Experiment Tracking
for Deep Learning

"W&B is a key piece of our fast-paced, cutting-edge, large-scale research workflow: great flexibility, performance, and user experience."
Adrien Gaidon
Toyota Research Institute
"W&B allows us to scale up insights from a single researcher to the entire team and from a single machine to thousands."
Wojciech Zaremba
Cofounder of OpenAI
Read the full interview with Peter Welinder from OpenAI Robotics →
Lightweight Integration
Add W&B to your training script with just a few lines, and we'll automatically capture and log system metrics for you.
import wandb
wandb.init()
Log Anything
Easily log metrics from your script to visualize in real time as your model trains.
wandb.log({"acc": accuracy, "val_acc": val_accuracy})
Track Insights at Scale
Collaborate with your coworkers and keep track of thousands of experiments in one centralized project.
for x in range (0, 1000):
   train()
Visualize Output
Peek under the hood and see what your model is producing at each timestep. Here we can see a colorizer model's output in the center compared to the ground truth on the right.
wandb.log({"examples": flower_images})
Compare Runs
Visualize performance differences across thousands of runs.
Example Projects
Sign Language Classifier
Digits Classifier
Face Emotion Classifier
Try Weights & Biases now.
Easily integrate our powerful tools, free for public projects.
Create AccountRequest Demo