- huf: haiku utilities and framework.
- spax: Sparse operations and utilities.
- grax: Graph neural network implementations.
- deqx: Deep Equilibrium operations and haiku modules.
- tfbm: High level decorator-based interface and CLI and tensorflow benchmarks.
- kblocks: Keras blocks with dependency injection and an efficient, dynamically configurable CLI.
- meta-model: A framework for simultaneously building data map functions and learned models for model-dependent data preprocessing pipelines.
- tfrng: Unified interface for different random number generation implementations and transforms for deterministic pipelines.
- tf_marching_cubes: peicewise-differentiable marching cubes implementations.
- tf_nearest_neighbour: brute-force kernels and
tf.py_funchacks for KDTree implementations.
- sdf_renderer: differentiable signed distance function rendering in tensorflow.
Data IO and cleaning is a necessary evil of almost all machine learning research. I maintain the following repositories for downloading and preprocessing publicly available datasets using tensorflow-datasets.