Stack Overflow
Stack Overflow

Questions tagged [wandb]

134 questions
8
votes
2 answers

Weights and Biases error: The wandb backend process has shutdown

running the colab linked below, I get the following error: "The wandb backend process has shutdown" I see nothing suspicious in the way the colab uses wandb and I couldn't find anyone with the same problem. Any help is greatly appreciated. I am…
Frido
  • 101
  • 2
  • 6
7
votes
2 answers

Multiple lines on same plot with incremental logging - wandb

I am using weights & biases (wandb). I want to group multiple plots into one while using incremental logging, any way to do that? Say we have 10 metrics, I can add them to the project incrementally, gradually building 10 graphs: import wandb import…
Julep
  • 760
  • 1
  • 6
  • 18
6
votes
2 answers

When is one supposed to run wandb.watch so that weights and biases tracks params and gradients properly?

I was trying out the wandb library and I run wandb.watch but that doesn't seem to work on my code. It's not supposed to be anything to complicated so I am puzzled why it's not…
Charlie Parker
  • 5,884
  • 57
  • 198
  • 323
6
votes
1 answer

Weights and Biases: Login and network errors

I recently installed Weights and Biases (wandb) for recording the metrics of my machine learning projects. Everything worked fine when connected to wandb cloud instance or when I used a local docker image. Now, when I tried to access my local wandb…
Adeel Hashmi
  • 767
  • 1
  • 8
  • 20
5
votes
1 answer

Wandb throws Permission denied error although I am logged in

I am using the cleanrl library, in particular the script dqn_atari.py dqn_atari.py where I followed the instructions in order to save and load the target and Q-network. I am running it locally within a conda environment. I haven't loaded…
blindeyes
  • 409
  • 3
  • 13
5
votes
3 answers

logging learning rate schedule in keras via weights and biases

I am training a keras model and using a custom learning rate scheduler for the optimizer (of type tf.keras.optimizers.schedules.LearningRateSchedule), and i want to log the learning rate change via the weights&biases framework. i couldn't find how…
Matan Halfon
  • 59
  • 3
4
votes
1 answer

What is the official way to run a wandb sweep with hugging face (HF) transformers so that all the HF features work e.g. distributed training?

Intially I wanted to run a hugging face run such that if the user wanted to run a sweep they could (and merge them with the command line arguments given) or just execute the run with the arguments from command line. The merging is so that the train…
Charlie Parker
  • 5,884
  • 57
  • 198
  • 323
4
votes
1 answer

weights & biases : ERROR Failed to sample metric: Not Supported

I am training a yolox model and using wandb (weight & biases library) to follow training evolution. My problem is that when I am loading wandb library (version 0.13.5) I get an error message, which is: wandb: ERROR Failed to sample metric: Not…
Sebastien Grand
  • 51
  • 2
4
votes
1 answer

PyTorch Lightning wants create a folder on import due to usage of wandb, which raises error on AWS Lambda

So I want to build a Docker image with PyTorch Lightning that can be used with AWS lambda. However, when the function is invoked it raises an OS Error, that claims it uses a Read-only file system and wandb.py wants to write something. I tried these…
Lau
  • 1,353
  • 7
  • 26
3
votes
1 answer

AttributeError: module 'collections' has no attribute 'MutableSet'

I've written an MLflow component to upload & log an artifact from a URL to WandB. I am trying to run the script inside a conda environment but unfortunately I am facing this error File…
Prashant singh
  • 41
  • 1
  • 3
3
votes
1 answer

How to prevent Weights & Biases from saving unnecessary parameters

I am using Weights & Biases (link) to manage hyperparameter optimization and log the results. I am training using Keras with a Tensorflow backend, and I am using the out-of-the-box logging functionality of Weights & Biases, in which I…
book_kees
  • 307
  • 1
  • 9
3
votes
1 answer

wandb - how to get it really silent (weights and biases)

Working with Anaconda-Spyder (python 3.7), I installed the latest release of wandb (0.10.7) and try to use it with tensorflow (2.1.0) and keras (2.3.1). Since then, my console is polluted with lengthy comments due to wandb. So far I am using config…
JDE65
  • 89
  • 2
  • 8
2
votes
1 answer

Hyperparameter Tuning with Wandb Sweep for custom parameters

I'm trying to tune the hyperparameters using the Stable-Baseline-3 Library for the network architecture. My configuration file is: program: main.py method: bayes name: sweep metric: goal: minimize name: train/loss parameters: batch_size: …
Marcelo Ruiz
  • 373
  • 3
  • 14
2
votes
2 answers

How to stop logging locally but only save to wandb's servers and have wandb work using soft links?

I am having a weird issue where I change the location of all my code & data to a different location with more disk space, then I soft link my projects & data to those locations with more space. I assume there must be some file handle issue because…
Charlie Parker
  • 5,884
  • 57
  • 198
  • 323
2
votes
3 answers

Google Colab: "wandb" ModuleNotFoundError after installation

The "wandb" package cannot be found even after I see it has been listed in pip list. To be more clear, I upgrade my python version to 3.9.15 in Colab, and then force to reinstall pip since somehow the pip module will disappear if I try to upgrade…
Laiborn
  • 21
  • 1
  • 3
1
2 3
…
8 9