How-to Guides¶

These guides demonstrate more in-depth features and customizations of DataEval features for more advanced users.

In addition to viewing them in our documentation, these notebooks can also be opened in Google Colab to be used interactively!

Detectors¶

• How to identify outliers and/or anomalies in a dataset

• How to detect duplicates in a dataset

• How to identify poor quality images in a dataset

Metrics¶

• How to determine if a dataset can meet performance requirements

• How to compare data distributions between 2 datasets

• How to compare label distributions between 2 datasets

• How to detect undersampled data subsets

• How to add statistics as intrinsic metadata for bias analysis

Workflows¶

• How to determine the amount of data needed to meet image classification performance requirements

Models¶

• How to create image embeddings with an autoencoder