call_for_contributions.md

Call for code example contributions

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

Timeseries forecasting with LSTM for weather prediction

Recommendations:

• Use timeseries_dataset_from_array for data processing
• Single LSTM layer + classification layer, trained with fit()
• Add nice visualizations

Transformer model for MIDI music generation

Reference TF/Keras implementation

StyleGAN / StyleGAN2

Reference paper 1 2 Reference TF implementation

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

3D image classification

Using a dataset of CT scans (a few are available on Kaggle).

The model should use Conv3D layers.