# szagoruyko/pytorchviz

A small package to create visualizations of PyTorch execution graphs

Repository: https://github.com/szagoruyko/pytorchviz
Canonical: https://ross.abutalabs.com/products/pytorchviz
Language: Jupyter Notebook
License: MIT
License Family: permissive
Last push: 2024-12-30T21:14:55+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3137, "days_push": 611, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3500, forks 290 (observed 2026-08-28T04:08:07.525304+00:00)

## What it is
A small Python package (torchviz) that generates visualizations of PyTorch execution graphs and autograd traces using Graphviz. It renders computation graphs of model outputs, optionally showing saved tensors and backward-pass attributes.

## Use cases
- visualize pytorch computation graph
- debug autograd backward pass dependencies
- see what tensors autograd saves for backward
- render neural network execution graph as image
- inspect model graph in jupyter notebook

## When to choose
- you need quick Graphviz diagrams of PyTorch autograd graphs
- you're debugging gradient flow or memory usage in a PyTorch model
- you want a lightweight pip-installable visualization utility

## When to avoid
- you need interactive or runtime training dashboards like TensorBoard
- you're not using PyTorch
- you need full model architecture diagrams rather than execution graphs

## Facets
- artifact type: library
- maturity: maintenance
- function: data-visualization, machine-learning, developer-tools
- domain: deep-learning, data-visualization, developer-tools
- platform: python, cross-platform
- tags: pytorch, autograd, computational-graph, graphviz, debugging

## Member repositories
- szagoruyko/pytorchviz (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:07.525304+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 2026-08-29T18:35:49.102168+00:00, confidence not recorded.
  - readme: https://github.com/szagoruyko/pytorchviz (fetched 2026-08-28T04:08:07.525304+00:00, sha 21d7abcbf3ff)
- Data as of 2026-08-30T08:39:29.467469+00:00.