This repository is a PyTorch Lightning implementation of DeepMind's Skillful Nowcasting GAN (DGMR) for precipitation nowcasting. It provides a modular, well-structured library of neural network components including residual blocks (GBlock, DBlock, LBlock), conditioning stacks (ContextConditioningStack, LatentConditioningStack), a ConvGRU-based sampler, spatial and temporal discriminators, and multiple loss functions (hinge, grid cell, SSIM, etc.). The core model (DGMR) wraps these components into a PyTorch Lightning module with manual optimization for the GAN training loop. It supports HuggingFace Hub integration for pretrained weights and includes training scripts for the UK Nimrod radar dataset. The codebase is well-documented, type-annotated, and includes comprehensive unit tests for individual components, making it suitable both as a library for nowcasting research and as a reference implementation for complex GAN architectures.
This repository implements a two-stage generative image model: Representation Autoencoders (RAE) followed by Diffusion Transformers (DiT^DH). Stage 1 trains a ViT decoder to reconstruct images from frozen pretrained encoders (DINOv2, MAE, SigLIP), with optional GAN and LPIPS losses. Stage 2 trains a class-conditional diffusion transformer on the resulting latent space, supporting both classifier-free guidance and autoguidance. The codebase is config-driven (OmegaConf), PyTorch DDP, and provides training, distributed sampling, FID evaluation, and latent statistics computation. It is a research codebase with strong engineering quality, modular design (transport, discriminator, encoders, decoders), and reproducibility through pinned dependencies and checkpointing.
FlowEdit is a PyTorch-based implementation of the FlowEdit algorithm for inversion-free text-guided image editing using pre-trained flow models (Stable Diffusion 3 and FLUX). The system takes an input image and source/target prompts, and edits the image by solving an ODE that combines the source and target trajectories in the latent space. The core algorithm is implemented in FlowEdit_utils.py, with run_script.py orchestrating the experiment execution based on YAML configurations. The repository includes a dataset of images and prompts for numerical evaluation.
IgGM is a generative foundation model for antibody and nanobody design, implemented as a deep learning library with a command-line interface. The system performs tasks such as CDR loop design, framework region redesign (humanization), inverse design from complex structures, affinity maturation, and structure prediction. It uses a diffusion-based approach to co-design amino acid sequences and 3D backbone structures, leveraging an ESM2-based protein language model, Evoformer blocks with triangular attention, invariant point attention, and a structure module similar to AlphaFold2. The pipeline is orchestrated by the `AbDesigner` class and invoked via the `design.py` script, accepting PDB and FASTA inputs and generating output sequences and structures. The repository includes well-structured reusable components for rigid-body transformations (Rotation, Rigid), PDB parsing, coordinate conversion, and diffusion schedules, though some research artifacts (misspellings, commented code) and dependency pinning to older PyTorch/CUDA versions indicate moderate technical drift.