This repository is a curated list of resources for the Torch deep learning framework, targeting researchers and developers working with Torch. It provides a comprehensive collection of tutorials, model implementations, libraries, and community links, serving as a valuable starting point for anyone exploring or utilizing Torch for deep learning tasks.

How It Works

The list is organized into categories such as Tutorials, Model Zoo (categorized by network type like Recurrent and Convolutional), Libraries, and IDEs. It links to numerous research papers, code repositories, and demos, showcasing a wide array of applications and techniques within the Torch ecosystem. The "Model Zoo" section, in particular, highlights implementations of state-of-the-art models with references to their original papers.

Quick Start & Requirements

This is a curated list, not a runnable project. To use the linked resources, users will need to install and configure Torch (typically via Lua or Python bindings like lutorpy or pytorch). Specific requirements will vary per linked project, often including Python versions, CUDA, and specific datasets.

Highlighted Details

• Extensive coverage of Recurrent Neural Networks (RNNs), including LSTMs, GRUs, and applications in audio generation, language modeling, and reinforcement learning.
• A broad selection of Convolutional Neural Network (CNN) implementations, featuring models like DCGAN, Inception, VGG, ResNet, and SRCNN, with applications in image classification, style transfer, and super-resolution.
• Resources for Reinforcement Learning (RL) are well-represented, with links to Deep Q-Networks (DQN), Actor-Mimic, and environments like ViZDoom.
• Includes essential Torch libraries such as nn, optim, nngraph, and torchnet, alongside GPU support (cutorch, cudnn) and IDE integrations (iTorch).

Maintenance & Community

The list appears to be a community-driven effort, with many entries referencing academic papers and GitHub repositories. Links to Google Groups and Gitter Chat are provided for community interaction.

Licensing & Compatibility

The licensing of individual projects linked within this list will vary. Torch itself is typically distributed under permissive licenses (e.g., BSD), allowing for commercial use and integration into closed-source projects. However, users must verify the licenses of each specific linked resource.

Limitations & Caveats

This repository is a static list of links and does not provide a unified installation or execution environment. Many of the linked projects may be older, potentially targeting outdated versions of Torch or requiring significant effort to adapt to current deep learning frameworks. The Torch ecosystem has largely been superseded by PyTorch.