convnetjs

Source

• Source kind: github-repository
• URL: https://github.com/karpathy/convnetjs
• Discovery source: https://github.com/karpathy/convnetjs
• License: MIT
• Distribution policy: public-summary-plus-license-aware-excerpts
• Public mirror status: partial excerpt
• Content hash: da4102e137ca864d94e3b6562a0c567f7c3ba99618d6d83a06b8b2cb11b3f364
• First seen: 2026-05-15
• Last changed: 2026-05-15

Classification

• Primary category: Neural network fundamentals
• Corpus source note: 2026-05-15-karpathy-public-corpus
• Project taxonomy: karpathy-project-taxonomy
• Idea map: karpathy-idea-map
• Topic hub: karpathy-public-work

Summary

ConvNetJS ConvNetJS is a Javascript implementation of Neural networks, together with nice browser-based demos. It currently supports: - Common Neural Network modules (fully connected layers, non-linearities) - Classification (SVM/Softmax) and Regression (L2) cost functions - Ability to specify and train Convolutional Networks that process images - An experimental Reinforcement Learning module, based on Deep Q Learning For much more information, see the main page at $1 Note : I am not actively maintaining ConvNetJS anymore because I simply don’t have time. I think the npm repo might not work at this point. Online…

What This Teaches

• How core neural network ideas can be rebuilt from first principles.
• Useful for grounding later LLM work in gradients, activations, optimization, and model internals.

Why It Matters

This is high-priority for Vipin because it supports durable first-principles understanding instead of shallow API use.

Repository Snapshot

• Full name: karpathy/convnetjs
• Default branch: master
• HEAD: 4c3358a315b4d71f31a0d532eb5d1700e9e592ee
• Stars at crawl: 11160
• Forks at crawl: 2075
• File count: 56
• README path: Readme.md
• License path: LICENSE
• Created: 2014-01-05T00:12:15Z
• Updated: 2026-05-12T17:12:54Z
• Pushed: 2023-01-07T21:33:23Z

Top-Level Structure

• demo: 24
• src: 15
• test: 9
• [root]: 3
• build: 3
• compile: 2

File Extension Profile

• .js: 35
• .html: 11
• .css: 3
• .jar: 1
• .json: 1
• .license: 1
• .md: 1
• .png: 1
• .xml: 1
• [none]: 1

Tags / Release-Like Markers

• No git tags found in the shallow local clone.

Sample File Tree

• bower.json
• build/deepqlearn.js
• build/util.js
• build/vis.js
• compile/build.xml
• compile/yuicompressor-2.4.8.jar
• demo/autoencoder.html
• demo/automatic.html
• demo/cifar10.html
• demo/classify2d.html
• demo/css/automatic.css
• demo/css/style.css
• demo/image_regression.html
• demo/js/autoencoder.js
• demo/js/automatic.js
• demo/js/classify2d.js
• demo/js/image_regression.js
• demo/js/image-helpers.js
• demo/js/images-demo.js
• demo/js/jquery-1.8.3.min.js
• demo/js/npgmain.js
• demo/js/pica.js
• demo/js/regression.js
• demo/js/rldemo.js
• demo/js/trainers.js
• demo/mnist.html
• demo/regression.html
• demo/rldemo.html
• demo/speedtest.html
• demo/trainers.html
• LICENSE
• Readme.md
• src/convnet_export.js
• src/convnet_init.js
• src/convnet_layers_dotproducts.js
• src/convnet_layers_dropout.js
• src/convnet_layers_input.js
• src/convnet_layers_loss.js
• src/convnet_layers_nonlinearities.js
• src/convnet_layers_normalization.js
• src/convnet_layers_pool.js
• src/convnet_magicnet.js
• src/convnet_net.js
• src/convnet_trainers.js
• src/convnet_util.js
• src/convnet_vol.js
• src/convnet_vol_util.js
• test/jasmine/lib/jasmine-2.0.0/boot.js
• test/jasmine/lib/jasmine-2.0.0/console.js
• test/jasmine/lib/jasmine-2.0.0/jasmine.css
• test/jasmine/lib/jasmine-2.0.0/jasmine.js
• test/jasmine/lib/jasmine-2.0.0/jasmine_favicon.png
• test/jasmine/lib/jasmine-2.0.0/jasmine-html.js
• test/jasmine/MIT.LICENSE
• test/jasmine/spec/NeuralNetSpec.js
• test/jasmine/SpecRunner.html

Public Handling Notes

• EXTRACTED: This page records public metadata and a source-grounded summary.
• INFERRED: Full local preservation, when available, is for private/local use unless a license or explicit source policy makes public redistribution safe.
• Do not treat this page as permission to republish unlicensed source text or code wholesale.