aboutsummaryrefslogtreecommitdiffstats
path: root/modules/esrgan_model.py
Commit message (Collapse)AuthorAgeFilesLines
* Revert "MPS Upscalers Fix"brkirch2022-11-171-1/+1
| | | | This reverts commit 768b95394a8500da639b947508f78296524f1836.
* add resrgan 8x, allow use 1x and up to 8x extra models, move BSRGAN model, ↵victorca252022-10-301-4/+13
| | | | add nearest
* MPS Upscalers Fixbrkirch2022-10-251-1/+1
| | | | Get ESRGAN, SCUNet, and SwinIR working correctly on MPS by ensuring memory is contiguous for tensor views before sending to MPS device.
* update ESRGAN architecture and model to support all ESRGAN models in the DB, ↵victorca252022-10-091-62/+128
| | | | BSRGAN and real-ESRGAN models
* Removed duplicate definition model_pathMilly2022-10-091-2/+0
|
* Add hypernetwork support to split cross attention v1brkirch2022-10-081-1/+1
| | | | | * Add hypernetwork support to split_cross_attention_forward_v1 * Fix device check in esrgan_model.py to use devices.device_esrgan instead of shared.device
* Add --use-cpu command line optionbrkirch2022-10-041-5/+4
| | | | Remove MPS detection to use CPU for GFPGAN / CodeFormer and add a --use-cpu command line option.
* Update esrgan_model.pyd8ahazard2022-10-021-2/+2
| | | | Use alternate ESRGAN Model download path.
* remove unwanted formatting/functionality from the PRAUTOMATIC2022-09-301-61/+62
|
* Holy $hit.d8ahazard2022-09-291-117/+110
| | | | | | | | | | | | | | | | | Yep. Fix gfpgan_model_arch requirement(s). Add Upscaler base class, move from images. Add a lot of methods to Upscaler. Re-work all the child upscalers to be proper classes. Add BSRGAN scaler. Add ldsr_model_arch class, removing the dependency for another repo that just uses regular latent-diffusion stuff. Add one universal method that will always find and load new upscaler models without having to add new "setup_model" calls. Still need to add command line params, but that could probably be automated. Add a "self.scale" property to all Upscalers so the scalers themselves can do "things" in response to the requested upscaling size. Ensure LDSR doesn't get stuck in a longer loop of "upscale/downscale/upscale" as we try to reach the target upscale size. Add typehints for IDE sanity. PEP-8 improvements. Moar.
* Cleanup existing directories, fixesd8ahazard2022-09-261-2/+2
|
* Re-implement universal model loadingd8ahazard2022-09-261-15/+41
|
* Fix/Revert opts name from GAN to ESRGANd8ahazard2022-09-221-2/+2
|
* Add new models, fix shared opts issuesd8ahazard2022-09-221-2/+2
| | | | | | Add General x4x3, GeneralWDN x4x3, and AnimeVideo models from newer ESRGAN releases. Fix issues caused by renaming ESRGAN_tille values to GAN_tile without using an IDE...
* Revert "fix for swininr PR breaking ESRGAN for new users"AUTOMATIC2022-09-211-2/+2
| | | | This reverts commit 8d1d64f9a238272d7a166d21e25cf529c09b2ad5.
* fix for swininr PR breaking ESRGAN for new usersAUTOMATIC2022-09-211-2/+2
|
* Modular device managementAbdullah Barhoum2022-09-111-1/+2
|
* add a helpful message when user puts RealESRGAN model into ESRGAN directory.AUTOMATIC2022-09-081-5/+8
|
* Added support for launching on Apple Siliconxeonvs2022-09-071-2/+5
|
* ESRGAN supportAUTOMATIC2022-09-041-0/+134
generated by cgit v1.2.3 (git 2.25.1) at 2025-04-29 22:59:06 +0000