Documentation
Search…
wandb.watch
Hooks into the torch model to collect gradients and the topology.
1
watch(
2
models, criterion=None, log="gradients", log_freq=1000, idx=None,
3
log_graph=(False)
4
)
Copied!
Should be extended to accept arbitrary ML models.
Args
Text
models
(torch.Module) The model to hook, can be a tuple
criterion
(torch.F) An optional loss value being optimized
log
(str) One of "gradients", "parameters", "all", or None
log_freq
(int) log gradients and parameters every N batches
idx
(int) an index to be used when calling wandb.watch on multiple models
log_graph
(boolean) log graph topology
Returns
Text
wandb.Graph The graph object that will populate after the first backward pass
Raises
Text
ValueError
If called before wandb.init or if any of models is not a torch.nn.Module.
Last modified 4h ago
Copy link