A multi-backend implementation of the Keras API, with support for TensorFlow, JAX, and PyTorch.

Pax is a Jax-based machine learning framework for training large scale models. Pax allows for advanced and fully configurable experimentation and parallelization, and has demonstrated industry lead…

Backward compatible ML compute opset inspired by HLO/MHLO

Stores documents used by the TensorFlow developer community

Simulating the performance of various streaming algorithms. #experimentalmathematics

Class Activation Mapping

The pretrained models trained on Moments in Time Dataset

Temporal Relation Networks

🏖 Keras Implementation of Painting outside the box

A collection of infrastructure and tools for research in neural network interpretability.

Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization

Chainer implementation of "Perceptual Losses for Real-Time Style Transfer and Super-Resolution".

Style transfer, deep learning, feature transform

Code for "Image Generation from Scene Graphs", Johnson et al, CVPR 2018

ROLO is short for Recurrent YOLO, aimed at simultaneous object detection and tracking

Generative Handwriting using LSTM Mixture Density Network with TensorFlow