leopard-ai / betty

Betty: an automatic differentiation library for generalized meta-learning and multilevel optimization
https://leopard-ai.github.io/betty/
Apache License 2.0
329 stars 27 forks source link

WandB logging issue #5

Closed asking28 closed 2 years ago

asking28 commented 2 years ago

wandb: (1) Create a W&B account wandb: (2) Use an existing W&B account wandb: (3) Don't visualize my results wandb: Enter your choice: 2 wandb: You chose 'Use an existing W&B account' wandb: Logging into wandb.ai. (Learn how to deploy a W&B server locally: https://wandb.me/wandb-server) wandb: You can find your API key in your browser here: https://wandb.ai/authorize wandb: Paste an API key from your profile and hit enter, or press ctrl+c to quit: wandb: Appending key for api.wandb.ai to your netrc file: /root/.netrc wandb: ERROR Error while calling W&B API: entity pod-tuning not found during upsertBucket (<Response [404]>)

sangkeun00 commented 2 years ago

Hello @asking28,

Thanks for opening the issue. We haven't fully tested wandb logging yet, but will fix this in the near future. I will leave an additional comment once this is added, and you can install the nightly version to use the feature! If you are willing to work on this issue, that would be also very much appreciated!

asking28 commented 2 years ago

@sangkeun00 thanks I'll try to work on that. Thanks for update.

asking28 commented 2 years ago

@sangkeun00 We are setting entity which corresponds to username or teamname, which is causing error in wandb logging. For now I have found this work around: class WandBLogger(LoggerBase): def init(self): try: wandb.init(project="betty", entity=socket.gethostname()) except: wandb.init(project="betty", reinit=True)

sangkeun00 commented 2 years ago

@asking28 Thanks for your work! I will try your code, and after some testing, update the source code for this