19±1.13     140.79±1.20     141.66±9.56                   0.93\n",

  1482.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}  141.41±0.90     141.06±0.94     139.92±7.82                   0.91\n",

  1483.       "{'weight': 1e-09, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}                   141.30±0.99     140.93±0.98     140.76±8.56                   0.96\n",

  1484.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}  141.12±1.13     140.75±1.09     142.28±9.69                   0.94\n",

  1485.       "{'weight': 1e-10, 'alpha': '1.00e+09', 'compute_method': 'sylvester'}                   141.47±0.91     141.05±0.93     139.36±7.99                   0.94\n",

  1486.       "{'weight': 0.1, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}                     141.13±1.15     140.76±1.20     142.15±9.76                   0.9\n",

  1487.       "{'weight': 0.03162277660168379, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}     140.87±1.21     140.51±1.21     144.36±10.23                  0.91\n",

  1488.       "{'weight': 0.01, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}                    141.28±1.00     140.88±0.99     140.94±8.66                   0.94\n",

  1489.       "{'weight': 0.0031622776601683794, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}   141.23±1.17     140.90±1.29     141.30±9.93                   0.92\n",

  1490.       "{'weight': 0.001, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}                   141.22±0.93     140.80±0.97     141.52±8.03                   0.94\n",

  1491.       "{'weight': 0.00031622776601683794, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}  141.24±1.25     140.86±1.26     141.16±10.83                  0.95\n",

  1492.       "{'weight': 0.0001, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}                  141.34±1.39     140.90±1.40     140.21±11.74                  0.92\n",

  1493.       "{'weight': 3.1622776601683795e-05, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}  141.60±1.13     141.24±1.09     138.05±10.13                  0.95\n",

  1494.       "{'weight': 1e-05, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}                   141.36±1.03     140.91±1.03     140.24±9.02                   0.93\n",

  1495.       "{'weight': 3.162277660168379e-06, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}   141.54±1.24     141.14±1.22     138.53±10.76                  0.92\n",

  1496.       "{'weight': 1e-06, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}                   141.19±1.07     140.78±1.08     141.75±9.03                   0.91\n",

  1497.       "{'weight': 3.162277660168379e-07, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}   141.44±1.04     141.04±1.08     139.57±9.14                   0.92\n",

  1498.       "{'weight': 1e-07, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}                   141.51±1.03     141.13±1.07     138.94±9.03                   0.93\n",

  1499.       "{'weight': 3.162277660168379e-08, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}   141.18±1.13     140.87±1.15     141.75±9.94                   0.92\n",

  1500.       "{'weight': 1e-08, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}                   141.19±1.13     140.79±1.20     141.66±9.56                   0.93\n",

  1501.       "{'weight': 3.1622776601683795e-09, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}  141.41±0.90     141.06±0.94     139.92±7.82                   0.91\n",

  1502.       "{'weight': 1e-09, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}                   141.30±0.99     140.93±0.98     140.76±8.56                   0.96\n",

  1503.       "{'weight': 3.1622776601683795e-10, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}  141.12±1.13     140.75±1.09     142.28±9.69                   0.94\n",

  1504.       "{'weight': 1e-10, 'alpha': '3.16e+09', 'compute_method': 'sylvester'}                   141.47±0.91     141.05±0.93     139.36±7.99                   0.94\n",

  1505.       "{'weight': 0.1, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}                     141.13±1.15     140.76±1.20     142.15±9.76                   0.9\n",

  1506.       "{'weight': 0.03162277660168379, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}     140.87±1.21     140.51±1.21     144.36±10.23                  0.91\n",

  1507.       "{'weight': 0.01, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}                    141.28±1.00     140.88±0.99     140.94±8.66                   0.94\n",

  1508.       "{'weight': 0.0031622776601683794, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}   141.23±1.17     140.90±1.29     141.30±9.93                   0.92\n",

  1509.       "{'weight': 0.001, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}                   141.22±0.93     140.80±0.97     141.52±8.03                   0.94\n",

  1510.       "{'weight': 0.00031622776601683794, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}  141.24±1.25     140.86±1.26     141.16±10.83                  0.95\n",

  1511.       "{'weight': 0.0001, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}                  141.34±1.39     140.90±1.40     140.21±11.74                  0.92\n",

  1512.       "{'weight': 3.1622776601683795e-05, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}  141.60±1.13     141.24±1.09     138.05±10.13                  0.95\n",

  1513.       "{'weight': 1e-05, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}                   141.36±1.03     140.91±1.03     140.24±9.02                   0.93\n",

  1514.       "{'weight': 3.162277660168379e-06, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}   141.54±1.24     141.14±1.22     138.53±10.76                  0.92\n",

  1515.       "{'weight': 1e-06, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}                   141.19±1.07     140.78±1.08     141.75±9.03                   0.91\n",

  1516.       "{'weight': 3.162277660168379e-07, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}   141.44±1.04     141.04±1.08     139.57±9.14                   0.92\n",

  1517.       "{'weight': 1e-07, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}                   141.51±1.03     141.13±1.07     138.94±9.03                   0.93\n",

  1518.       "{'weight': 3.162277660168379e-08, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}   141.18±1.13     140.87±1.15     141.75±9.94                   0.92\n",

  1519.       "{'weight': 1e-08, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}                   141.19±1.13     140.79±1.20     141.66±9.56                   0.93\n",

  1520.       "{'weight': 3.1622776601683795e-09, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}  141.41±0.90     141.06±0.94     139.92±7.82                   0.91\n",

  1521.       "{'weight': 1e-09, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}                   141.30±0.99     140.93±0.98     140.76±8.56                   0.96\n",

  1522.       "{'weight': 3.1622776601683795e-10, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}  141.12±1.13     140.75±1.09     142.28±9.69                   0.94\n",

  1523.       "{'weight': 1e-10, 'alpha': '1.00e+10', 'compute_method': 'sylvester'}                   141.47±0.91     141.05±0.93     139.36±7.99                   0.94\n"

  1524.      ]

  1525.     },

  1526.     {

  1527.      "name": "stdout",

  1528.      "output_type": "stream",

  1529.      "text": [

  1530.       "calculate performance: 100%|██████████| 23370/23370 [07:37<00:00, 51.12it/s]\n",

  1531.       "\n",

  1532.       "\n",

  1533.       "COIL-DEL\n",

  1534.       "\n",

  1535.       "--- This is a classification problem ---\n",

  1536.       "\n",

  1537.       "\n",

  1538.       "I. Loading dataset from file...\n",

  1539.       "\n",

  1540.       "2. Calculating gram matrices. This could take a while...\n",

  1541.       "\n",

  1542.       " None edge weight specified. Set all weight to 1.\n",

  1543.       "\n",

  1544.       "compute adjacency matrices: 100%|██████████| 3900/3900 [00:01<00:00, 3774.23it/s]\n",

  1545.       "calculating kernels:   1%|          | 46683/7606950.0 [00:23<22:10, 5680.25it/s]  "

  1546.      ]

  1547.     },

  1548.     {

  1549.      "ename": "KeyboardInterrupt",

  1550.      "evalue": "",

  1551.      "output_type": "error",

  1552.      "traceback": [

  1553.       "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",

  1554.       "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",

  1555.       "\u001b[0;32m<ipython-input-1-5e720a487257>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     60\u001b[0m         \u001b[0mdatafile_y\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mds\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'dataset_y'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;34m'dataset_y'\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mds\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     61\u001b[0m         \u001b[0mextra_params\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mds\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'extra_params'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;34m'extra_params'\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mds\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 62\u001b[0;31m         ds_name=ds['name'])\n\u001b[0m\u001b[1;32m     63\u001b[0m     \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",

  1556.       "\u001b[0;32m/media/ljia/DATA/research-repo/codes/Linlin/py-graph/pygraph/utils/model_selection_precomputed.py\u001b[0m in \u001b[0;36mmodel_selection_for_precomputed_kernel\u001b[0;34m(datafile, estimator, param_grid_precomputed, param_grid, model_type, NUM_TRIALS, datafile_y, extra_params, ds_name)\u001b[0m\n\u001b[1;32m    113\u001b[0m         \u001b[0mnb_gm_ignore\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m  \u001b[0;31m# the number of gram matrices those should not be considered, as they may contain elements that are not numbers (NaN)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    114\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0midx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mparams_out\u001b[0m \u001b[0;32min\u001b[0m \u001b[0menumerate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mparam_list_precomputed\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 115\u001b[0;31m             \u001b[0mrtn_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mestimator\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mparams_out\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    116\u001b[0m             \u001b[0mKmatrix\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrtn_data\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    117\u001b[0m             \u001b[0mcurrent_run_time\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrtn_data\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",

  1557.       "\u001b[0;32m/media/ljia/DATA/research-repo/codes/Linlin/py-graph/pygraph/kernels/randomWalkKernel.py\u001b[0m in \u001b[0;36mrandomwalkkernel\u001b[0;34m(node_label, edge_label, edge_weight, h, p, q, weight, compute_method, *args)\u001b[0m\n\u001b[1;32m    100\u001b[0m         )\n\u001b[1;32m    101\u001b[0m         Kmatrix = _randomwalkkernel_sylvester(Gn, weight, p, q, node_label,\n\u001b[0;32m--> 102\u001b[0;31m                                               edge_label, eweight)\n\u001b[0m\u001b[1;32m    103\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    104\u001b[0m     \u001b[0;32melif\u001b[0m \u001b[0mcompute_method\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m'conjugate'\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",

  1558.       "\u001b[0;32m/media/ljia/DATA/research-repo/codes/Linlin/py-graph/pygraph/kernels/randomWalkKernel.py\u001b[0m in \u001b[0;36m_randomwalkkernel_sylvester\u001b[0;34m(Gn, lmda, p, q, node_label, edge_label, eweight)\u001b[0m\n\u001b[1;32m    196\u001b[0m                     \u001b[0mC\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfull\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mA_list\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mj\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mA_list\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpd_uni\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    197\u001b[0m                     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 198\u001b[0;31m                         \u001b[0mX\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdlyap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mA\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mQ\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mC\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    199\u001b[0m                         \u001b[0mX\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreshape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0morder\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'F'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    200\u001b[0m                         \u001b[0;31m# use uniform distribution if there is no prior knowledge.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",

  1559.       "\u001b[0;32m/usr/local/lib/python3.5/dist-packages/control/mateqn.py\u001b[0m in \u001b[0;36mdlyap\u001b[0;34m(A, Q, C, E)\u001b[0m\n\u001b[1;32m    338\u001b[0m         \u001b[0;31m# Solve the Sylvester equation by calling Slycot function sb04qd\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    339\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 340\u001b[0;31m             \u001b[0mX\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msb04qd\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mm\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0mA\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0masarray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mQ\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mT\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mC\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    341\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mValueError\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mve\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    342\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mve\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minfo\u001b[0m \u001b[0;34m<\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",

  1560.       "\u001b[0;32m/usr/local/lib/python3.5/dist-packages/slycot/synthesis.py\u001b[0m in \u001b[0;36msb04qd\u001b[0;34m(n, m, A, B, C, ldwork)\u001b[0m\n\u001b[1;32m   1234\u001b[0m         'ldwork', 'INFO'+hidden]\n\u001b[1;32m   1235\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mldwork\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1236\u001b[0;31m         \u001b[0mout\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_wrapper\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msb04qd\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mm\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mA\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mB\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mC\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1237\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1238\u001b[0m         \u001b[0mout\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_wrapper\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msb04qd\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mm\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mA\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mB\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mC\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mldwork\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mldwork\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",

  1561.       "\u001b[0;31mKeyboardInterrupt\u001b[0m: "

  1562.      ]

  1563.     }

  1564.    ],

  1565.    "source": [

  1566.     "%load_ext line_profiler\n",

  1567.     "%matplotlib inline\n",

  1568.     "import numpy as np\n",

  1569.     "import sys\n",

  1570.     "sys.path.insert(0, \"../\")\n",

  1571.     "from pygraph.utils.model_selection_precomputed import model_selection_for_precomputed_kernel\n",

  1572.     "from pygraph.kernels.randomWalkKernel import randomwalkkernel\n",

  1573.     "\n",

  1574.     "dslist = [   \n",

  1575.     "    {'name': 'Acyclic', 'dataset': '../datasets/acyclic/dataset_bps.ds', 'task': 'regression'}, # node_labeled\n",

  1576.     "    {'name': 'COIL-DEL', 'dataset': '../datasets/COIL-DEL/COIL-DEL_A.txt'}, # edge_labeled\n",

  1577.     "    {'name': 'PAH', 'dataset': '../datasets/PAH/dataset.ds',}, # unlabeled\n",

  1578.     "    {'name': 'Mutagenicity', 'dataset': '../datasets/Mutagenicity/Mutagenicity_A.txt'}, # fully_labeled\n",

  1579.     "#     {'name': 'MAO', 'dataset': '../datasets/MAO/dataset.ds',},\n",

  1580.     "\n",

  1581.     "#     {'name': 'MUTAG', 'dataset': '../datasets/MUTAG/MUTAG.mat',\n",

  1582.     "#         'extra_params': {'am_sp_al_nl_el': [0, 0, 3, 1, 2]}},\n",

  1583.     "#     {'name': 'Alkane', 'dataset': '../datasets/Alkane/dataset.ds', 'task': 'regression', \n",

  1584.     "#         'dataset_y': '../datasets/Alkane/dataset_boiling_point_names.txt',},\n",

  1585.     "#     {'name': 'BZR', 'dataset': '../datasets/BZR_txt/BZR_A_sparse.txt'},\n",

  1586.     "#     {'name': 'COX2', 'dataset': '../datasets/COX2_txt/COX2_A_sparse.txt'},    \n",

  1587.     "    {'name': 'ENZYMES', 'dataset': '../datasets/ENZYMES_txt/ENZYMES_A_sparse.txt'},\n",

  1588.     "#     {'name': 'DHFR', 'dataset': '../datasets/DHFR_txt/DHFR_A_sparse.txt'},\n",

  1589.     "#     {'name': 'SYNTHETIC', 'dataset': '../datasets/SYNTHETIC_txt/SYNTHETIC_A_sparse.txt'},\n",

  1590.     "#     {'name': 'MSRC9', 'dataset': '../datasets/MSRC_9_txt/MSRC_9_A.txt'},\n",

  1591.     "#     {'name': 'MSRC21', 'dataset': '../datasets/MSRC_21_txt/MSRC_21_A.txt'},\n",

  1592.     "#     {'name': 'FIRSTMM_DB', 'dataset': '../datasets/FIRSTMM_DB/FIRSTMM_DB_A.txt'},\n",

  1593.     "\n",

  1594.     "#     {'name': 'PROTEINS', 'dataset': '../datasets/PROTEINS_txt/PROTEINS_A_sparse.txt'},\n",

  1595.     "#     {'name': 'PROTEINS_full', 'dataset': '../datasets/PROTEINS_full_txt/PROTEINS_full_A_sparse.txt'},\n",

  1596.     "#     {'name': 'D&D', 'dataset': '../datasets/D&D/DD.mat',\n",

  1597.     "#      'extra_params': {'am_sp_al_nl_el': [0, 1, 2, 1, -1]}},\n",

  1598.     "#     {'name': 'AIDS', 'dataset': '../datasets/AIDS/AIDS_A.txt'},\n",

  1599.     "#     {'name': 'NCI1', 'dataset': '../datasets/NCI1/NCI1.mat',\n",

  1600.     "#         'extra_params': {'am_sp_al_nl_el': [1, 1, 2, 0, -1]}},\n",

  1601.     "#     {'name': 'NCI109', 'dataset': '../datasets/NCI109/NCI109.mat',\n",

  1602.     "#         'extra_params': {'am_sp_al_nl_el': [1, 1, 2, 0, -1]}},\n",

  1603.     "#     {'name': 'NCI-HIV', 'dataset': '../datasets/NCI-HIV/AIDO99SD.sdf',\n",

  1604.     "#         'dataset_y': '../datasets/NCI-HIV/aids_conc_may04.txt',},\n",

  1605.     "    \n",

  1606.     "#     # not working below\n",

  1607.     "#     {'name': 'PTC_FM', 'dataset': '../datasets/PTC/Train/FM.ds',},\n",

  1608.     "#     {'name': 'PTC_FR', 'dataset': '../datasets/PTC/Train/FR.ds',},\n",

  1609.     "#     {'name': 'PTC_MM', 'dataset': '../datasets/PTC/Train/MM.ds',},\n",

  1610.     "#     {'name': 'PTC_MR', 'dataset': '../datasets/PTC/Train/MR.ds',},\n",

  1611.     "]\n",

  1612.     "estimator = randomwalkkernel\n",

  1613.     "param_grid_precomputed = {'compute_method': ['sylvester'], \n",

  1614.     "                          'weight': np.logspace(0, -10, num = 21, base = 10)}\n",

  1615.     "param_grid = [{'C': np.logspace(-10, 10, num = 41, base = 10)}, \n",

  1616.     "              {'alpha': np.logspace(-10, 10, num = 41, base = 10)}]\n",

  1617.     "\n",

  1618.     "for ds in dslist:\n",

  1619.     "    print()\n",

  1620.     "    print(ds['name'])\n",

  1621.     "    model_selection_for_precomputed_kernel(\n",

  1622.     "        ds['dataset'], estimator, param_grid_precomputed, \n",

  1623.     "        (param_grid[1] if ('task' in ds and ds['task'] == 'regression') else param_grid[0]), \n",

  1624.     "        (ds['task'] if 'task' in ds else 'classification'), NUM_TRIALS=30,\n",

  1625.     "        datafile_y=(ds['dataset_y'] if 'dataset_y' in ds else None),\n",

  1626.     "        extra_params=(ds['extra_params'] if 'extra_params' in ds else None),\n",

  1627.     "        ds_name=ds['name'])\n",

  1628.     "    print()"

  1629.    ]

  1630.   },

  1631.   {

  1632.    "cell_type": "code",

  1633.    "execution_count": 1,

  1634.    "metadata": {

  1635.     "scrolled": true

  1636.    },

  1637.    "outputs": [

  1638.     {

  1639.      "name": "stdout",

  1640.      "output_type": "stream",

  1641.      "text": [

  1642.       "\n",

  1643.       "--- This is a regression problem ---\n",

  1644.       "\n",

  1645.       "1. Loading dataset from file...\n",

  1646.       "\n",

  1647.       "2. Calculating gram matrices. This could take a while...\n",

  1648.       "\n",

  1649.       "gram matrix with parameters {'n': 0.0} is: \n"

  1650.      ]

  1651.     },

  1652.     {

  1653.      "ename": "IndexError",

  1654.      "evalue": "index 1 is out of bounds for axis 0 with size 1",

  1655.      "output_type": "error",

  1656.      "traceback": [

  1657.       "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",

  1658.       "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",

  1659.       "\u001b[0;32m<ipython-input-1-2b1121e86472>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     13\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     14\u001b[0m model_selection_for_precomputed_kernel(datafile, estimator, param_grid_precomputed, param_grid, \n\u001b[0;32m---> 15\u001b[0;31m                                        'regression', NUM_TRIALS=30)\n\u001b[0m",

  1660.       "\u001b[0;32m/media/ljia/DATA/research-repo/codes/Linlin/py-graph/pygraph/utils/model_selection_precomputed.py\u001b[0m in \u001b[0;36mmodel_selection_for_precomputed_kernel\u001b[0;34m(datafile, estimator, param_grid_precomputed, param_grid, model_type, NUM_TRIALS, datafile_y)\u001b[0m\n\u001b[1;32m     94\u001b[0m         \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'gram matrix with parameters'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mparams_out\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'is: '\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     95\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 96\u001b[0;31m         \u001b[0mKmatrix\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcurrent_run_time\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mestimator\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mparams_out\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     97\u001b[0m         \u001b[0mKmatrix_diag\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mKmatrix\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdiagonal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcopy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     98\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",

  1661.       "\u001b[0;32m/media/ljia/DATA/research-repo/codes/Linlin/py-graph/pygraph/kernels/untilnWalkKernel.py\u001b[0m in \u001b[0;36muntilnwalkkernel\u001b[0;34m(node_label, edge_label, labeled, n, weight, compute_method, *args)\u001b[0m\n\u001b[1;32m     65\u001b[0m             \u001b[0;32mfor\u001b[0m \u001b[0mj\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mGn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     66\u001b[0m                 Kmatrix[i][j] = _untilnwalkkernel_direct(\n\u001b[0;32m---> 67\u001b[0;31m                     Gn[i], Gn[j], node_label, edge_label, labeled, weight)\n\u001b[0m\u001b[1;32m     68\u001b[0m                 \u001b[0mKmatrix\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mj\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mKmatrix\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mj\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     69\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",

  1662.       "\u001b[0;32m/media/ljia/DATA/research-repo/codes/Linlin/py-graph/pygraph/kernels/untilnWalkKernel.py\u001b[0m in \u001b[0;36m_untilnwalkkernel_direct\u001b[0;34m(G1, G2, node_label, edge_label, labeled, weight)\u001b[0m\n\u001b[1;32m    129\u001b[0m     \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mj\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mproduct\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgp\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgp\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    130\u001b[0m         \u001b[0mmat_tmp\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mT\u001b[0m \u001b[0;34m*\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexp\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mweight\u001b[0m \u001b[0;34m*\u001b[0m \u001b[0mD\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m*\u001b[0m \u001b[0mT\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mI\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 131\u001b[0;31m         \u001b[0mkernel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mkernel\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mmat_tmp\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mj\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    132\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    133\u001b[0m     \u001b[0;31m# from matplotlib import pyplot as plt\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",

  1663.       "\u001b[0;32m/usr/local/lib/python3.5/dist-packages/numpy/matrixlib/defmatrix.py\u001b[0m in \u001b[0;36m__getitem__\u001b[0;34m(self, index)\u001b[0m\n\u001b[1;32m    282\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    283\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 284\u001b[0;31m             \u001b[0mout\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mN\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mndarray\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__getitem__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mindex\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    285\u001b[0m         \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    286\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_getitem\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",

  1664.       "\u001b[0;31mIndexError\u001b[0m: index 1 is out of bounds for axis 0 with size 1"

  1665.      ]

  1666.     }

  1667.    ],

  1668.    "source": [

  1669.     "%load_ext line_profiler\n",

  1670.     "%matplotlib inline\n",

  1671.     "import numpy as np\n",

  1672.     "import sys\n",

  1673.     "sys.path.insert(0, \"../\")\n",

  1674.     "from pygraph.utils.model_selection_precomputed import model_selection_for_precomputed_kernel\n",

  1675.     "from pygraph.kernels.untilnWalkKernel import untilnwalkkernel\n",

  1676.     "\n",

  1677.     "datafile = '../../../../datasets/acyclic/Acyclic/dataset_bps.ds'\n",

  1678.     "estimator = untilnwalkkernel\n",

  1679.     "param_grid_precomputed = {'n': np.linspace(0, 10, 11)}\n",

  1680.     "param_grid = {'alpha': np.logspace(-10, 10, num = 41, base = 10)}\n",

  1681.     "\n",

  1682.     "model_selection_for_precomputed_kernel(datafile, estimator, param_grid_precomputed, param_grid, \n",

  1683.     "                                       'regression', NUM_TRIALS=30)"

  1684.    ]

  1685.   },

  1686.   {

  1687.    "cell_type": "code",

  1688.    "execution_count": null,

  1689.    "metadata": {},

  1690.    "outputs": [],

  1691.    "source": [

  1692.     "# results\n",

  1693.     "\n",

  1694.     "# untiln kernel when h = 2\n",

  1695.     "       lmda    rmse_test    std_test    rmse_train    std_train    k_time\n",

  1696.     "-----------  -----------  ----------  ------------  -----------  --------\n",

  1697.     "     1e-10       7.46524     1.71862       5.99486     0.356634   38.1447\n",

  1698.     "     1e-09       7.37326     1.77195       5.96155     0.374395   37.4921\n",

  1699.     "     1e-08       7.35105     1.78349       5.96481     0.378047   37.9971\n",

  1700.     "     1e-07       7.35213     1.77903       5.96728     0.382251   38.3182\n",

  1701.     "     1e-06       7.3524      1.77992       5.9696      0.3863     39.6428\n",

  1702.     "     1e-05       7.34958     1.78141       5.97114     0.39017    37.3711\n",

  1703.     "     0.0001      7.3513      1.78136       5.94251     0.331843   37.3967\n",

  1704.     "     0.001       7.35822     1.78119       5.9326      0.32534    36.7357\n",

  1705.     "     0.01        7.37552     1.79037       5.94089     0.34763    36.8864\n",

  1706.     "     0.1         7.32951     1.91346       6.42634     1.29405    36.8382\n",

  1707.     "     1           7.27134     2.20774       6.62425     1.2242     37.2425\n",

  1708.     "    10           7.49787     2.36815       6.81697     1.50182    37.8286\n",

  1709.     "   100           7.42887     2.64789       6.68766     1.34809    36.3701\n",

  1710.     "  1000           7.24914     2.65554       6.81906     1.41008    36.1695\n",

  1711.     " 10000           7.08183     2.6248        6.93431     1.38441    37.5723\n",

  1712.     "100000           8.021       3.43694       8.69813     0.909839   37.8158\n",

  1713.     "     1e+06       8.49625     3.6332        9.59333     0.96626    38.4688\n",

  1714.     "     1e+07      10.9067      3.17593      11.5642      2.07792    36.9926\n",

  1715.     "     1e+08      61.1524     10.4355       65.3527     13.9538     37.1321\n",

  1716.     "     1e+09      99.943      13.6994       98.8848      5.27014    36.7443\n",

  1717.     "     1e+10     100.083      13.8503       97.9168      3.22768    37.096\n"

  1718.    ]

  1719.   }

  1720.  ],

  1721.  "metadata": {

  1722.   "kernelspec": {

  1723.    "display_name": "Python 3",

  1724.    "language": "python",

  1725.    "name": "python3"

  1726.   },

  1727.   "language_info": {

  1728.    "codemirror_mode": {

  1729.     "name": "ipython",

  1730.     "version": 3

  1731.    },

  1732.    "file_extension": ".py",

  1733.    "mimetype": "text/x-python",

  1734.    "name": "python",

  1735.    "nbconvert_exporter": "python",

  1736.    "pygments_lexer": "ipython3",

  1737.    "version": "3.5.2"

  1738.   }

  1739.  },

  1740.  "nbformat": 4,

  1741.  "nbformat_minor": 2

  1742. }

A Python package for graph kernels, graph edit distances and graph pre-image problem.