Chainer Models

This repository contains a number of models implemented in Chainer.

Contributing guidelines

If you have created a model, please send us a pull request. For those just getting started with pull requests, GitHub has a howto.

We have a list of candidate papers to implement: https://github.com/chainer/models/projects/1

Models

• Averaging Weights Leads to Wider Optima and Better Generalization [code] [paper]
• Snapshot Ensembles: Train 1, get M for free [paper] [code]
• Compressing Word Embeddings via Deep Compositional Code Learning [paper] [code]
• Simple Does It: Weakly Supervised Instance and Semantic Segmentation [paper] [code]
• Mixture Density Networks [article] [code]
• GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks [paper] [code]
• Improving Language Understanding by Generative Pre-Training [article] [code]
• BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding [paper] [code]
• Deep contextualized word representations [paper] [code]
• Adversarial Training Methods for Semi-Supervised Text Classification [paper] [code]
• Multi-label image classification [code]
• Real-Time Seamless Single Shot 6D Object Pose Prediction [paper] [code]
• Neural Relational Inference for Interacting Systems [paper] [code]
• SiamRPN and SiamMask [paper] [code]
• Learning to learn by gradient descent by gradient descent [paper] [code]
• Attention is all you need [paper] [code]

License

MIT License (see LICENSE file).