github/SHARK-Studio
1
1
Fork 0
mirror of https://github.com/nod-ai/SHARK-Studio.git synced 2026-04-20 03:00:34 -04:00
Code Issues Packages Projects Releases Wiki Activity
1,230 Commits 66 Branches 705 Tags
20230430.708
Commit Graph

493 Commits

Author SHA1 Message Date
Chi_Liu
aa8ada9da9 Add support for torch to stablehlo and tosa in shark_importer (#1360) 2023-04-27 08:09:45 -07:00
powderluv
09e3f63d5b Fix pascal (#1346) 
* Add fp32 for upscaler VAE

* Plumb Pascal vulkan support
 2023-04-23 20:28:25 -07:00
Ean Garvey
1afe07c296 Disable winograd on VAE with rdna2 and fix unet tuning. (#1313) 
* Disable winograd on VAE with rdna2 and fix unet tuning.

* Fix batch size 1 downloads and clear_all on windows.
 2023-04-18 15:55:10 -05:00
powderluv
1aa85222e9 Add AMD W7900 target triple (#1304) 
This maps to RDNA3
 2023-04-16 00:14:21 -07:00
Ean Garvey
f17915bedc Fix batch size appending to model name. (#1294) 
* Update shark_downloader.py

* Update shark_downloader.py
 2023-04-11 15:34:25 -05:00
Ean Garvey
98b069488e Add tank_version.json (#1272) 2023-04-03 18:36:23 -07:00
powderluv
2cd61a5b96 strip source map (#1270) 2023-04-03 14:41:32 -07:00
Ean Garvey
2191fc8952 Separate pytest benchmark modes and fix model updates for SHARK downloader / pytest. (#1264) 
* Only xfail windows models in CI

* downloader: make model updates more robust.

* Separate baseline and native benchmarks in pytest.

* Fix native benchmarks

* Fix torchvision model utils.
 2023-04-03 08:24:21 -07:00
powderluv
02d52bb626 Add Intel ARC A770 target triple (#1263) 
This just enables the plumbing. It generates black images.
 2023-03-29 14:49:05 -07:00
Ean Garvey
d6f740b998 allow pytest to retry getting model artifacts + disable autotuning for pytorch benchmarks (#1257) 
* Adds a few xfails to enable macOS builder

* Convert string batch sizes to ints where needed.

* allow pytest to retry getting model artifacts

* Reduce attempts and add assert msg.
 2023-03-28 23:38:45 -05:00
Ean Garvey
96b1560da5 Make batch size configurable via pytest and fix sharktank generation. (#1227) 
* Fix sharktank generation and add batch_size pytest option for torch.

* Disable torch dynamo until py3.11 supported

* Compile torchmodel without dynamo if torch.compile fails

* Use release versions of TF/Keras for importer.

* Pin torchvision and remove debug prints.

* Remove duplicates from torch model list.

* Update generate_sharktank.py

* xfail a few models that fail sharktank generation/ numerics
 2023-03-28 14:33:39 -05:00
Thomas
807947c0c8 Remove deprecated cli option iree-hal-cuda-disable-loop-nounroll-wa (#1235) 2023-03-22 12:05:15 -05:00
Ean Garvey
d31bae488c Set iree-input-type to tm_tensor for SD (#1228) 2023-03-21 19:07:31 -07:00
gpetters94
7899e1803a Add fix for attention slicing fp16 (#1217) 2023-03-20 19:11:29 -07:00
Phoenix Meadowlark
d319f4684e Add peak memory reporting for IREE, TF and PyTorch (#1216) 2023-03-20 15:40:49 -05:00
Ean Garvey
54d7b6d83e Generate model artifacts in pytests if they don't exist in the cloud. (#1121) 
* Add gen_shark_files fn to shark_downloader for OTF artifact generation

* add generate_sharktank as a tank/ python module.

* Fix some paths in tank generation.
 2023-03-20 12:13:19 -05:00
Daniel Garvey
bdbe992769 Add IREE_SAVE_TEMPS for import_debug command (#1184) 
based on hf_model_id. Works on windows
 2023-03-14 11:40:23 -07:00
Boian Petkantchin
a8ce7680db Add flag to augment the device allocator (#1182) 
Example:
$ python my_app.py --device_allocator caching debug
This will wrap the device allocator with first caching allocator then
debug allocator.

$ python my_app.py --device_allocator caching
Only wrap with caching allocator.

Co-authored-by: Boian Petkantchin <boian@nod-labs.com>
 2023-03-13 15:49:26 -07:00
Daniel Garvey
62b5a9fd49 generate sharktank for apps dir (#966) 
* merge confix resolution

* add support to other scripts

---------

Co-authored-by: dan <dan@nod-labs.com>
 2023-03-13 10:54:15 -07:00
powderluv
f898a1d332 Update README.md 2023-03-12 16:54:42 -07:00
Prashant Kumar
97fdff7f19 Add instructions how to run the LLaMA model. (#1168) 
* Add instructions how to run the LLaMA model.

* Update README.md
 2023-03-10 12:36:37 -08:00
Anush Elangovan
ce6d82eab2 Fix bloom lint 2023-03-10 11:53:08 -08:00
Eliasj42
b23d3aa584 added more memory efficient method to run large bloom models with sharded blooms (#1165) 
Co-authored-by: Elias Joseph <elias@nod-labs.com>
 2023-03-10 09:32:56 -08:00
Vivek Khandelwal
495670d9b6 Fix SD fine tuning script device arg usage 2023-03-10 18:37:53 +05:30
Boian Petkantchin
815e23a0b8 Update iree-compile flags --iree-llvm-xxx -> --iree-llvmcpu-xxx (#1164) 2023-03-09 11:31:50 -08:00
Eliasj42
93891984f3 made sharded bloom example more user friendly (#1153) 
Co-authored-by: Elias Joseph <elias@nod-labs.com>
 2023-03-06 10:23:48 -08:00
Vivek Khandelwal
cc0ef54e0e Fix Stable diffusion fine tuning script 2023-03-06 17:52:16 +05:30
Daniel Garvey
812152485d temporarily xfail tiny convnext macos (#1142) 2023-03-03 13:30:56 -06:00
Vivek Khandelwal
0816fb403a Add Stable diffusion fine tuning script 
This commit adds the sd fine tuning script which runs through the
torchdynamo path.
 2023-03-03 21:59:00 +05:30
mariecwhite
d567192fd3 Fix call to Torch Inductor 2023-02-28 00:35:57 -08:00
jinchen62
a01154a507 Add SD outpainting (#1072) 
python apps/stable_diffusion/scripts/outpaint.py --prompt="Face of a yellow cat, high resolution, sitting on a park bench" --img_path=test_imgs/overture-creations-5sI6fQgYIuo.png --import_mlir --hf_model_id="stabilityai/stable-diffusion-2-inpainting" --pixels=128 --mask_blur=8 --left --right --top --bottom --steps=20
 2023-02-22 23:16:05 -08:00
Eliasj42
5ff40a0d2d added an example to run sharded bloom (#1079) 
added ability to compile sharded mlir files from hugingface models

Co-authored-by: Elias Joseph <elias@nod-labs.com>
 2023-02-22 22:48:58 -08:00
Daniel Garvey
b18bc36127 force creation of workdir (#1070) 2023-02-21 18:10:36 -08:00
Ean Garvey
6d2a485264 Add --benchmark_dispatches option to pytest. (#800) 
* Add --benchmark_dispatches option to pytest.

* Update README.md and fix filepath for dispatch benchmarks
 2023-02-19 12:16:18 -06:00
Eliasj42
87fd13d8eb added an example to run sharded bloom (#1003) 
Co-authored-by: Elias Joseph <elias@nod-labs.com>
 2023-02-13 10:37:47 -08:00
Anush Elangovan
41ee65b377 Revert "Enable --device_allocator=caching" 
This reverts commit 83fe477066.
 2023-02-09 23:00:06 -08:00
Anush Elangovan
83fe477066 Enable --device_allocator=caching 2023-02-09 22:58:46 -08:00
Ean Garvey
c28cc4c919 Fix local_tank_cache handling in shark_downloader. (#981) 2023-02-09 14:52:03 -06:00
Prashant Kumar
3595b4aaff Incorporate latest changes in the shark_dynamo backend. 2023-02-08 20:37:30 +05:30
Ean Garvey
1eb9436836 Fix generate_sharktank args. 2023-02-07 14:06:07 +05:30
Ean Garvey
9604d9ce81 make --update_tank update only if hash mismatch 2023-02-07 14:06:07 +05:30
Ean Garvey
481d0553d8 Remove unnecessary repro_dir / shark_tmp usage 2023-02-07 14:06:07 +05:30
yzhang93
7cf31a6ae4 Fix iree-benchmark flag names (#952) 2023-02-04 22:24:18 -08:00
Quinn Dawkins
fbd6224b04 Revert "Revert pipelines (#948)" (#951) 
This reverts commit 8115b26079.
Additionally fixes img2col by adding detach elementwise from named op
passes.
 2023-02-04 22:44:08 -05:00
powderluv
8115b26079 Revert pipelines (#948) 
* Revert "[SD] Modify the flags to use --iree-preprocessing-pass-pipeline (#914)"

This reverts commit a783c089a9.

* Revert "Fix iree flags due to the change in shark-runtime (#944)"

This reverts commit 1d38d49162.
 2023-02-04 07:09:51 -08:00
yzhang93
1d38d49162 Fix iree flags due to the change in shark-runtime (#944) 2023-02-03 21:34:02 -08:00
Ean Garvey
ba07e94a5e disable Torch Inductor autotuner in benchmarks (#919) 2023-02-02 13:25:43 -06:00
aldesilv
b3fc0f29cc enable additional flags for tank test models (#866) 
Co-authored-by: Alex <alexander@nod-labs.com>
 2023-02-02 11:19:33 -08:00
Ean Garvey
a90812133b Enable pytests on Windows (#901) 2023-02-01 18:36:41 -06:00
powderluv
e26a70aa4f Drop old cli and webui (#911) 2023-02-01 13:13:46 -08:00