Update Dockerfile

Dockerfile

@@ -1,30 +1,99 @@
-FROM python:3.9
+from aim import Run
+from aim.pytorch import track_gradients_dists, track_params_dists
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torchvision import datasets, transforms
+from tqdm import tqdm
 
-RUN useradd -m -u 1000 aim_user
+batch_size = 64
+epochs = 10
+learning_rate = 0.01
 
-# Switch to the "aim_user" user
-USER aim_user
+aim_run = Run()
 
-# Set home to the user's home directory
-ENV HOME=/home/aim_user \
-	PATH=/home/aim_user/.local/bin:$PATH
+class CNN(nn.Module):
+    def __init__(self):
+        super(CNN, self).__init__()
+        self.conv1 = nn.Conv2d(1, 32, 3, 1)
+        self.conv2 = nn.Conv2d(32, 64, 3, 1)
+        self.pool = nn.MaxPool2d(2, 2)
+        self.fc1 = nn.Linear(64 * 7 * 7, 128)
+        self.fc2 = nn.Linear(128, 10)
 
-# Set the working directory to the user's home directory
-WORKDIR $HOME
+    def forward(self, x):
+        x = self.pool(torch.relu(self.conv1(x)))
+        x = self.pool(torch.relu(self.conv2(x)))
+        x = torch.flatten(x, 1)
+        x = torch.relu(self.fc1(x))
+        x = self.fc2(x)
+        return x
 
-# install the `aim` package on the latest version
-RUN pip install aim
+train_dataset = datasets.MNIST(root='./data',
+                               train=True,
+                               transform=transforms.ToTensor(),
+                               download=True)
 
-RUN aim telemetry off
+test_dataset = datasets.MNIST(root='./data',
+                              train=False,
+                              transform=transforms.ToTensor())
 
-ENTRYPOINT ["/bin/sh", "-c"]
+train_loader = torch.utils.data.DataLoader(dataset=train_dataset,
+                                           batch_size=batch_size,
+                                           shuffle=True)
 
-COPY aim_repo.tar.gz .
-RUN tar xvzf aim_repo.tar.gz
-# have to run `aim init` in the directory that stores aim data for
-# otherwise `aim up` will prompt for confirmation to create the directory itself.
-# We run aim listening on 0.0.0.0 to expose all ports. Also, we run
-# using `--dev` to print verbose logs. Port 43800 is the default port of
-# `aim up` but explicit is better than implicit.
-CMD ["aim up --host 0.0.0.0 --port 7860 --workers 2"]
+test_loader = torch.utils.data.DataLoader(dataset=test_dataset,
+                                          batch_size=batch_size,
+                                          shuffle=False)
 
+model = CNN()
+optimizer = optim.Adam(model.parameters(), lr=learning_rate)
+criterion = nn.CrossEntropyLoss()
+
+for epoch in range(epochs):
+    model.train()
+    train_loss = 0
+    correct = 0
+    total = 0
+
+    for batch_idx, (data, target) in enumerate(tqdm(train_loader, desc="Training", leave=False)):
+        optimizer.zero_grad()
+        output = model(data)
+        loss = criterion(output, target)
+        loss.backward()
+        optimizer.step()
+
+        train_loss += loss.item()
+        _, predicted = torch.max(output.data, 1)
+        total += target.size(0)
+        correct += (predicted == target).sum().item()
+
+    acc = correct / total
+    items = {'accuracy': acc, 'loss': train_loss / len(train_loader)}
+    aim_run.track(items, epoch=epoch, context={'subset': 'train'})
+
+    track_params_dists(model, aim_run, epoch=epoch, context={'subset': 'train'})
+    track_gradients_dists(model, aim_run, epoch=epoch, context={'subset': 'train'})
+
+    model.eval()
+    test_loss = 0
+    correct = 0
+    total = 0
+
+    with torch.no_grad():
+        for batch_idx, (data, target) in enumerate(tqdm(test_loader, desc="Testing", leave=False)):
+            output = model(data)
+            loss = criterion(output, target)
+            test_loss += loss.item()
+            _, predicted = torch.max(output.data, 1)
+            total += target.size(0)
+            correct += (predicted == target).sum().item()
+
+    acc = correct / total
+    items = {'accuracy': acc, 'loss': test_loss / len(test_loader)}
+    aim_run.track(items, epoch=epoch, context={'subset': 'test'})
+
+    track_params_dists(model, aim_run, epoch=epoch, context={'subset': 'test'})
+    track_gradients_dists(model, aim_run, epoch=epoch, context={'subset': 'test'})
+
+torch.save(model.state_dict(), 'mnist_cnn.pth')