profile
viewpoint

karpathy/char-rnn 9986

Multi-layer Recurrent Neural Networks (LSTM, GRU, RNN) for character-level language models in Torch

karpathy/convnetjs 9897

Deep Learning in Javascript. Train Convolutional Neural Networks (or ordinary ones) in your browser.

karpathy/neuraltalk 5085

NeuralTalk is a Python+numpy project for learning Multimodal Recurrent Neural Networks that describe images with sentences.

karpathy/neuraltalk2 4802

Efficient Image Captioning code in Torch, runs on GPU

karpathy/arxiv-sanity-preserver 3527

Web interface for browsing, search and filtering recent arxiv submissions

jcjohnson/densecap 1353

Dense image captioning in Torch

karpathy/reinforcejs 1012

Reinforcement Learning Agents in Javascript (Dynamic Programming, Temporal Difference, Deep Q-Learning, Stochastic/Deterministic Policy Gradients)

karpathy/recurrentjs 852

Deep Recurrent Neural Networks and LSTMs in Javascript. More generally also arbitrary expression graphs with automatic differentiation.

karpathy/paper-notes 616

Random notes on papers, likely a short-term repo.

karpathy/pytorch-normalizing-flows 412

Normalizing flows in PyTorch. Current intended use is education not production.

issue commentkarpathy/arxiv-sanity-preserver

"Could not render: incorrectly formatted pdf" error on every recent paper

The issue was with thumbnail generation, convert was failing with an error. Had to adjust system configuration files, no code changes. Still slightly mysterious why it happened rather suddenly without any (manual) system update or anything.

jasonjmcghee

comment created time in 9 days

push eventkarpathy/pytorch-normalizing-flows

Andrej Karpathy

commit sha fce68b6ccfd1580e0984246ef86e67eb7ce5b7e1

fix a bug where we would init with uninitialized memory

view details

Andrej Karpathy

commit sha dbd4b5a1c52d40b6b3ba1687cc8ca7d1efa943eb

add a positional encoding MLP to the mix, helps with low-d curled up data potentially

view details

Andrej Karpathy

commit sha b60e119b37be10ce2930ef9fa17e58686aaf2b3d

fix bug where net_class was not correctly used and propagated

view details

push time in 2 months

push eventkarpathy/pytorch-normalizing-flows

Andrej Karpathy

commit sha ada7ef5aaa0dddfc43db41a8ab99d0d49efad68e

small updates to Readme

view details

push time in 2 months

push eventkarpathy/pytorch-normalizing-flows

Andrej Karpathy

commit sha 4f1fa276ade199f3f503bf24d9e7141d2955cd2b

include also and play with neural splines too

view details

push time in 2 months

push eventkarpathy/pytorch-normalizing-flows

Andrej Karpathy

commit sha cae0006f4a74c295a660b0bf6bd15c4443a42664

refactor Invertible1x1Conv: remove scipy dependence, clean up

view details

push time in 2 months

push eventkarpathy/pytorch-normalizing-flows

Andrej Karpathy

commit sha 04fadeaf6fde661c79b3d2b56013134bf2eda9d7

readme formatting cosmetics

view details

push time in 2 months

push eventkarpathy/pytorch-normalizing-flows

Andrej Karpathy

commit sha ac30aa077b74ad1da2c08b68798c9e1747cbb2bd

sequester code into files, keep notebooks a bit thiner and focused on visualizations instead of core functionality

view details

push time in 2 months

PublicEvent
more