# tqchen/tinyflow

Tutorial code on how to build your own Deep Learning System in 2k Lines

Repository: https://github.com/tqchen/tinyflow
Canonical: https://ross.abutalabs.com/products/tinyflow
Language: C++
License: Apache-2.0
License Family: permissive
Topics: deep-learning, nnvm, computational-graphs
Archived: true
Last push: 2018-10-04T18:34:44+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3646, "days_push": 2890, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2017, forks 364 (observed 2026-08-28T04:06:06.123877+00:00)

## What it is
TinyFlow is a minimal (~2k lines) educational deep learning system built on NNVM that demonstrates how to construct a computational graph-based framework with a TensorFlow-like API. It uses Torch7 for operator execution and serves as course material for teaching deep learning system internals.

## Use cases
- learn how to build a deep learning framework from scratch
- teach a course on deep learning system design
- study automatic differentiation and computational graphs
- experiment with operator fusion and memory allocation in a DL system
- use as a research testbed for DL system components
- understand intermediate representations like NNVM

## When to choose
- you want a small, readable codebase to learn DL system internals
- you are teaching or taking a course on building deep learning systems
- you need a minimal experimental bed for DL system research

## When to avoid
- you need a production-ready deep learning framework
- you want extensive operator coverage or ecosystem support
- you need active maintenance and community support

## Facets
- artifact type: learning-resource
- maturity: maintenance
- function: deep-learning, machine-learning, compiler, interpreter
- domain: deep-learning, machine-learning, education, tutorials
- platform: cpp, lua
- tags: deep-learning-system, computational-graph, nnvm, educational, minimal-implementation, autodiff, gpu-computing, computational-graphs, linux, macos

## Member repositories
- tqchen/tinyflow (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:06.123877+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 2026-08-30T03:00:41.493725+00:00, confidence not recorded.
  - readme: https://github.com/tqchen/tinyflow (fetched 2026-08-28T04:06:06.123877+00:00, sha 7c87f4b23e77)
- Data as of 2026-08-30T08:39:29.467469+00:00.