# lanpa/tensorboardX

tensorboard for pytorch (and chainer, mxnet, numpy, ...)

Repository: https://github.com/lanpa/tensorboardX
Canonical: https://ross.abutalabs.com/products/tensorboardx
Homepage: https://tensorboardx.readthedocs.io/en/latest/tensorboard.html
Language: Python
License: MIT
License Family: permissive
Topics: pytorch, tensorboard, machine-learning, visualization, numpy
Last push: 2026-07-14T16:48:08+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 92, release rhythm 46, longevity 100
- inputs: {"age_days": 3368, "days_push": 50, "days_rel": 151, "gap_med": 211, "n_releases_24m": 2}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7998, forks 852 (observed 2026-08-28T04:10:11.491060+00:00)

## What it is
A Python library for writing TensorBoard event files from PyTorch, Chainer, MXNet, NumPy, and other frameworks without needing TensorFlow. It supports logging scalars, images, histograms, audio, text, graphs, embeddings, and more via a simple SummaryWriter API.

## Use cases
- visualize pytorch training metrics in tensorboard
- log scalars and histograms during model training
- log images and audio from training without tensorflow
- track hyperparameters and pr curves in tensorboard
- write tensorboard events from numpy arrays

## When to choose
- you train models in PyTorch or other non-TensorFlow frameworks and want TensorBoard dashboards
- you need lightweight event-file writing with a simple function-call API

## When to avoid
- you use TensorFlow, which has built-in TensorBoard support
- you need a full experiment-tracking platform with remote dashboards like W&B or MLflow

## Facets
- artifact type: library
- maturity: active
- function: data-visualization, logging, machine-learning
- domain: machine-learning, data-visualization, deep-learning
- platform: python, cross-platform
- tags: tensorboard, pytorch, training-visualization, summary-writer

## Member repositories
- lanpa/tensorboardX (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:11.491060+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 2026-08-29T17:32:32.752145+00:00, confidence not recorded.
  - readme: https://github.com/lanpa/tensorboardX (fetched 2026-08-28T04:10:11.491060+00:00, sha 38fe0d8301e1)
  - registry_pypi: https://pypi.org/pypi/tensorboardx/json (fetched 2026-08-29T08:29:57.870156+00:00, sha 1388d56ba640)
- Data as of 2026-08-30T08:39:29.467469+00:00.