Skip to content

Latest commit

 

History

History
67 lines (32 loc) · 2 KB

call_for_contributions.md

File metadata and controls

67 lines (32 loc) · 2 KB
Raw

Call for code example contributions

This is a constantly-updated list of code examples that we're currently interested in.

Transformer model for MIDI music generation

Reference TF/Keras implementation

StyleGAN / StyleGAN2

Reference paper 1 2 Reference TF implementation

  • Currently being worked on by @parthivc and @Korusuke as of 07/14/20. See issue #144

Improve next-frame prediction with ConvLSTM

Current example

Recommendations:

  • Find a better, more interesting dataset
  • Make it work better
  • Add detailed explanations
  • As of 6/23/20, this is being worked on. See issue #112

Text-to-speech

Example TF2/Keras implementation

Speech recognition

Example TF2/Keras implementation

Learning to rank

Reference Kaggle competition

Large-scale multi-label text classification

Using word bi-grams + TF-IDF + a small MLP, from raw text strings. The tokenization and extraction of TF-IDF ngrams should be done with the TextVectorization layer.

The dataset should have at least 50k samples and there should be at least a dozen of labels.

DETR: End-to-End Object Detection with Transformers

Reference implementation TF/Keras implementation

Image similarity search engine using triplet loss

A convnet trained with a triplet loss to estimate image similarity. It could be trained on something like this dataset (other datasets are also possible).