# ericjang/tdb

Interactive, node-by-node debugging and visualization for TensorFlow

Repository: https://github.com/ericjang/tdb
Canonical: https://ross.abutalabs.com/products/tdb
Language: JavaScript
License: Apache-2.0
License Family: permissive
Last push: 2017-01-27T04:42:40+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 3917, "days_push": 3505, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1350, forks 139 (observed 2026-08-28T04:04:28.122065+00:00)

## What it is
TensorDebugger (TDB) is a visual debugger for TensorFlow deep learning models, combining a Python library with a Jupyter notebook extension. It adds breakpoints on graph ops and tensors plus real-time visualization of data flowing through the computational graph during training.

## Use cases
- debug tensorflow models node by node
- set breakpoints on tensors in a computational graph
- visualize weights and gradients during training
- plot histograms of activations in real time
- find training bugs in neural networks faster
- inspect deep learning model internals in jupyter

## When to choose
- you are working with older TensorFlow 1.x codebases and want visual graph debugging
- you want real-time plots of tensors while a network trains
- you prefer a Jupyter-based debugging workflow for deep learning

## When to avoid
- you use modern TensorFlow, where the official tfdbg debugger is the maintained alternative
- you use PyTorch or other frameworks, since TDB is TensorFlow-specific
- you need actively maintained software with recent fixes

## Facets
- artifact type: library
- maturity: abandoned
- function: machine-learning, developer-tools, data-visualization
- domain: deep-learning, machine-learning, developer-tools, data-visualization
- platform: python, browser
- tags: tensorflow, debugger, jupyter-notebook, visualization, breakpoints, deep-learning, debugging

## Member repositories
- ericjang/tdb (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:28.122065+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 2026-08-30T04:42:16.437522+00:00, confidence not recorded.
  - readme: https://github.com/ericjang/tdb (fetched 2026-08-28T04:04:28.122065+00:00, sha 2a102b5481d2)
- Data as of 2026-08-30T08:39:29.467469+00:00.