Commit Graph

  • 450b6cafc4 [SD] Add weight emphasis to prompts encoder (#1276) 20230404.659 20230404.658 m68k-fr 2023-04-04 18:47:04 +02:00
  • 237d26baa2 update model db to reflect changes (#1277) Daniel Garvey 2023-04-04 11:46:55 -05:00
  • 67d6ee1104 remove 1/1 tqdm progress bar (#1274) 20230403.657 Daniel Garvey 2023-04-03 22:30:09 -05:00
  • 98b069488e Add tank_version.json (#1272) 20230403.656 Ean Garvey 2023-04-03 20:36:23 -05:00
  • e0f227643a Fix webui circular import issue (#1271) jinchen62 2023-04-03 16:00:10 -07:00
  • a0af3bb0cb xload and unload models (#1242) 20230403.655 jinchen62 2023-04-03 14:42:18 -07:00
  • 2cd61a5b96 strip source map (#1270) powderluv 2023-04-03 14:41:32 -07:00
  • f49d41a807 [SD] Add Stable diffusion text2image rest API (#1265) Gaurav Shukla 2023-04-04 00:32:24 +05:30
  • 2191fc8952 Separate pytest benchmark modes and fix model updates for SHARK downloader / pytest. (#1264) 20230403.654 Ean Garvey 2023-04-03 10:24:21 -05:00
  • aea7796e60 add gradio client to spec 20230403.653 PhaneeshB 2023-04-03 18:36:54 +05:30
  • a376619f1e [SD] Improve vmfb caching algo and retry mechanism (#1248) 20230402.652 20230401.651 20230331.650 Abhishek Varma 2023-03-31 22:08:14 +05:30
  • 02d52bb626 Add Intel ARC A770 target triple (#1263) 20230330.649 20230329.648 powderluv 2023-03-29 14:49:05 -07:00
  • 3b63645f79 [SD] Fix custom model path for WebUI (#1260) Abhishek Varma 2023-03-29 22:18:11 +05:30
  • d6f740b998 allow pytest to retry getting model artifacts + disable autotuning for pytorch benchmarks (#1257) 20230328.647 Ean Garvey 2023-03-28 23:38:45 -05:00
  • 594c6b8ea2 fix ckpt dir (#1258) 20230328.646 Daniel Garvey 2023-03-28 16:31:01 -05:00
  • 96b1560da5 Make batch size configurable via pytest and fix sharktank generation. (#1227) Ean Garvey 2023-03-28 14:33:39 -05:00
  • 0ef6a0e234 [SD] Fix Stencil scribble crash by updating image resize (#1255) 20230328.643 Abhishek Varma 2023-03-28 22:43:11 +05:30
  • 641d535f44 [SD] Fix device path issue for cpu (#1256) Gaurav Shukla 2023-03-28 22:39:49 +05:30
  • 5bb7846227 single entry point exe for all cli apps (#1158) Daniel Garvey 2023-03-28 11:15:21 -05:00
  • 8f84258fb8 Fix check for use_tuned conditions (#1252) 20230327.642 yzhang93 2023-03-27 11:21:25 -07:00
  • 7619e76bbd Disable and xfail some models that fail validation/compilation. (#1251) Ean Garvey 2023-03-27 12:42:53 -05:00
  • 9267eadbfa disable openjourney gen for nightly (#1249) Daniel Garvey 2023-03-27 11:55:34 -05:00
  • 431132b8ee Fix img2img mode switch (#1247) 20230327.641 Phaneesh Barwaria 2023-03-27 19:31:22 +05:30
  • fb35e13e7a fix Python version detection bug (#1246) cstueckrath 2023-03-27 16:00:40 +02:00
  • 17a67897d1 Add SD v2.1 768x768 tuned model (#1244) 20230326.640 20230325.639 20230324.638 20230324.637 yzhang93 2023-03-24 10:39:15 -07:00
  • da449b73aa [SD] Disable lora training tab for now (#1241) Gaurav Shukla 2023-03-24 21:46:24 +05:30
  • 0b0526699a Fix incorrect device argument initialization for LoRA training by extracting the device type and number and formatting it for pytorch (#1237) Kyle Herndon 2023-03-24 01:10:50 -07:00
  • 6c5705cf09 fixed a bug where lm would crash with certain smaller models AMD-Shark-Intel Elias Joseph 2023-03-23 07:36:04 +00:00
  • 4fac46f7bb In models testing fix paths to be relative to the script dir not cwd (#1128) 20230323.636 Boian Petkantchin 2023-03-22 13:26:52 -07:00
  • 49925950f1 fix false positives (#1193) Daniel Garvey 2023-03-22 15:25:39 -05:00
  • 2f891a6c23 fixed issue where a file wouldn't be downloaded when running full bloom Elias Joseph 2023-03-22 19:21:22 +00:00
  • 807947c0c8 Remove deprecated cli option iree-hal-cuda-disable-loop-nounroll-wa (#1235) Thomas 2023-03-22 10:05:15 -07:00
  • 0f3b62032b Only generate artifacts OTF if requirements are met. 20230322.631 Ean Garvey 2023-03-21 19:54:12 +00:00
  • a0a5f8b1d0 Merge branch 'main' into ean-gen-sharktank 20230322.630 Ean Garvey 2023-03-22 11:26:03 -05:00
  • 9d63365fe4 Only generate artifacts OTF if requirements are met. Ean Garvey 2023-03-21 19:54:12 +00:00
  • 593428bda4 [SD] Fix for transformers/__init__.py issue in PyInstaller (#1233) Abhishek Varma 2023-03-22 21:13:53 +05:30
  • cede9b4fec [SD] Fix custom_vae as a required parameter in inpaint (#1232) Abhishek Varma 2023-03-22 17:00:17 +05:30
  • c2360303f0 Add the int8 quantized model. Prashant Kumar 2023-03-21 10:37:34 +00:00
  • 420366c1b8 Move schedulers to global obj (#1225) jinchen62 2023-03-21 22:40:43 -07:00
  • d31bae488c Set iree-input-type to tm_tensor for SD (#1228) 20230321.628 20230321.627 Ean Garvey 2023-03-21 21:07:31 -05:00
  • c23fcf3748 Fix incorrect device argument initialization for LoRA training (#1231) Kyle Herndon 2023-03-21 19:07:18 -07:00
  • 3ac36e0cc8 Temporarily remove torch T5 models to enable tank generation 20230321.626 Ean Garvey 2023-03-21 16:37:34 -05:00
  • 2533f751a2 Only generate artifacts OTF if requirements are met. Ean Garvey 2023-03-21 19:54:12 +00:00
  • 57e99885e1 Set missing arg in SD tank generation Ean Garvey 2023-03-21 14:12:42 -05:00
  • 7dbbb1726a Fix SD obj not defined if fail to get models from pretrained (#1222) 20230321.625 jinchen62 2023-03-21 07:55:17 -07:00
  • 8b8cc7fd33 [SD] Update LoRA inference to handle various checkpoints (#1215) Abhishek Varma 2023-03-21 19:22:20 +05:30
  • e3c96a2b9d Move sentencepiece to importer requirements. (#1218) 20230320.624 Ean Garvey 2023-03-21 00:39:57 -05:00
  • 5e3f50647d Set --vulkan_large_heap_block_size default to 2gb. (#1220) Ean Garvey 2023-03-20 23:07:09 -05:00
  • 7899e1803a Add fix for attention slicing fp16 (#1217) gpetters94 2023-03-20 22:11:29 -04:00
  • d105246b9c Fix t5 models mariecwhite 2023-03-20 16:14:53 -07:00
  • c72a8365b1 Update requirements-importer.txt 20230320.621 Ean Garvey 2023-03-20 17:43:18 -05:00
  • 46d0842459 Update requirements.txt Ean Garvey 2023-03-20 17:42:07 -05:00
  • 90c958bca2 Add T5-base and T5-large Torch and TF Models (#1116) mariecwhite 2023-03-21 09:32:50 +11:00
  • f99903e023 Add EfficientNet B0 and B7 Torch and TF models mariecwhite 2023-02-20 04:45:23 -08:00
  • c6f44ef1b3 Add EfficientNet B0 and B7 Torch and TF models mariecwhite 2023-02-20 04:45:23 -08:00
  • 8dcd4d5aeb Make batch size configurable mariecwhite 2023-02-27 16:51:45 -08:00
  • d319f4684e Add peak memory reporting for IREE, TF and PyTorch (#1216) Phoenix Meadowlark 2023-03-20 13:40:49 -07:00
  • f1fb363403 fixed errors that occur when running SHARK on an older version of iree Elias Joseph 2023-03-20 20:14:00 +00:00
  • 54d7b6d83e Generate model artifacts in pytests if they don't exist in the cloud. (#1121) Ean Garvey 2023-03-20 12:13:19 -05:00
  • 4a622532e5 [Web] Stop images (#1212) 20230319.619 20230319.618 20230319.617 m68k-fr 2023-03-19 22:37:30 +01:00
  • 650b2ada58 add pytorch_lightning to requirements (#1211) 20230319.616 20230319.615 cstueckrath 2023-03-19 20:29:54 +01:00
  • f87f8949f3 [Web] CSS fix for gradio V3.22.1 (#1210) 20230319.614 m68k-fr 2023-03-19 14:13:59 +01:00
  • 7dc9bf8148 [Web] Move "stop Batch" button to "Advanced Options" toggle (#1209) m68k-fr 2023-03-19 04:54:42 +01:00
  • ba48ff8d25 Implement LoRA training and UI for training and UI for inference in img2img, inpaint, outpaint (#1200) 20230317.609 Kyle Herndon 2023-03-17 12:54:56 -07:00
  • 638840925c [SD] Add support for larger size upscaling (#1204) Gaurav Shukla 2023-03-17 22:50:48 +05:30
  • b661656c03 [Web] Fix custom model path for upscaler (#1199) 20230316.607 20230316.606 m68k-fr 2023-03-16 23:57:23 +01:00
  • 0225434389 [SD] Add sendTo Upscaler Gaurav Shukla 2023-03-16 19:09:29 +05:30
  • 7ffe20b1c2 [SD] Release memory used by upscaler when not in use Gaurav Shukla 2023-03-16 18:52:23 +05:30
  • d8f0c4655d [SD] Add Upscaler web Gaurav Shukla 2023-03-15 16:11:48 +05:30
  • 7e8d3ec0df [SD] Add upscalar pipeline Gaurav Shukla 2023-03-10 22:52:10 +05:30
  • 9c08eec565 Clear memory cache when switching model and mode (#1194) 20230315.605 jinchen62 2023-03-15 22:18:26 -07:00
  • 2d2c523ac5 [Web] Upgrade Gradio to v3.21.0 (#1188) 20230315.604 20230315.603 20230315.602 m68k-fr 2023-03-15 18:14:49 +01:00
  • f17b3128c0 [SD] Add LoRA inference to SD pipeline (#1189) Abhishek Varma 2023-03-15 22:43:45 +05:30
  • 7c7e630099 [SD] Add fix for using latest diffusers + add scribble variant to Stencil (#1191) Abhishek Varma 2023-03-15 22:43:20 +05:30
  • 2dd1491ec1 [Web] Add clear queue button (#1192) m68k-fr 2023-03-15 18:12:59 +01:00
  • 236357fb61 add missing import for shark_sd.spec (#1190) Daniel Garvey 2023-03-15 09:23:29 -05:00
  • 7bc38719de Add benchmark artifacts to .gitignore (#1186) Phoenix Meadowlark 2023-03-14 15:19:06 -07:00
  • bdbe992769 Add IREE_SAVE_TEMPS for import_debug command (#1184) Daniel Garvey 2023-03-14 13:40:23 -05:00
  • e6b925e012 [SD] Add Openpose to Stencil + image size issue fix (#1181) Abhishek Varma 2023-03-14 23:00:52 +05:30
  • 771120b76c workaround Gradio issue (#1183) 20230314.599 cstueckrath 2023-03-14 09:27:24 +01:00
  • a8ce7680db Add flag to augment the device allocator (#1182) 20230313.598 Boian Petkantchin 2023-03-13 15:49:26 -07:00
  • b6dcf2401b Stencil perf improvement (#1179) 20230313.597 Phaneesh Barwaria 2023-03-14 03:07:38 +05:30
  • 62b5a9fd49 generate sharktank for apps dir (#966) Daniel Garvey 2023-03-13 12:54:15 -05:00
  • 2f133e9d5c Fix png metadata (#1178) m68k-fr 2023-03-13 06:43:39 +01:00
  • f898a1d332 Update README.md powderluv 2023-03-12 16:54:42 -07:00
  • b94266d2b9 [Web] Randomize seed to -1 (#1176) m68k-fr 2023-03-12 20:42:31 +01:00
  • 1b08242aaa [Web] Improve dropdowns ux (#1175) m68k-fr 2023-03-12 20:41:51 +01:00
  • 691030fbab [SD] Improve Stencil feature to handle general image sizes 20230311.595 Abhishek Varma 2023-03-11 15:04:01 +00:00
  • 16ad7d57a3 [WebUi] txt2img_ui: Import png metadata (#1147) 20230310.594 20230310.593 m68k-fr 2023-03-11 01:26:34 +01:00
  • c561ebf43c Drop the torch-mlir pin Anush Elangovan 2023-03-10 15:39:04 -08:00
  • 97fdff7f19 Add instructions how to run the LLaMA model. (#1168) Prashant Kumar 2023-03-11 02:06:37 +05:30
  • ce6d82eab2 Fix bloom lint 20230310.592 Anush Elangovan 2023-03-10 11:52:40 -08:00
  • b8f4b18951 [SD] Use dynamic stencil HF repo id Abhishek Varma 2023-03-10 11:58:56 +00:00
  • b23d3aa584 added more memory efficient method to run large bloom models with sharded blooms (#1165) Eliasj42 2023-03-10 09:32:56 -08:00
  • 495670d9b6 Fix SD fine tuning script device arg usage Vivek Khandelwal 2023-03-10 12:45:26 +00:00
  • 815e23a0b8 Update iree-compile flags --iree-llvm-xxx -> --iree-llvmcpu-xxx (#1164) 20230310.591 20230309.590 Boian Petkantchin 2023-03-09 11:31:50 -08:00
  • 783538fe11 Move linting opts from github workflow to config files Boian Petkantchin 2023-03-01 11:11:41 -08:00
  • 996c645f6a In SD don't include device path in vmfb filename Boian Petkantchin 2023-03-02 07:51:41 -08:00
  • 1f7d249a62 Use utf-8 format for imgs_details.csv m68k-fr 2023-03-08 18:21:42 +01:00
  • de8bb8c6a2 single entry point exe for all cli apps 20230308.587 dan 2023-03-07 11:47:43 -06:00