Added new contextual bandit algorithm and updated example notebook.

ONN_Usage_Example.ipynb

@@ -3,23 +3,25 @@
 #Installing Dependencies
 
 %% Cell type:code id: tags:
 
 ``` 
-!pip install onn
+!pip install --upgrade onn
 ```
 
 %% Cell type:markdown id: tags:
 
 ##Importing Dependencies
 
 %% Cell type:code id: tags:
 
 ``` 
 from onn.OnlineNeuralNetwork import ONN
-from sklearn.datasets import make_classification
+from onn.OnlineNeuralNetwork import ONN_THS
+from sklearn.datasets import make_classification, make_circles
 from sklearn.model_selection import train_test_split
+import torch
 from sklearn.metrics import accuracy_score
 import numpy as np
 ```
 
 %% Cell type:markdown id: tags:
@@ -56,5 +58,148 @@
   
   if i % 1000 == 0:
     predictions = onn_network.predict(X_test)
     print("Online Accuracy: {}".format(accuracy_score(y_test, predictions)))
 ```
+
+%% Cell type:markdown id: tags:
+
+# Learning in batch with CUDA
+
+%% Cell type:code id: tags:
+
+``` 
+onn_network = ONN(features_size=10, max_num_hidden_layers=5, qtd_neuron_per_hidden_layer=40, n_classes=10, batch_size=10, use_cuda=True)
+```
+
+%% Cell type:code id: tags:
+
+``` 
+from torch.utils.data import Dataset, DataLoader
+class Dataset(Dataset):
+
+  def __init__(self, X, Y):
+    self.X = X
+    self.Y = Y
+
+  def __len__(self):
+      return len(self.X)
+
+  def __getitem__(self, idx):
+      X = self.X[idx],
+      Y = self.Y[idx]
+
+      return X, Y
+    
+transformed_dataset = Dataset(X_train, y_train)
+dataloader = DataLoader(transformed_dataset, batch_size=10,shuffle=True, num_workers=1)
+```
+
+%% Cell type:code id: tags:
+
+``` 
+for local_X, local_y in dataloader: 
+  onn_network.partial_fit(np.squeeze(torch.stack(local_X).numpy()), local_y.numpy())
+```
+
+%% Cell type:code id: tags:
+
+``` 
+predictions = onn_network.predict(X_test)
+print("Accuracy: {}".format(accuracy_score(y_test, predictions)))
+```
+
+%% Cell type:markdown id: tags:
+
+#Using contextual bandit - ONN_THS
+
+%% Cell type:markdown id: tags:
+
+In this example the ONN acts like a contextual bandits a reinforcement learning algorithm type. 
+
+%% Cell type:code id: tags:
+
+``` 
+X_linear, Y_linear = make_classification(n_samples=10000, n_features=2, n_informative=2, n_redundant=0, n_classes=2, n_clusters_per_class=1, class_sep=200, shuffle=True)
+X_non_linear, Y_non_linear = make_circles(n_samples=10000, noise=0.1, factor=0.3, shuffle=True)
+X_linear_2, Y_linear_2 = make_classification(n_samples=10000, n_features=2, n_informative=2, n_redundant=0, n_classes=2, n_clusters_per_class=1, class_sep=200, shuffle=True)
+
+X_linear_train = X_linear[:5000]
+Y_linear_train = Y_linear[:5000]
+
+X_linear_test = X_linear[5000:]
+Y_linear_test = Y_linear[5000:]
+
+X_non_linear_train = X_non_linear[:5000]
+Y_non_linear_train = Y_non_linear[:5000]
+
+X_non_linear_test = X_non_linear[5000:]
+Y_non_linear_test = Y_non_linear[5000:]
+
+X_linear_train_2 = X_linear_2[:5000]
+Y_linear_train_2 = Y_linear_2[:5000]
+
+X_linear_test_2 = X_linear_2[5000:]
+Y_linear_test_2 = Y_linear_2[5000:]
+```
+
+%% Cell type:code id: tags:
+
+``` 
+gp = ONN_THS(2, 5, 100, 2)
+```
+
+%% Cell type:code id: tags:
+
+``` 
+for epoch in range(5):
+
+    for i in range(len(X_linear_train)):
+        x = np.asarray([X_linear_train[i, :]])
+        y = np.asarray([Y_linear_train[i]])
+
+        arm, exp = gp.predict(x)
+        
+        if arm == y[0]:  
+          gp.partial_fit(x, y, exp)
+          
+        if i % 2000 == 1999:
+          pred = []
+          print("======================================================")
+          for i in range(len(X_linear_test)):  
+            pred.append(gp.predict(np.asarray([X_linear_test[i, :]]))[0])
+          print("Accuracy: " + str(accuracy_score(Y_linear_test, pred)))
+          print("======================================================")
+
+print('Finished Training')
+```
+
+%% Cell type:code id: tags:
+
+``` 
+for epoch in range(5):
+
+    for i in range(len(X_non_linear_train)):
+        x = np.asarray([X_non_linear_train[i, :]])
+        y = np.asarray([Y_non_linear_train[i]])
+
+        arm, exp = gp.predict(x)
+        
+        if arm == y[0]:  
+          gp.partial_fit(x, y, exp)
+          
+        if i % 2000 == 1999:
+          pred = []
+          print("======================================================")
+          for i in range(len(X_linear_test)):  
+            pred.append(gp.predict(np.asarray([X_non_linear_test[i, :]]))[0])
+          print("Accuracy: " + str(accuracy_score(Y_non_linear_test, pred)))
+          print("======================================================")
+
+print('Finished Training')
+```
+
+%% Cell type:code id: tags:
+
+``` 
+
+```

onn/OnlineNeuralNetwork.py

@@ -153,7 +153,7 @@ class ONN(nn.Module):
 
 class ONN_THS(ONN):
     def __init__(self, features_size, max_num_hidden_layers, qtd_neuron_per_hidden_layer, n_classes, b=0.99, n=0.01,
-                 s=0.2, e=[0.5, 0.35, 0.2, 0.1], use_cuda=False):
+                 s=0.2, e=[0.5, 0.35, 0.2, 0.1, 0.05], use_cuda=False):
         super().__init__(features_size, max_num_hidden_layers, qtd_neuron_per_hidden_layer, n_classes, b=b, n=n, s=s,
                          use_cuda=use_cuda)
         self.e = e

setup.py

@@ -6,7 +6,7 @@ with open(path.join(this_directory, 'README.md'), encoding='utf-8') as f:
     long_description = f.read()
 
 setup(name='onn',
-      version='0.1.1',
+      version='0.1.3',
       description='Online Neural Network',
       url='https://github.com/alison-carrera/onn',
       author='Alison Carrera',