Providers¶
Flash is a framework integrator. We rely on many open source frameworks for our tasks, visualizations and backbones. Here’s a list of some of the providers we use for backbones and heads within Flash (check them out and star their repos to spread the open source love!):
airctic/IceVision (https://github.com/airctic/icevision)
Facebook Research/dino (https://github.com/facebookresearch/dino)
Facebook Research/PyTorchVideo (https://github.com/facebookresearch/pytorchvideo)
Facebook Research/vissl (https://github.com/facebookresearch/vissl)
Hugging Face/transformers (https://github.com/huggingface/transformers)
Intelligent Systems Lab Org/Open3D-ML (https://github.com/isl-org/Open3D-ML)
jdb78/PyTorch-Forecasting (https://github.com/jdb78/pytorch-forecasting)
learnables/learn2learn (https://github.com/learnables/learn2learn)
manujosephv/PyTorch Tabular (https://github.com/manujosephv/pytorch_tabular)
OpenAI/CLIP (https://github.com/openai/CLIP)
OpenMMLab/MMDetection (https://github.com/open-mmlab/mmdetection)
PyG/PyTorch Geometric (https://github.com/pyg-team/pytorch_geometric)
pystiche/pystiche (https://github.com/pystiche/pystiche)
PyTorch/fairseq (https://github.com/pytorch/fairseq)
PyTorch/torchvision (https://github.com/pytorch/vision)
qubvel/segmentation_models.pytorch (https://github.com/qubvel/segmentation_models.pytorch)
rwightman/efficientdet-pytorch (https://github.com/rwightman/efficientdet-pytorch)
rwightman/pytorch-image-models (https://github.com/rwightman/pytorch-image-models)
UKPLab/sentence-transformers (https://github.com/UKPLab/sentence-transformers)
Ultralytics/YOLOV5 (https://github.com/ultralytics/yolov5)
voxel51/fiftyone (https://github.com/voxel51/fiftyone)
You can also read our guides for some of our larger integrations: