Loading devtools/jenkins/MolNet.sh +4 −2 Original line number Diff line number Diff line Loading @@ -6,14 +6,16 @@ source activate $envname python setup.py install rm examples/results.csv || true export DEEPCHEM_DATA_DIR='/tmp/molnet_'$envname mkdir $DEEPCHEM_DATA_DIR cd examples python benchmark.py -d tox21 -m weave -m graphconv -m tf_robust -m tf -m irv -m xgb -m logreg --seed 123 export retval_tox21=$? python benchmark.py -d hiv -m graphconv -m tf -m irv -m logreg --seed 123 export retval_hiv=$? python benchmark.py -d delaney -m weave_regression -m graphconreg -m tf_regression -m xgb_regression -m mpnn --seed 123 python benchmark.py -d delaney -m weave_regression -m graphconvreg -m tf_regression -m dag_regression -m mpnn --seed 123 export retval_delaney=$? python benchmark.py -d qm7 -m dtnn -m graphconvreg -m tf_regression_ft -m tf_regression --seed 123 python benchmark.py -d qm7 -m dtnn -m graphconvreg -m tf_regression_ft --seed 123 export retval_qm7=$? cd .. Loading devtools/jenkins/compare_results.py +2 −1 Original line number Diff line number Diff line Loading @@ -10,7 +10,7 @@ BENCHMARK_TO_DESIRED_KEY_MAP = { "logreg": "Logistic regression", "rf": "Random forest", "tf": "NN classification", "tf_robust": "robust NN", "tf_robust": "Robust NN", "tf_regression": "NN regression", "tf_regression_ft": "NN regression(CM)", "graphconv": "Graph convolution", Loading @@ -19,6 +19,7 @@ BENCHMARK_TO_DESIRED_KEY_MAP = { "dag": "DAG", "dag_regression":"DAG regression", "dtnn": "DTNN", "mpnn": "MPNN", "weave": "Weave", "weave_regression": "Weave regression", "xgb": "XGBoost", Loading devtools/jenkins/desired_results.csv +6 −0 Original line number Diff line number Diff line Loading @@ -224,18 +224,21 @@ Index splitting,delaney,NN regression,0.869,0.585 Index splitting,delaney,Graphconv regression,0.969,0.813 Index splitting,delaney,DAG regression,0.976,0.85 Index splitting,delaney,Weave regression,0.963,0.872 Index splitting,delaney,MPNN,0.988,0.882 Random splitting,delaney,Random forest,0.955,0.561 Random splitting,delaney,XGBoost regression,0.927,0.727 Random splitting,delaney,NN regression,0.875,0.495 Random splitting,delaney,Graphconv regression,0.976,0.787 Random splitting,delaney,DAG regression,0.968,0.899 Random splitting,delaney,Weave regression,0.955,0.907 Random splitting,delaney,MPNN,0.985,0.926 Scaffold splitting,delaney,Random forest,0.953,0.281 Scaffold splitting,delaney,XGBoost regression,0.89,0.316 Scaffold splitting,delaney,NN regression,0.872,0.308 Scaffold splitting,delaney,Graphconv regression,0.98,0.564 Scaffold splitting,delaney,DAG regression,0.968,0.676 Scaffold splitting,delaney,Weave regression,0.971,0.756 Scaffold splitting,delaney,MPNN,0.986,0.777 Index splitting,hopv,Random forest,0.943,0.338 Index splitting,hopv,NN regression,0.725,0.293 Index splitting,hopv,Graphconv regression,0.307,0.284 Loading Loading @@ -354,15 +357,18 @@ Index splitting,sampl,NN regression,0.923,0.758 Index splitting,sampl,Graphconv regression,0.97,0.897 Index splitting,sampl,DAG regression,0.97,0.871 Index splitting,sampl,Weave regression,0.992,0.915 Index splitting,sampl,MPNN,0.992,0.906 Random splitting,sampl,Random forest,0.966,0.729 Random splitting,sampl,XGBoost regression,0.906,0.745 Random splitting,sampl,NN regression,0.931,0.689 Random splitting,sampl,Graphconv regression,0.964,0.848 Random splitting,sampl,DAG regression,0.973,0.861 Random splitting,sampl,Weave regression,0.992,0.885 Random splitting,sampl,MPNN,0.997,0.883 Scaffold splitting,sampl,Random forest,0.967,0.465 Scaffold splitting,sampl,XGBoost regression,0.918,0.439 Scaffold splitting,sampl,NN regression,0.901,0.238 Scaffold splitting,sampl,Graphconv regression,0.963,0.822 Scaffold splitting,sampl,DAG regression,0.961,0.846 Scaffold splitting,sampl,Weave regression,0.992,0.837 Scaffold splitting,sampl,MPNN,0.993,0.862 Loading
devtools/jenkins/MolNet.sh +4 −2 Original line number Diff line number Diff line Loading @@ -6,14 +6,16 @@ source activate $envname python setup.py install rm examples/results.csv || true export DEEPCHEM_DATA_DIR='/tmp/molnet_'$envname mkdir $DEEPCHEM_DATA_DIR cd examples python benchmark.py -d tox21 -m weave -m graphconv -m tf_robust -m tf -m irv -m xgb -m logreg --seed 123 export retval_tox21=$? python benchmark.py -d hiv -m graphconv -m tf -m irv -m logreg --seed 123 export retval_hiv=$? python benchmark.py -d delaney -m weave_regression -m graphconreg -m tf_regression -m xgb_regression -m mpnn --seed 123 python benchmark.py -d delaney -m weave_regression -m graphconvreg -m tf_regression -m dag_regression -m mpnn --seed 123 export retval_delaney=$? python benchmark.py -d qm7 -m dtnn -m graphconvreg -m tf_regression_ft -m tf_regression --seed 123 python benchmark.py -d qm7 -m dtnn -m graphconvreg -m tf_regression_ft --seed 123 export retval_qm7=$? cd .. Loading
devtools/jenkins/compare_results.py +2 −1 Original line number Diff line number Diff line Loading @@ -10,7 +10,7 @@ BENCHMARK_TO_DESIRED_KEY_MAP = { "logreg": "Logistic regression", "rf": "Random forest", "tf": "NN classification", "tf_robust": "robust NN", "tf_robust": "Robust NN", "tf_regression": "NN regression", "tf_regression_ft": "NN regression(CM)", "graphconv": "Graph convolution", Loading @@ -19,6 +19,7 @@ BENCHMARK_TO_DESIRED_KEY_MAP = { "dag": "DAG", "dag_regression":"DAG regression", "dtnn": "DTNN", "mpnn": "MPNN", "weave": "Weave", "weave_regression": "Weave regression", "xgb": "XGBoost", Loading
devtools/jenkins/desired_results.csv +6 −0 Original line number Diff line number Diff line Loading @@ -224,18 +224,21 @@ Index splitting,delaney,NN regression,0.869,0.585 Index splitting,delaney,Graphconv regression,0.969,0.813 Index splitting,delaney,DAG regression,0.976,0.85 Index splitting,delaney,Weave regression,0.963,0.872 Index splitting,delaney,MPNN,0.988,0.882 Random splitting,delaney,Random forest,0.955,0.561 Random splitting,delaney,XGBoost regression,0.927,0.727 Random splitting,delaney,NN regression,0.875,0.495 Random splitting,delaney,Graphconv regression,0.976,0.787 Random splitting,delaney,DAG regression,0.968,0.899 Random splitting,delaney,Weave regression,0.955,0.907 Random splitting,delaney,MPNN,0.985,0.926 Scaffold splitting,delaney,Random forest,0.953,0.281 Scaffold splitting,delaney,XGBoost regression,0.89,0.316 Scaffold splitting,delaney,NN regression,0.872,0.308 Scaffold splitting,delaney,Graphconv regression,0.98,0.564 Scaffold splitting,delaney,DAG regression,0.968,0.676 Scaffold splitting,delaney,Weave regression,0.971,0.756 Scaffold splitting,delaney,MPNN,0.986,0.777 Index splitting,hopv,Random forest,0.943,0.338 Index splitting,hopv,NN regression,0.725,0.293 Index splitting,hopv,Graphconv regression,0.307,0.284 Loading Loading @@ -354,15 +357,18 @@ Index splitting,sampl,NN regression,0.923,0.758 Index splitting,sampl,Graphconv regression,0.97,0.897 Index splitting,sampl,DAG regression,0.97,0.871 Index splitting,sampl,Weave regression,0.992,0.915 Index splitting,sampl,MPNN,0.992,0.906 Random splitting,sampl,Random forest,0.966,0.729 Random splitting,sampl,XGBoost regression,0.906,0.745 Random splitting,sampl,NN regression,0.931,0.689 Random splitting,sampl,Graphconv regression,0.964,0.848 Random splitting,sampl,DAG regression,0.973,0.861 Random splitting,sampl,Weave regression,0.992,0.885 Random splitting,sampl,MPNN,0.997,0.883 Scaffold splitting,sampl,Random forest,0.967,0.465 Scaffold splitting,sampl,XGBoost regression,0.918,0.439 Scaffold splitting,sampl,NN regression,0.901,0.238 Scaffold splitting,sampl,Graphconv regression,0.963,0.822 Scaffold splitting,sampl,DAG regression,0.961,0.846 Scaffold splitting,sampl,Weave regression,0.992,0.837 Scaffold splitting,sampl,MPNN,0.993,0.862
