Name: Michael J Clark
Type: User
Company: I'm just a guy who likes to machine learn
Bio: If you can't possibly remember what something is called, what it is, what it does, or why it does it, it instantly becomes a 'wassname'
Twitter: wassname
Location: Perth, Australia
Blog: wassname.org
Michael J Clark's Projects
Parallel processing with progress bars
PyTorch implementation of https://github.com/charlesq34/pointnet2
Python implementation of the PR-SSM.
Lesion and prostate masks for the PROSTATEx training dataset, after a lesion-by-lesion quality check.
Python Pushshift.io API Wrapper (for comment/submission search)
original code : https://github.com/hszhao/PSPNet
Projection creation, and conversion to misc formats
Read, modify and write DICOM files with python code
pyreadline repository
PyRoboLearn: a Python framework for Robot Learning
Python interface to ISLEX, an English IPA pronunciation dictionary with syllable and stress marking.
PyTorch implementation of Advantage Actor Critic (A2C), Proximal Policy Optimization (PPO) and Scalable trust-region method for deep reinforcement learning using Kronecker-factored approximation (ACKTR).
PyTorch for Numpy users.
The lightweight PyTorch wrapper for ML researchers. Scale your models. Write less boilerplate
PyTorch BERT model - for generating horror
PyTorch implementation of soft actor critic
Repository of Transformer based PyTorch Time Series Models
PyTorch based Probabilistic Time Series forecasting framework based on GluonTS backend
Random number generator of many statistical distributions.
deep learning for aerial/satellite imagery
An open source framework that provides a simple, universal API for building distributed applications. Ray is packaged with RLlib, a scalable reinforcement learning library, and Tune, a scalable hyperparameter tuning library.
Simple react webpack template
Framework for building complex recurrent neural networks with Keras
Migrate an old Reddit account to a new one.
Scrapes wildbow's twig, pact and worm into epubs with some extra ease-of-use features
keras implementation of [A simple neural network module for relational reasoning](https://arxiv.org/pdf/1706.01427.pdf)
Applying the RUDDER paper (arxiv.org/abs/1806.07857) to Sonic.
Ribbon V2 for Theta Vaults