W&B provides first class support for PyTorch. To automatically log gradients and store the network topology, you can call watch and pass in your PyTorch model.

import wandb
wandb.init(config=args)
​
# Magic
wandb.watch(model)
​
model.train()
for batch_idx, (data, target) in enumerate(train_loader):
    output = model(data)
    loss = F.nll_loss(output, target)
    loss.backward()
    optimizer.step()
    if batch_idx % args.log_interval == 0:
        wandb.log({"loss": loss})

Gradients, metrics and the graph won't be logged until wandb.log is called after a forward and backward pass.

See this colab notebook for an end to end example of integrating wandb with PyTorch, including a video tutorial. You can also find more examples in our example projects section.

Options

By default the hook only logs gradients.

Images

You can pass PyTorch tensors with image data into wandb.Image and torchvision utils will be used to log them automatically.

To log images and view them in the Media panel, you can use the following syntax:

wandb.log({"examples" : [wandb.Image(i) for i in images]})

Multiple Models

If you need to track multiple models in the same script, you can wall wandb.watch() on each model separately.

Example

We've created a few examples for you to see how the integration works:

• ​Run in Google Colab: A simple notebook example to get you started

• ​Example on Github: MNIST example in a Python script

• ​Wandb Dashboard: View result on W&B