updating with final model

MNISTModel.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca929cef293140ec31967298cfae2f56b0b4a319d445a81cc60dd298064df843
+size 1688783

app.py

@@ -9,8 +9,9 @@ transform = transforms.Compose([
     transforms.Normalize((0.5,), (0.5,))  # Normalize the image
 ])
 
-# Load the trained model    
-net = torch.load('MNISTModel_98.pth')
+# Load the trained model   
+net = MNISTNetwork() 
+net.load_state_dict(torch.load('MNISTModel.pth'))
 LABELS = ['0', '1', '2', '3', '4', '5', '6', '7', '8', '9']
 
 def predict(drawing): 

neural_network.py

@@ -22,7 +22,7 @@ import torch.nn.functional as F
 #         return F.log_softmax(x, dim=1)
 
 class MNISTNetwork(nn.Module):
-    # achieved 98.76 percent accuracy 
+    # achieved 98.783 percent accuracy 
     def __init__(self):
         super().__init__()
         self.conv1 = nn.Conv2d(1, 32, kernel_size=3, padding=1)

train_model.py

@@ -8,12 +8,11 @@ from neural_network import MNISTNetwork
 # hyperparameters 
 BATCH_SIZE = 64
 NUM_WORKERS = 2
-EPOCH = 12
+EPOCH = 15
 LEARNING_RATE = 0.01
 MOMENTUM = 0.5
 LOSS = torch.nn.CrossEntropyLoss()
 
-
 ## Step 1: define our transforms 
 transform = transforms.Compose(
     [
@@ -24,7 +23,10 @@ transform = transforms.Compose(
 
 ## Step 2: get our datasets
 full_ds = torchvision.datasets.MNIST(root='./data', train=True, download=False, transform=transform)
-train_ds, valid_ds = torch.utils.data.random_split(full_ds, [50000, 10000])
+train_size = int(0.8 * len(full_ds))  # Use 80% of the data for training
+val_size = len(full_ds) - train_size  # Use the remaining 20% for validation
+
+train_ds, valid_ds = torch.utils.data.random_split(full_ds, [train_size, val_size])
 test_ds = torchvision.datasets.MNIST(root='./data', train=False, download=False, transform=transform)
 
 ## Step 3: create our dataloaders 
@@ -43,7 +45,7 @@ table.field_names = ['Epoch', 'Training Loss', 'Validation Accuracy']
 
 if __name__ == "__main__":
     multiprocessing.freeze_support()
-    ## begin training process 
+    # begin training process 
     for e in range(EPOCH): 
         model.train()
         running_loss = 0.0
@@ -82,5 +84,5 @@ if __name__ == "__main__":
     test_acc = round((correct/total)*100, 3)
     
     print(f'Test Accuracy: {test_acc}')
-    torch.save(model, 'MNISTModel.pth')
+    torch.save(model.state_dict(), 'MNISTModel.pth')