PyTorch

PassGPTjavirandor/passgpt

A GPT-2 model trained from scratch on password leaks for password modeling, generation, and strength estimation.

CyclOpsVectorInstitute/cyclops

A toolkit for evaluating and monitoring machine learning models in clinical healthcare settings.

GreatXEdisonLeeeee/GreatX

A PyTorch-based toolbox for graph reliability, focusing on adversarial attacks, defenses, and robustness techniques for graph neural networks.

Tutorials to get started on signal processing for machine learningjinglescode/python-signal-processing

A Python package and tutorial collection for signal processing and machine learning, built on NumPy and SciPy.

InjecGuardsafolab-wisc/injecguard

An open-source prompt guard model that detects prompt injection attacks while mitigating over-defense against benign inputs.

BulkFormerKangBoming/BulkFormer

A large-scale foundation model pretrained on over 500,000 human bulk RNA-seq profiles for biomedical transcriptome analysis.

Cogitarecogitare-ai/cogitare

A modern, fast, and modular deep learning and machine learning framework for Python built on PyTorch.

Learnergygugarosa/learnergy

A PyTorch-based Python library for energy-based machine learning models, including Restricted Boltzmann Machines and Deep Belief Networks.

MIDASlabomics/midas

A PyTorch deep generative model for integrating and imputing single-cell multimodal data with missing modalities.

DeepMushroomOlament/DeepMushroom

A fungal image classification project using ResNet to identify mushroom species from citizen science and expert sources.

dream-creatorProGamerGov/dream-creator

A PyTorch-based tool for training custom DeepDream models using GoogleNet and custom image datasets.

MAPSmahmoodlab/MAPS

A machine learning approach for rapid, pathologist-level cell type annotation from spatial proteomics data like MIBI and CODEX.

open-solution-ship-detectionneptune-ml/open-solution-ship-detection

An open-source solution for the Airbus Ship Detection Challenge, providing a benchmark and base for ship detection in satellite imagery.

typilustypilus/typilus

A deep learning algorithm that predicts type hints for Python code using neural networks.

xrays-and-gradcampriyavrat-misra/xrays-and-gradcam

A PyTorch-based project for classifying chest X-rays and localizing pathologies using Grad-CAM with fine-tuned CNN models.

GPimziatdinovmax/GPim

A Python package for applying Gaussian processes and Bayesian optimization to images and hyperspectral data using Pyro and Gpytorch.

max-toxic-comment-classifierIBM/MAX-Toxic-Comment-Classifier

A BERT-based model that detects six types of toxicity in text comments, deployable as a Docker container.

AirPoserobot-perception-group/AirPose

A multi-view fusion network for aerial 3D human pose and shape estimation using autonomous UAVs with on-board RGB cameras.

omarchy-aimitkox/omarchy-ai

A one-command installer that transforms a fresh Arch Linux installation into a complete, local-first AI development environment with Hyprland.

UnitedNetLiuLab-Bioelectronics-Harvard/UnitedNet

An interpretable multi-task deep neural network for single-cell multi-omics integration and cross-modal analysis.

QuickVisionQuick-AI/quickvision

An easy-to-use PyTorch library for faster computer vision model development and training.

PyTorchCVdonnyyou/PyTorchCV

A PyTorch-based framework providing implementations of state-of-the-art deep learning models for computer vision tasks.

max-fast-neural-style-transferIBM/MAX-Fast-Neural-Style-Transfer

Deploy a neural network model that transfers artistic styles from one image to another using a ResNet-based architecture.

Catalyst-RLcatalyst-team/catalyst-rl

A PyTorch framework for reinforcement learning research, focused on reproducibility and fast experimentation.

open-solution-googleai-object-detectionneptune-ml/open-solution-googleai-object-detection

An open-source solution for the Google AI Open Images Object Detection Challenge, providing a RetinaNet-based benchmark with experiment tracking.

max-scene-classifierIBM/MAX-Scene-Classifier

A pre-trained image classifier that recognizes 365 different scene and location types using a ResNet18 model fine-tuned on Places365.

CryptoKnightAbertayMachineLearningGroup/CryptoKnight

A framework for automated cryptographic primitive classification using dynamic binary instrumentation and deep learning.

Notebooksrlan/notebooks

A Docker-based starter kit providing pre-configured Jupyter notebook environments for machine learning with TensorFlow, PyTorch, and essential libraries.

Multigratetheislab/multigrate

A Python library for integrating multiomic single-cell data using product-of-experts variational autoencoders.

Classposesohmandal/classpose

A foundation model-driven method for semantic cell classification in whole slide images, extending Cellpose with a WSI workflow and QuPath integration.

CellshapeSentinal4D/cellshape

A Python package for 3D single-cell shape analysis using geometric deep learning on point clouds and voxels.

YeaZlpbsscientist/YeaZ-GUI

A graphical interface for segmenting yeast cells in microscopy images using convolutional neural networks.

DeepSeek-R1-Flax-1.5B-DistillJ-Rosser-UK/Torch2Jax-DeepSeek-R1-Distill-Qwen-1.5B

Flax (JAX) and PyTorch implementations of the DeepSeek-R1-Distill-Qwen-1.5B language model with weight conversion utilities.

TGSAviolet-sto/TGSA

PyTorch implementation of twin graph neural networks with similarity augmentation for drug response prediction using protein-protein associations.

SyMBacgeorgeoshardo/SyMBac

Generates synthetic phase contrast or fluorescence microscope images of bacteria for training deep learning segmentation models.

RoosterizeEngineeringSoftware/roosterize

A neural network-based tool that suggests lemma names for Coq verification projects by analyzing serialized statements and elaborated terms.