Run Crashes

How can I fix an error like `AttributeError: module 'wandb' has no attribute ...`?

If you encounter an error like AttributeError: module ‘wandb’ has no attribute ‘init’ or AttributeError: module ‘wandb’ …

How do I fix `Cuda out of memory` during a sweep?

If you see Cuda out of memory during a sweep, refactor your code to use process-based execution. Rewrite your code as a …

How do I kill a job with wandb?

To stop a script instrumented with W&B, press Ctrl+D. Use this method to cancel a training run you no longer need.

How do I resolve a run initialization timeout error in wandb?

A run initialization timeout error means wandb.init() can’t establish a connection to the W&B service within the allotte …

If wandb crashes, will it possibly crash my training run?

W&B runs in a separate process, so if W&B crashes, your training run continues. During an internet outage, W&B keeps ret …

InitStartError: Error communicating with wandb process

This error indicates that the W&B library can’t launch the process that synchronizes data to the server. To resolve the …

My run's state is `crashed` on the UI but is still running on my machine. What do I do to get my data back?

This state usually means you lost connection to your machine during training. To recover your data, run wandb sync PATH_ …

Why does my process stop responding when using Hydra with W&B?

This page explains how to resolve unresponsive processes that occur when you start a process with Hydra alongside W&B. I …

Why does my training hang with distributed training?

This article helps you resolve training hangs when you use W&B with distributed training frameworks, so your runs can st …

Why is a run marked crashed in W&B when it's training fine locally?

A crashed status usually indicates a connection problem. If the server loses internet access and data stops syncing, W&B …