Stack Overflow
Stack Overflow

Questions tagged [wandb]

134 questions
0
votes
1 answer

Create a Scalar Chart in Weights & Biases using python

I'm using weights & biases and try to log some scalars using python SDK for W&B, the logs appear like that: a plot with a single dot I get So I basically just run this: wandb.log({"loss": 0.5}) And looks weird, In the UI there's an opttion of…
Maxim
  • 11
0
votes
0 answers

How do I run a "regular" W&B session after I've done a sweep?

I have the following code def train(config=None, run_num=None, project=None): with wandb.init(project=project, config=config) as run: config = wandb.config # Working training code... sweep_id = wandb.sweep(sweep_config,…
orome
  • 45,163
  • 57
  • 202
  • 418
0
votes
0 answers

Checkpoints and model training

I'm very new to training on bigger datasets and would appreciate some advice. I have a very large dataset that I am training on Google Colab. Colab has a 12-hour period, and it stops the runtime. I have built-in checkpoints every so often. If Colab…
Mark Stent
  • 3
  • 1
0
votes
0 answers

How to load a pytorch grid in wandb?

I'm training a network using pytorch (no lightning or other modules) and I'm using W&B to monitor the training metrics. During the epochs, I'm also loading some images that allow me to keep track of the learning process. Since I had an error "Fail…
SahFra98
  • 80
  • 5
0
votes
0 answers

ConnectionAbortedError: [WinError 10053] during W&B hyperparameter search

I am trying to do hyperparameter seach using Weights & Biases. Tried to run this code from documentation https://docs.wandb.ai/guides/sweeps/add-w-and-b-to-your-code in the notebook: import wandb import numpy as np import random # Define sweep…
Aska
  • 141
  • 1
  • 10
0
votes
1 answer

Weights and biases - saving and loading models to continue training

I am trying to set up my model to run, train and save everything into and from Weights and Biases. I would like to: Load the latest model from W and B and continue training with it from the last epoch that i left on/place it ended on the last model…
Mark Stent
  • 3
  • 1
0
votes
1 answer

Forgot local wandb email

I have been logging and browsing locally on wandb, but after some time and updates I got logged out and can't remember the email I used locally. How can I recover my local credentials?
Samuel
  • 86
  • 1
  • 7
0
votes
0 answers

HyperParameterOptimization with Wandb and huggingface connect error?

I am trying to run hyperparameter optimization on huggingface transformer using wandb. I tried following the example in documentation in hugginface: https://huggingface.co/docs/transformers/hpo_train I am getting this weird error when trying to do…
Rawstone
  • 1
0
votes
0 answers

How do I store a sweep_id in a cli environment variable so that it runs in a wandb agent later?

how do I run a wandb sweep in a cli/terminal/bash and store the sweep id in a env variable to use it later in the wandb agent cli command? Is this really correct/recommended way? export SWEEP_ID=$(wandb sweep sweep.yaml --project …
Charlie Parker
  • 5,884
  • 57
  • 198
  • 323
0
votes
0 answers

Colored text in a wandb table

For error analysis (for my ASR model), I want to report a table to wandb where the text is color coded to show the difference between the predicted transcript and the ground truth (e.g. deleted words gray, inserted blue, replaced red). I tried HTML,…
DaveFar
  • 7,078
  • 4
  • 50
  • 90
0
votes
0 answers

How do I select a GPU before running a wandb agent?

I wanted to run a set of wandb agents but the server that I have access to does not have slurm or any workload manager. How do I have the agent code select an CUDA_VISIBLE_DEVICE automatically according to the gpu memory I need and is available? I…
Charlie Parker
  • 5,884
  • 57
  • 198
  • 323
0
votes
0 answers

wandb.sdk.wandb_manager.ManagerConnectionRefusedError: Connection to wandb service failed: [WinError 10061]

I have a problem about AI project use yolov5 and WandB. Developing Environment has packages include code: import os os.environ["KMP_DUPLICATE_LIB_OK"]="TRUE" !python train.py --data data/blood_cell.yaml --batch-size 4 --epochs 300 --img-size 640…
hoon050824
  • 11
  • 3
0
votes
1 answer

Is there a way to scale the x axis in wandb in the website?

I'm working with weight and bias(wandb). My data is already logged but I would like to scale the plot, as my data was logged every 5 epochs I would like to at least change the numbers in the scale so they stay true to their nature. For example, my…
Roberto Gutiérrez
  • 3
  • 2
0
votes
1 answer

Optuna & Wandb - how to enable logging of each trial separately?

Within my optuna study, I want that each trial is separately logged by wandb. Currently, the study is run and the end result is tracked in my wandb dashboard. Instead of showing each trial run separately, the end result over all epochs is shown. So,…
resei09
  • 23
  • 4
0
votes
3 answers

Pytorch Lightning - How to log a W&B histogram?

Lightning AI has a W&B logger (https://lightning.ai/docs/pytorch/latest/extensions/generated/pytorch_lightning.loggers.WandbLogger.html), but the documentation doesn't specify how to log a W&B Histogram. Can someone please tell me how to log a W&B…
Rylan Schaeffer
  • 1,945
  • 2
  • 28
  • 50
1 2 3
…
8 9