DeepChecks

Integrate W&B with Deepchecks to validate ML models and datasets with automated testing and experiment tracking.

DeepChecks helps you validate your machine learning models and data with minimal effort. You can verify your data’s integrity, inspect its distributions, validate data splits, evaluate your model, and compare different models. This page shows how to use the DeepChecks integration with W&B so you can log validation results and test suites alongside your experiments.

For more information, see the DeepChecks W&B integration guide.

Get started

To use DeepChecks with W&B, first sign up for a W&B account. With the W&B integration in DeepChecks, you can get started by running a single check and pushing the result to W&B:

import wandb

wandb.login()

# import your check from deepchecks
from deepchecks.checks import ModelErrorAnalysis

# run your check
result = ModelErrorAnalysis()

# push that result to wandb
result.to_wandb()

In addition to logging individual checks, you can log an entire DeepChecks test suite to W&B:

import wandb

wandb.login()

# import your full_suite tests from deepchecks
from deepchecks.suites import full_suite

# create and run a DeepChecks test suite
suite_result = full_suite().run(...)

# push thes results to wandb
# here you can pass any wandb.init configs and arguments you need
suite_result.to_wandb(project="my-suite-project", config={"suite-name": "full-suite"})

Example

To see what the integration looks like in practice, explore the Validate your data and models with Deepchecks and W&B report, which demonstrates how to use DeepChecks and W&B together.

If you have questions or issues about this W&B integration, open an issue in the DeepChecks GitHub repository and we’ll get you an answer.