Stack Overflow
Stack Overflow

Questions tagged [wandb]

134 questions
0
votes
1 answer

Wandb training kills kernel in jupyter lab

In my jupyter I can train my model on batch_size=8, but when I use wandb always after 9 iterations the process is killed and kernel restarts. What's more weird is that the same code worked on colab, but with my GPU (RTX 3080) I can never finish the…
bindas
  • 33
  • 5
0
votes
1 answer

What does "save_graph" keyword in WandbCallback mean?

I'm using Weights and Biases to track my deep learning models. To monitor everything I use the WandbCallback in .fit. In the WandbCallback documentation there is the keyword save_graph which defaults to True. The description is very brief and I…
Nicolai
  • 62
  • 5
0
votes
1 answer

How to log artifacts in wandb while using saimpletransformers?

I am creating a Question Answering model using simpletransformers. I would also like to use wandb to track model artifacts. As I understand from wandb docs, there is an integration touchpoint for simpletransformers but there is no mention of logging…
terokd
  • 13
  • 4
0
votes
1 answer

Weights & Biases with Transformers and PyTorch?

I'm training an NLP model at work (e-commerce SEO) applying a BERT variation for portuguese language (BERTimbau) through Transformers by Hugging Face. I didn't used the Trainer from Transformers API. I used PyTorch to set all parameters through…
Guilherme Giuliano Nicolau
  • 143
  • 1
  • 11
0
votes
0 answers

AttributeError: 'NoneType' object has no attribute '_global_run_stack'

Description I am using PTAN library with an A3C model and I am trying to work with wandb sweep but I've encountered some weird problems, and I am not sure if it's a bug regarding sweep (because if I am going to use just a simple model without any…
Mircea
  • 1,671
  • 7
  • 25
  • 41
0
votes
0 answers

wandb logging PermissionError and OSError

Description: When running experiments using Weights and Biases (wandb), I occasionally get a PermissionError for Python's logging library and OSError for accessing the TLS CA cert. I had the following stacktrace, repeated many times with…
peractio
  • 593
  • 1
  • 5
  • 14
0
votes
1 answer

Control the logging frequency and contents when using wandb with HuggingFace

I am using the wandb with my HuggingFace code. I would like to log the loss and other metrics. Now I have two questions How does wandb decide when to log the loss? Is this decided by logging_steps in TrainingArguments(...)？ training_args =…
Mr.Robot
  • 349
  • 1
  • 16
0
votes
1 answer

'WANDB_MODE' is not recognized as an internal or external command, operable program or batch file

I am trying to run the custom yolo model on my data set in my local machine. I am following some reference code from the kaggle platform. Here first time I encounter the wandb frame work. while doing so I use the following to run the train.py file…
anil
  • 37
  • 1
  • 8
0
votes
2 answers

access logged values during runtime

How can I retrieve a logged value from wandb before the run was finished? import os import wandb wandb.init(project='someproject') def loss_a(): # do_stuff and log: wandb.log({"loss_a": 1.0}) def loss_b(): # do_stuff and log: …
nivniv
  • 3,421
  • 5
  • 33
  • 40
-1
votes
0 answers

How to fix the mysterious Rate Limit Issues in wandb when logging a single scale at the end?

Has anyone experienced wandb errors about logging to often? (funny I only log a single number [diversity score] at the very end, I'm not even sure what is calling wandb, besides my print statements being logged). Errors: wandb: 429 encountered…
Charlie Parker
  • 5,884
  • 57
  • 198
  • 323
-1
votes
1 answer

Error when not using the defined layer in call method in Keras model with WandB integration

When defining a new Keras model and integrating it with WandB (Weights and Biases), I encountered an error related to a specific layer. The error message I received is as follows: File…
Salman
  • 924
  • 1
  • 7
  • 20
-1
votes
1 answer

Weights and Biases (wandb), Explain metrics displayed in WandB

I'm looking for a thorough explanation of all the plots displayed in a basic WandB dashboard, such as: train/loss train/global_step train/train_samples_per_second train/train_loss This is part of my dashboard Mainly, I'm looking for practical…
arnle
  • 480
  • 4
  • 10
-1
votes
2 answers

How to use StratifiedKFold with wandb sweeps?

I have the following piece of code - it is a train function for Logistic regression. I run sweeps to be able to compare hyperparameter tuning results. My issue is that I don't know how to incorporate StratifiedKFold to work with sweeps. I would…
Yana
  • 785
  • 8
  • 23
-1
votes
1 answer

what is the meaning of allow anonymous in wandb

i am trying to use wandb for visualization but I am confused about this line andb_init_kwargs=dict(anonymous="allow") it means anyone can see visualization with having link ? WandbCallback( wandb_init_kwargs=dict(anonymous="allow"), # for Anonymous…
user12
  • 761
  • 8
  • 24
1 2 3
…
8
9