zafarali

learning-angular

A repo containing all code used to learn AngularJS, Scroll down to README for the tutorial.

emdp

Easy MDPs and grid worlds with accessible transition dynamics to do exact calculations

minerva-bot

Your personal assistant for everything McGill.

policy-gradient-methods

Modular PyTorch implementation of policy gradient methods

sentiment.datalogue

Sentiment analysis challenge for Datalogue recruiting

paper-notes

notes from machine learning, reinforcement learning and other papers

generative

Repository to explore Generative Models

explore-intelligence

Repository for all code and notes used to learn data analysis, visualisation, machine learning, artificial intelligence and probability and statistics

embedKB

Repository to benchmark vector representations of knowledge bases (Knowledge Base Embeddings)

gym

A toolkit for developing and comparing reinforcement learning algorithms.

keras

Deep Learning library for Python. Runs on TensorFlow, Theano or CNTK.

urbanistview

zafarali.github.io

Website

baselines

OpenAI Baselines: high-quality implementations of reinforcement learning algorithms

better-sampling

investigating sampling

compare_impute

Pipeline for comparing multiple imputation methods using a truth versus test set

deep-subsets

Learning deep functions on sets for set2subset transformations

disease-network-model

A Model To Simulate Diseases on a Network Structure

dl_ling_papers

Some summaries of papers in the confluence of Deep Learning and Linguistics / NLP

dopamine

Dopamine is a research framework for fast prototyping of reinforcement learning algorithms.

jax

Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

konstable

Visualize loss functions for multiple keras models

mjrl

Reinforcement learning algorithms for MuJoCo tasks

mlresearchkit

Tools to speed up research in machine learning

neural-combinatorial-rl-pytorch

PyTorch implementation of Neural Combinatorial Optimization with Reinforcement Learning https://arxiv.org/abs/1611.09940

pytorch-a2c-ppo-acktr

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).

redirectify

Browser extension to redirect pages based on rules. Intended to stop the browser following links straight to the PDF of a paper, but instead to go first to the HTML index page for the paper.

test-tube

Python library to easily log experiments and parallelize hyperparameter search for neural networks

tumorheterogeneity

A more powerful analysis of tumor heterogeneity

v97

Proceedings of ICML 2019