karpathy/char-rnn
Multi-layer recurrent neural network for character-level language modeling and text generation in Torch.
Not currently ranked — collecting fresh signals.
star history
This repository implements multi-layer RNNs including LSTM and GRU variants for character-level language modeling in Lua/Torch. The model trains on a text file to predict the next character in a sequence, then generates new text by sampling character-by-character that resembles the training corpus. It supports GPU acceleration, mini-batches, and model checkpointing.
Frequently asked
- What is karpathy/char-rnn?
- Multi-layer recurrent neural network for character-level language modeling and text generation in Torch.
- Is char-rnn open source?
- Yes — karpathy/char-rnn is an open-source project tracked on heatdrop.
- What language is char-rnn written in?
- karpathy/char-rnn is primarily written in Lua.
- How popular is char-rnn?
- karpathy/char-rnn has 12.1k stars on GitHub.
- Where can I find char-rnn?
- karpathy/char-rnn is on GitHub at https://github.com/karpathy/char-rnn.