# denizyuret/Knet.jl

Koç University deep learning framework.

Repository: https://github.com/denizyuret/Knet.jl
Canonical: https://ross.abutalabs.com/products/knetjl
Homepage: https://denizyuret.github.io/Knet.jl/latest
Language: Jupyter Notebook
License: NOASSERTION
License Family: other
Topics: knet, deep-learning, julia, machine-learning, neural-networks, data-science
Last push: 2024-11-15T16:59:30+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": 3991, "days_push": 656, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1432, forks 223 (observed 2026-08-28T04:04:43.181541+00:00)

## What it is
Knet (Koç University deep learning framework) is a deep learning library implemented in Julia with GPU support and automatic differentiation via dynamic computational graphs. Models are defined in plain Julia code without a separate modeling language.

## Use cases
- train neural networks in Julia
- build deep learning models with GPU acceleration
- define models with automatic differentiation in plain Julia
- learn deep learning with a minimal framework
- compare deep learning framework performance benchmarks
- implement CNNs and RNNs in Julia

## When to choose
- you are already working in the Julia ecosystem
- you want models defined in plain Julia with dynamic graphs
- you need GPU-backed training with automatic differentiation
- you want a lightweight educational deep learning framework

## When to avoid
- you need a large ecosystem of pretrained models and tooling
- you prefer Python-based frameworks like PyTorch or TensorFlow
- you need production-grade support and frequent updates
- you want a framework with a large active community

## Facets
- artifact type: library
- maturity: maintenance
- function: deep-learning, machine-learning, gpu-computing
- domain: deep-learning, machine-learning, data-science
- platform: windows
- tags: automatic-differentiation, neural-networks, julia, dynamic-computational-graphs, gpu, linux, macos

## Member repositories
- denizyuret/Knet.jl (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:43.181541+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 2026-08-30T04:37:01.286220+00:00, confidence not recorded.
  - readme: https://github.com/denizyuret/Knet.jl (fetched 2026-08-28T04:04:43.181541+00:00, sha 2aa0e4c0cb42)
  - homepage: https://denizyuret.github.io/Knet.jl/latest (fetched 2026-08-29T11:48:14.931871+00:00, sha d73d160ae4dc)
- Data as of 2026-08-30T08:39:29.467469+00:00.