[ENH] Some polish

.gitignore

@@ -1,6 +1,7 @@
 *.pyc
 *.swp
 
+.idea
 build
 dist/
 torchensemble.egg-info

MINIFEST.in

@@ -0,0 +1,4 @@
+include MANIFEST.in
+
+# exclude from sdist
+exclude examples

examples/DATASET.txt

@@ -0,0 +1,2 @@
+* The YearPredictionMSD dataset can be downloaded from "https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/regression.html#YearPredictionMSD"
+* Please modify the `data_dir` in classification_cifar10_cnn.py, or `train_path` and `test_path` in regression_YearPredictionMSD_mlp.py for personal use

examples/classification_cifar10_cnn.py

@@ -81,7 +81,7 @@ def num_flat_features(self, X):
                                       transforms.ToTensor(),
                                       transforms.Normalize((0.4914, 0.4822, 0.4465), 
                                                            (0.2023, 0.1994, 0.2010))])
-    
+
     train_loader = DataLoader(datasets.CIFAR10(
         data_dir, train=True, download=True, transform=transformer), 
         batch_size=batch_size, shuffle=True)
@@ -94,11 +94,11 @@ def num_flat_features(self, X):
 
     # FusionClassifier
     model = FusionClassifier(estimator=LeNet5,
-                              n_estimators=n_estimators,
-                              output_dim=output_dim,
-                              lr=lr,
-                              weight_decay=weight_decay,
-                              epochs=epochs)
+                             n_estimators=n_estimators,
+                             output_dim=output_dim,
+                             lr=lr,
+                             weight_decay=weight_decay,
+                             epochs=epochs)
 
     tic = time.time()
     model.fit(train_loader)
@@ -117,12 +117,12 @@ def num_flat_features(self, X):
 
     # VotingClassifier
     model = VotingClassifier(estimator=LeNet5,
-                              n_estimators=n_estimators,
-                              output_dim=output_dim,
-                              lr=lr,
-                              weight_decay=weight_decay,
-                              epochs=epochs,
-                              n_jobs=1)
+                             n_estimators=n_estimators,
+                             output_dim=output_dim,
+                             lr=lr,
+                             weight_decay=weight_decay,
+                             epochs=epochs,
+                             n_jobs=1)
 
     tic = time.time()
     model.fit(train_loader)
@@ -169,7 +169,8 @@ def num_flat_features(self, X):
                                        output_dim=output_dim,
                                        lr=lr,
                                        weight_decay=weight_decay,
-                                       epochs=epochs)
+                                       epochs=epochs,
+                                       shrinkage_rate=1.)
 
     tic = time.time()
     model.fit(train_loader)

examples/regression_YearPredictionMSD_mlp.py

@@ -33,6 +33,7 @@ def load_data(batch_size):
     train = load_svmlight_file(train_path)
     test = load_svmlight_file(test_path)
 
+    # Numpy array -> Tensor
     X_train, X_test = (torch.FloatTensor(train[0].toarray()), 
                        torch.FloatTensor(test[0].toarray()))
 
@@ -173,7 +174,8 @@ def forward(self, X):
                                       output_dim=output_dim,
                                       lr=lr,
                                       weight_decay=weight_decay,
-                                      epochs=epochs)
+                                      epochs=epochs,
+                                      shrinkage_rate=1.)
 
     tic = time.time()
     model.fit(train_loader)

setup.py

@@ -53,7 +53,7 @@ def run(self):
 
 setup(
     name='torchensemble',
-    version='0.0.1',
+    version='1.0.0',
     author='AaronX121',
 
     description=('Implementations of scikit-learn like ensemble methods in'

torchensemble/bagging.py

@@ -54,9 +54,9 @@ def _parallel_fit(epoch, estimator_idx,
 
                 msg = ('Estimator: {:03d} | Epoch: {:03d} |' 
                        ' Batch: {:03d} | Loss: {:.5f} | Correct:'
-                       ' {:d}/{:d}')
+                       ' {:03d}/{:03d}')
                 print(msg.format(estimator_idx, epoch, batch_idx, loss, 
-                                 correct, batch_size))
+                                 correct, sampling_X_train.size()[0]))
             else:
                 msg = ('Estimator: {:03d} | Epoch: {:03d} |' 
                        ' Batch: {:03d} | Loss: {:.5f}')