diff --git a/2_logistic_regression.py b/2_logistic_regression.py
index f4ff9d4..0494090 100644
--- a/2_logistic_regression.py
+++ b/2_logistic_regression.py
@@ -52,7 +52,7 @@ def main():
     n_examples, n_features = trX.size()
     n_classes = 10
     model = build_model(n_features, n_classes)
-    loss = torch.nn.CrossEntropyLoss(size_average=True)
+    loss = torch.nn.CrossEntropyLoss(reduction='elementwise_mean')
     optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.9)
     batch_size = 100
 
diff --git a/3_neural_net.py b/3_neural_net.py
index e725d64..b8db316 100644
--- a/3_neural_net.py
+++ b/3_neural_net.py
@@ -51,7 +51,7 @@ def main():
     n_examples, n_features = trX.size()
     n_classes = 10
     model = build_model(n_features, n_classes)
-    loss = torch.nn.CrossEntropyLoss(size_average=True)
+    loss = torch.nn.CrossEntropyLoss(reduction='elementwise_mean')
     optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.9)
     batch_size = 100
 
diff --git a/4_modern_neural_net.py b/4_modern_neural_net.py
index 4f330c2..84f970e 100644
--- a/4_modern_neural_net.py
+++ b/4_modern_neural_net.py
@@ -55,7 +55,7 @@ def main():
     n_examples, n_features = trX.size()
     n_classes = 10
     model = build_model(n_features, n_classes)
-    loss = torch.nn.CrossEntropyLoss(size_average=True)
+    loss = torch.nn.CrossEntropyLoss(reduction='elementwise_mean')
     optimizer = optim.Adam(model.parameters())
     batch_size = 100
 
diff --git a/5_convolutional_net.py b/5_convolutional_net.py
index f53c511..d877961 100644
--- a/5_convolutional_net.py
+++ b/5_convolutional_net.py
@@ -72,7 +72,7 @@ def main():
     n_examples = len(trX)
     n_classes = 10
     model = ConvNet(output_dim=n_classes)
-    loss = torch.nn.CrossEntropyLoss(size_average=True)
+    loss = torch.nn.CrossEntropyLoss(reduction='elementwise_mean')
     optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.9)
     batch_size = 100
 
diff --git a/6_lstm.py b/6_lstm.py
index b8ccb52..6b9f44e 100644
--- a/6_lstm.py
+++ b/6_lstm.py
@@ -73,7 +73,7 @@ def main():
     trY = torch.from_numpy(trY).long()
 
     model = LSTMNet(input_dim, hidden_dim, n_classes)
-    loss = torch.nn.CrossEntropyLoss(size_average=True)
+    loss = torch.nn.CrossEntropyLoss(reduction='elementwise_mean')
     optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.9)
 
     for i in range(epochs):