Stars
A python package for predicting group-level fMRI responses to visual stimuli using deep neural networks
[NeurIPS 2023] ImageReward: Learning and Evaluating Human Preferences for Text-to-image Generation
ViT models pretrained with up to ~5k hours of human-like video data
This is an unofficial implementation of the diffusion-style noise frontend in "Intriguing properties of generative classifiers" by Priyank Jaini, Kevin Clark, Robert Geirhos to improve the shape-bi…
A library for experimenting with, training and evaluating neural networks, with a focus on adversarial robustness.
Pytorch implementation of MixNMatch
Use lmdb to speed up imagenet dataset
Build stimuli for vision science, as numpy arrays
Adversarial Robustness Toolbox (ART) - Python Library for Machine Learning Security - Evasion, Poisoning, Extraction, Inference - Red and Blue Teams
A playbook for systematically maximizing the performance of deep learning models.
Python library for Representational Similarity Analysis
Tools for modeling image systems engineering in the human visual system front end
A library for deep learning with Spiking Neural Networks (SNN).
AEPsych is a tool for adaptive experimentation in psychophysics and perception research, built on top of gpytorch and botorch.
Visualize/test models for visual representation by synthesizing images.
image pyramid code in python 3
Matplotlib style sheets to nicely format figures for scientific papers, thesis and presentations while keeping them fully editable in Adobe Illustrator.
The home of MIT's 9.66 course (Fall 2021).
Vyākarana: A Colorless Green Benchmark for Syntactic Evaluation in Indic Languages
A general and accurate MACs / FLOPs profiler for PyTorch models
DRIFT is a tool for Diachronic Analysis of Scientific Literature.
Deep learning software to decode EEG, ECG or MEG signals
Benchmark your model on out-of-distribution datasets with carefully collected human comparison data (NeurIPS 2021 Oral)
Code for SkipNet: Learning Dynamic Routing in Convolutional Networks (ECCV 2018)
Modular Object-Oriented Games (MOOG): Python-based game engine for reinforcement learning, psychology, and neurophysiology.