github/AMD-SHARK-Studio
1
1
Fork 0
mirror of https://github.com/nod-ai/AMD-SHARK-Studio.git synced 2026-02-19 11:56:43 -05:00
Code Issues Packages Projects Releases Wiki Activity
566 Commits 66 Branches 705 Tags
rdna
Commit Graph

566 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Gaurav Shukla
4b1a0b43ff [WEB] Remove long prompts support 
It removes support to long prompts due to higher lag in loading long prompts.

Signed-Off-by: Gaurav Shukla <gaurav@nod-labs>
 2022-11-03 18:57:58 +05:30
Gaurav Shukla
099f2160c3 [WEB] fix background color 
Signed-Off-by: Gaurav Shukla
 2022-11-03 17:36:24 +05:30
Gaurav Shukla
9d2d62dedf [WEB] Add support for long prompts (#467) 2022-11-03 03:27:36 -07:00
Gaurav Shukla
15ed05b221 [WEB] Update the title (#466) 2022-11-02 14:30:03 -07:00
Gaurav Shukla
7c825fc288 [WEB] CSS changes to the web-ui (#465) 
This commit updates UI with styling.

Signed-Off-by: Gaurav Shukla <gaurav@nod-labs.com>

Signed-off-by: Gaurav Shukla <gaurav@nod-labs.com>
 2022-11-02 12:36:11 -07:00
Gaurav Shukla
88f8718635 [WEB] Load prompts from json 
The prompt examples will now be loaded from a json file `prompts.json`.

Signed-Off-by: Gaurav Shukla
 2022-11-02 20:52:34 +05:30
Prashant Kumar
a081733a42 Add the clip text shark_model. (#458) 2022-11-02 00:08:33 -07:00
Gaurav Shukla
06ccfb0533 [WEB] Load vae and unet during server start up 
The vae and unet models(both fp16 and fp32 variant) can be loaded at
server startup in order to reduce web response time.

Signed-Off-by: Gaurav Shukla <gaurav@nod-labs.com>
 2022-11-01 23:11:52 +05:30
Gaurav Shukla
b18d75e3f7 [WEB] Use tuned version of UNET fp16 
This commit updates SD script in order to use the tuned version of Unet
fp16 model.

Signed-Off-by: Gaurav Shukla <gaurav@nod-labs.com>
 2022-11-01 19:00:21 +05:30
Quinn Dawkins
3e7efaa048 Switch stable diffusion to the new tuned model (#455) 2022-10-31 15:15:31 -07:00
Gaurav Shukla
a3fdfc81db [WEB] Minor changes in the shark web (#454) 
1. Default steps = 50.
2. Live preview will yield intermediate image at every 5 steps.
3. Add logs to .gitignore

Signed-Off-by: Gaurav Shukla <gaurav@nod-labs.com>

Signed-off-by: Gaurav Shukla <gaurav@nod-labs.com>
 2022-10-31 14:29:00 -07:00
Gaurav Shukla
f4c91df1df [WEB] Add pillow dependency (#453) 
Signed-Off-by: Gaurav Shukla <gaurav@nod-labs.com>

Signed-off-by: Gaurav Shukla <gaurav@nod-labs.com>
 2022-10-31 12:57:21 -07:00
Prashant Kumar
32e1ba8c0d Adding batch_size support for stable diffusion. 2022-11-01 00:57:52 +05:30
Gaurav Shukla
1939376d72 [WEB] Cache model parameters (#452) 
This commit cache some of the model parameters to reduce the response
time of shark web.

Signed-Off-by: Gaurav Shukla <gaurav@nod-labs.com>

Signed-off-by: Gaurav Shukla <gaurav@nod-labs.com>
 2022-10-31 11:55:10 -07:00
Gaurav Shukla
25931d48a3 [WEB] Update stable diffusion UI and enable live preview (#447) 
This commit enables live preview feature and also updates stable
diffusion web UI.

Signed-Off-by: Gaurav Shukla <gaurav@nod-labs.com>

Signed-off-by: Gaurav Shukla <gaurav@nod-labs.com>
 2022-10-31 04:10:15 -07:00
powderluv
024c5e153a Update Windows in README 2022-10-30 22:27:03 -07:00
powderluv
83f34b645d Add Windows instructions 2022-10-30 22:25:42 -07:00
powderluv
3f9f450e0d Add setup_venv.ps1 for windows (#448) 
Powershell users can run ./setup_venv.ps1 to setup the env
 2022-10-30 22:17:35 -07:00
powderluv
fd89b06641 Drop RDNA1 for now 2022-10-29 14:29:09 -07:00
Gaurav Shukla
f8dc996004 Update vulkan-target-triple for Radeon devices. (#446) 
Signed-Off-by: Gaurav Shukla <gaurav@nod-labs.com>

Signed-off-by: Gaurav Shukla <gaurav@nod-labs.com>
 2022-10-29 14:27:20 -07:00
Phaneesh Barwaria
e6a964088b Add os agnostic vulkan device name check (#445) 2022-10-29 13:19:14 -07:00
Gaurav Shukla
e3e767c7eb [WEB] Remove live preview and disable resnet|albert_maskfill 
This commit removes live preview feature for now as it's not functional.
This feature will be added in the next patch.

Signed-Off-by: Gaurav Shukla <gaurav@nod-labs.com>
 2022-10-30 00:37:59 +05:30
Quinn Dawkins
239c19eb12 Update Stable diffusion script to enable use of tuned models (#443) 2022-10-29 01:42:49 -04:00
Eliasj42
7f37599a60 Added a dispatch benchmarking tool (#441) 
To produce benchmarks of individual dispatches, you can add --dispatch_benchmarks=All --dispatch_benchmarks_dir=<output_dir> to your command line argument.

Co-authored-by: Elias Joseph <elias@nod-labs.com>
 2022-10-28 14:31:03 -07:00
Prashant Kumar
77c9a2c5ea Add profiling vulkan_device info and minor changes to reflect upstream 
changes.
 2022-10-28 18:02:07 +05:30
Ean Garvey
fd7baae548 Serialize torch-mlir CAPI module as bytecode instead of string. (#435) 
* Serialize torch-mlir CAPI as bytecode instead of string.

* Minor fixes to MLIR data handling in SHARK python.
 2022-10-27 14:37:15 -05:00
Stanley Winata
01fdf5ee16 [example][SD] compile fp16 with iree-spirv-unify-aliased-resources (#436) 2022-10-27 05:12:28 -07:00
Gaurav Shukla
e52f533c16 [WEB] Save vmfb and add live preview 
This commit updates SD script to save the compiled module and also adds
live preview of generated images.

Signed-off-by: Gaurav Shukla<gaurav@nod-labs.com>
 2022-10-26 23:20:53 +05:30
Quinn Dawkins
fbd77dc936 Enable iterator space fusion for SD (#432) 2022-10-26 01:08:26 -04:00
Quinn Dawkins
cdc6dd19e3 Force stable diffusion fp16 and fp32 to generate images with similar noise (#431) 2022-10-25 17:28:18 -04:00
PhaneeshB
fd578a48a9 add cli args for vulkan target triple 2022-10-25 21:47:26 +05:30
Ean Garvey
9956099516 Add pytest option for updating tank and fix save_mlir function. (#413) 
* Use IREE tf tools to save .mlir modules when generating shark_tank.

* Add option to pytest for enabling auto-updates to local shark tank.

* xfail mobilenet torch on cpu, cuda and fix CI macos setup

* Update test-models.yml to disable macos vulkan CI.
 2022-10-25 21:29:18 +05:30
powderluv
f97b8fffed Update README.md 2022-10-24 12:51:49 -07:00
Gaurav Shukla
7b9e309724 [WEB] Expose SD parameters in the web ui (#427) 2022-10-24 04:34:35 -07:00
Quinn Dawkins
1d33913d48 Add option to save and load precompiled flatbuffer (#425) 2022-10-23 16:24:09 -07:00
Prashant Kumar
a48eaaed20 Pass the flags to vae. 2022-10-23 23:57:48 +05:30
Prashant Kumar
2741b8be53 Pass the flags to vae. (#422) 2022-10-23 11:23:13 -07:00
Anush Elangovan
4f906a265c Fix lint 2022-10-22 12:43:52 -07:00
Anush Elangovan
0dff8d7af0 Simple download script to prime the hf model cache 2022-10-21 17:42:05 -07:00
Quinn Dawkins
4f0d0d8167 Update vulkan gui README for iree-vulkan-gui + Stable Diffusion (#399) 2022-10-21 14:02:40 -04:00
Vivek Khandelwal
d513060b21 Add params for Stable Diffusion (#420) 2022-10-21 23:11:09 +05:30
Prashant Kumar
d1a25ce4f3 Update stable_args.py 2022-10-21 17:26:31 +05:30
Gaurav Shukla
51c98695b2 [WEB] Update stable diffusion inference 
This commit updates the stable diffusion web incorporating the latest
improvements.

Signed-Off-by: Gaurav Shukla <gaurav@nod-labs.com>
 2022-10-21 01:26:38 +05:30
Quinn Dawkins
b448770ec2 Add ms/iter timing for stable diffusion script (#414) 2022-10-20 13:32:37 -04:00
Prashant Kumar
5fe22a7980 Minor fix. 2022-10-20 22:57:22 +05:30
Prashant Kumar
38ae6b5af4 Add stable_diffusion fp16 and fp32 with args. 2022-10-20 21:47:11 +05:30
Ean Garvey
0bfe30d75d Fix issues with extra_args in benchmarks, pin tf==2.10 (#411) 2022-10-20 06:55:26 -07:00
Quinn Dawkins
7be1d7d0be Add option for extra arguments through SharkInference.compile (#408) 2022-10-19 15:32:48 -05:00
Prashant Kumar
0d74c873f0 Add stable_diff_f16 version. (#407) 2022-10-19 10:04:24 -07:00
powderluv
139aff2938 Update nightly.yml 
fix links
 2022-10-18 23:42:22 -07:00