Training and validation loss tracking in PyTorch - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to initialize the training loss list.

PyTorch

train_losses = [1]

Drag options to blanks, or click blank then click option'

A{}

B[]

C0

DNone

Attempts:

3 left

💡 Hint

Common Mistakes

Using a dictionary {} instead of a list.

Initializing with 0 or None which cannot store multiple values.

We use an empty list [] to store training loss values over epochs.

2fill in blank

medium

Complete the code to calculate the average training loss for an epoch.

PyTorch

epoch_train_loss = sum(train_losses) / [1]

Drag options to blanks, or click blank then click option'

Alen(train_losses)

Bmax(train_losses)

Cmin(train_losses)

Dtrain_losses

Attempts:

3 left

💡 Hint

Common Mistakes

Dividing by the maximum or minimum loss instead of the count.

Dividing by the list itself which causes an error.

To get the average loss, divide the sum by the number of loss values using len(train_losses).

3fill in blank

hard

Fix the error in the code to append validation loss after each epoch.

PyTorch

validation_losses.[1](val_loss)

Drag options to blanks, or click blank then click option'

Aadd

Binsert

Cappend

Dextend

Attempts:

3 left

💡 Hint

Common Mistakes

Using add which is not a list method.

Using extend which expects an iterable.

Using insert without specifying an index.

Use append to add a single loss value to the list.

4fill in blank

hard

Fill both blanks to compute and store average training and validation losses.

PyTorch

avg_train_loss = sum(train_losses) [1] len(train_losses)
avg_val_loss = sum(validation_losses) [2] len(validation_losses)

Drag options to blanks, or click blank then click option'

A/

B*

C-

D+

Attempts:

3 left

💡 Hint

Common Mistakes

Using multiplication or addition instead of division.

Using subtraction which gives incorrect results.

Divide the sum of losses by the count to get the average for both training and validation.

5fill in blank

hard

Fill all three blanks to track losses during training and validation phases.

PyTorch

for epoch in range(num_epochs):
    model.train()
    train_loss = 0
    for data, target in train_loader:
        optimizer.zero_grad()
        output = model(data)
        loss = criterion(output, target)
        loss.[1]()
        optimizer.step()
        train_loss += loss.item()
    train_losses.[2](train_loss / len(train_loader))

    model.eval()
    val_loss = 0
    with torch.no_grad():
        for data, target in val_loader:
            output = model(data)
            loss = criterion(output, target)
            val_loss += loss.item()
    validation_losses.[3](val_loss / len(val_loader))

Drag options to blanks, or click blank then click option'

Abackward

Bappend

Dstep

Attempts:

3 left

💡 Hint

Common Mistakes

Using step instead of backward for gradients.

Using step or wrong methods to add losses to lists.

Call loss.backward() to compute gradients, then append average losses to the lists.