# tensorflow/model-analysis

Model analysis tools for TensorFlow

Repository: https://github.com/tensorflow/model-analysis
Canonical: https://ross.abutalabs.com/products/model-analysis
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-08-14T20:41:19+00:00

## Health v2 (maintenance only)
Score: 83/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 97, release rhythm 56, longevity 100
- inputs: {"age_days": 3085, "days_push": 19, "days_rel": 82, "gap_med": 207, "n_releases_24m": 4}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1265, forks 279 (observed 2026-08-28T04:04:11.049309+00:00)

## What it is
TensorFlow Model Analysis (TFMA) is a Python library for evaluating TensorFlow models on large datasets in a distributed manner. It computes trainer-defined metrics over different data slices and visualizes results, e.g., in Jupyter notebooks.

## Use cases
- evaluate tensorflow models on large datasets
- compute model metrics over data slices
- visualize model evaluation results in jupyter notebooks
- compare model performance across subgroups
- detect model bias with fairness metrics
- integrate model evaluation into tfx pipelines

## When to choose
- you train models with TensorFlow and want consistent metrics between training and evaluation
- you need distributed evaluation over large datasets
- you want slice-based analysis or fairness metrics for model debugging

## When to avoid
- your models are built with PyTorch, JAX, or scikit-learn rather than TensorFlow
- you only need simple evaluation on small datasets where a single-machine library suffices
- you need a stable 1.0 API since TFMA may introduce breaking changes before 1.0

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, data-visualization, analytics, benchmarking
- domain: machine-learning, data-science, analytics, developer-tools
- platform: python, cross-platform
- tags: tensorflow, model-evaluation, tfma, tfx, slicing-metrics, jupyter, fairness-indicators

## Member repositories
- tensorflow/model-analysis (main) score 83

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:11.049309+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T05:04:00.574832+00:00, confidence not recorded.
  - readme: https://github.com/tensorflow/model-analysis (fetched 2026-08-28T04:04:11.049309+00:00, sha 6a44f6be7775)
- Data as of 2026-08-30T08:39:29.467469+00:00.