spandrel
- Spandrel is a library for loading and running pre-trained PyTorch models. It automatically detects the model architecture and hyper parameters from model files, and provides a unified interface for running models.architectures
- The package containing the implementations of all supported architectures. Not necessary for most user code.ATD
- UndocumentedCompact
- UndocumentedCRAFT
- UndocumentedDAT
- UndocumentedDCTLSA
- UndocumentedDITN
- UndocumentedDnCNN
- UndocumentedDRCT
- UndocumentedDRUNet
- UndocumentedESRGAN
- No package docstring; 1/4 function, 0/1 class documentedFBCNN
- UndocumentedFFTformer
- UndocumentedGFPGAN
- UndocumentedGRL
- No package docstring; 0/1 constant, 1/4 function, 0/1 class documentedHAT
- No package docstring; 1/2 function, 0/1 class documentedIPT
- UndocumentedKBNet
- UndocumentedLaMa
- UndocumentedMixDehazeNet
- UndocumentedMMRealSR
- UndocumentedNAFNet
- UndocumentedOmniSR
- UndocumentedPLKSR
- UndocumentedRealCUGAN
- UndocumentedRestoreFormer
- UndocumentedRGT
- UndocumentedSAFMN
- UndocumentedSAFMNBCIE
- UndocumentedSCUNet
- UndocumentedSPAN
- UndocumentedSwiftSRGAN
- UndocumentedSwin2SR
- UndocumentedSwinIR
- UndocumentedUformer
- Undocumented__arch_helpers
- Undocumentedblock
- No module docstring; 0/1 type alias, 2/10 functions, 3/6 classes documenteddpir_basic_block
- No module docstring; 2/9 functions, 0/1 class documentedpadding
- No module docstring; 1/1 function documentedutil
- A module containing commonly-used functionality to implement architectures.timm
- No package docstring; 4/5 functions, 2/2 classes, 2/3 modules documented__drop
- DropBlock, DropPath__helpers
- Layer/Module Helpers Hacked together by / Copyright 2020 Ross Wightman__weight_init
- No module docstring; 1/4 function documented