diff --git a/2_logistic_regression.py b/2_logistic_regression.py
index a3df9d8..9219c47 100644
--- a/2_logistic_regression.py
+++ b/2_logistic_regression.py
@@ -14,9 +14,9 @@ def build_model(input_dim, output_dim):
     return model
 
 
-def train(model, loss, optimizer, x, y):
-    x = Variable(x, requires_grad=False)
-    y = Variable(y, requires_grad=False)
+def train(model, loss, optimizer, x_val, y_val):
+    x = Variable(x_val, requires_grad=False)
+    y = Variable(y_val, requires_grad=False)
 
     # Reset gradient
     optimizer.zero_grad()
@@ -34,9 +34,9 @@ def train(model, loss, optimizer, x, y):
     return output.data[0]
 
 
-def predict(model, x):
-    var_x = Variable(x, requires_grad=False)
-    output = model.forward(var_x)
+def predict(model, x_val):
+    x = Variable(x_val, requires_grad=False)
+    output = model.forward(x)
     return output.data.numpy().argmax(axis=1)
 
 
diff --git a/3_neural_net.py b/3_neural_net.py
index decf299..13ce533 100644
--- a/3_neural_net.py
+++ b/3_neural_net.py
@@ -16,9 +16,9 @@ def build_model(input_dim, output_dim):
     return model
 
 
-def train(model, loss, optimizer, x, y):
-    x = Variable(x, requires_grad=False)
-    y = Variable(y, requires_grad=False)
+def train(model, loss, optimizer, x_val, y_val):
+    x = Variable(x_val, requires_grad=False)
+    y = Variable(y_val, requires_grad=False)
 
     # Reset gradient
     optimizer.zero_grad()
@@ -36,9 +36,9 @@ def train(model, loss, optimizer, x, y):
     return output.data[0]
 
 
-def predict(model, x):
-    var_x = Variable(x, requires_grad=False)
-    output = model.forward(var_x)
+def predict(model, x_val):
+    x = Variable(x_val, requires_grad=False)
+    output = model.forward(x)
     return output.data.numpy().argmax(axis=1)
 
 
diff --git a/4_modern_neural_net.py b/4_modern_neural_net.py
index 91a042b..3c57aee 100644
--- a/4_modern_neural_net.py
+++ b/4_modern_neural_net.py
@@ -20,9 +20,9 @@ def build_model(input_dim, output_dim):
     return model
 
 
-def train(model, loss, optimizer, x, y):
-    x = Variable(x, requires_grad=False)
-    y = Variable(y, requires_grad=False)
+def train(model, loss, optimizer, x_val, y_val):
+    x = Variable(x_val, requires_grad=False)
+    y = Variable(y_val, requires_grad=False)
 
     # Reset gradient
     optimizer.zero_grad()
@@ -40,9 +40,9 @@ def train(model, loss, optimizer, x, y):
     return output.data[0]
 
 
-def predict(model, x):
-    var_x = Variable(x, requires_grad=False)
-    output = model.forward(var_x)
+def predict(model, x_val):
+    x = Variable(x_val, requires_grad=False)
+    output = model.forward(x)
     return output.data.numpy().argmax(axis=1)
 
 
@@ -57,7 +57,7 @@ def main():
     n_classes = 10
     model = build_model(n_features, n_classes)
     loss = torch.nn.CrossEntropyLoss(size_average=True)
-    optimizer = optim.RMSprop(model.parameters(), lr=0.01)
+    optimizer = optim.Adam(model.parameters())
     batch_size = 100
 
     for i in range(100):
diff --git a/5_convolutional_net.py b/5_convolutional_net.py
index 9bfc565..bfdb62a 100644
--- a/5_convolutional_net.py
+++ b/5_convolutional_net.py
@@ -35,9 +35,9 @@ def forward(self, x):
         return self.fc.forward(x)
 
 
-def train(model, loss, optimizer, x, y):
-    x = Variable(x, requires_grad=False)
-    y = Variable(y, requires_grad=False)
+def train(model, loss, optimizer, x_val, y_val):
+    x = Variable(x_val, requires_grad=False)
+    y = Variable(y_val, requires_grad=False)
 
     # Reset gradient
     optimizer.zero_grad()
@@ -55,9 +55,9 @@ def train(model, loss, optimizer, x, y):
     return output.data[0]
 
 
-def predict(model, x):
-    var_x = Variable(x, requires_grad=False)
-    output = model.forward(var_x)
+def predict(model, x_val):
+    x = Variable(x_val, requires_grad=False)
+    output = model.forward(x)
     return output.data.numpy().argmax(axis=1)
 
 
@@ -75,7 +75,7 @@ def main():
     n_classes = 10
     model = ConvNet(output_dim=n_classes)
     loss = torch.nn.CrossEntropyLoss(size_average=True)
-    optimizer = optim.RMSprop(model.parameters(), lr=0.001)
+    optimizer = optim.SGD(model.parameters())
     batch_size = 100
 
     for i in range(100):