fix: use a stable seed in the PyTorch integration

[marcus-campbell]

Description

The PyTorch integration page says this setup ensures deterministic behavior, but it doesn't appear to actually do that. This is because it derives its seeds from hash("..."). By default, Python randomizes string hashes between interpreter processes, so the example can initialize its RNGs differently across runs.

This fix replaces the hash-derived values with one named integer seed.

Before:

torch.backends.cudnn.deterministic = True
random.seed(hash("setting random seeds") % 2**32 - 1)
np.random.seed(hash("improves reproducibility") % 2**32 - 1)
torch.manual_seed(hash("by removing stochasticity") % 2**32 - 1)
torch.cuda.manual_seed_all(hash("so runs are repeatable") % 2**32 - 1)

After:

seed = 42
torch.backends.cudnn.deterministic = True
random.seed(seed)
np.random.seed(seed)
torch.manual_seed(seed)
torch.cuda.manual_seed_all(seed)

Why this matters

A small exploratory search found the same recognizable recipe in at least 24 other repositories. This suggests that the pattern has propagated beyond W&B's own docs, so correcting the bug here can help prevent it from spreading further.

The problem was also noted in the technical-review notes for #2673, but it wasn't fixed as it was out-of-scope for that PR (it was a style guide pass).

The corresponding notebooks are updated in wandb/examples#644.

Testing

• git diff --check succeeds.
• Confirmed that the diff contains only the intended seed changes.
• Confirmed that the old hash-derived seeds vary with PYTHONHASHSEED.
• Confirmed that the replacement has no interpreter hash-secret dependency.
• PR tests succeed.

[mdlinville]

Thank you!

[mdlinville]

Thank you