github/AMD-SHARK-Studio
1
1
Fork 0
mirror of https://github.com/nod-ai/AMD-SHARK-Studio.git synced 2026-04-03 03:00:17 -04:00
Code Issues Packages Projects Releases Wiki Activity
976 Commits 66 Branches 705 Tags
enable_caching_alloc
Commit Graph

459 Commits

Author SHA1 Message Date
Anush Elangovan
d6e59c6241 black format comments 2023-01-22 16:34:40 -08:00
powderluv
458eb5d34c detect RX 7900 better 2023-01-22 16:32:27 -08:00
Erkin Alp Güney
8259f08864 Collapsibles for Win10 and Linux users (#851) 
Co-authored-by: powderluv <powderluv@users.noreply.github.com>
 2023-01-22 09:50:33 -08:00
Prashant Kumar
b3ab0a1843 Add width and height support for the scheduler. 2023-01-22 23:16:50 +05:30
Daniel Garvey
e842c8c19b add main.py testing for sdiff (#836) 
Co-authored-by: dan <dan@nod-labs.com>
Co-authored-by: powderluv <powderluv@users.noreply.github.com>
 2023-01-22 01:16:17 -08:00
powderluv
f6c3112d44 Revert "potential fix to pre-load DLL dir for torch-mlir (#848)" (#852) 
This reverts commit 6c470d8131.
 2023-01-22 00:09:35 -08:00
yzhang93
7059610632 Modify the default for --hf_model_id flag 2023-01-21 11:21:47 +05:30
powderluv
2d272930d9 Update to signed build 455 2023-01-20 16:50:42 -08:00
powderluv
6c470d8131 potential fix to pre-load DLL dir for torch-mlir (#848) 
Doesn't regress the main.py script but system already pre-loaded
the DLL so needs more testing.
 2023-01-20 14:48:45 -08:00
stanley
c4a9365aa1 [Shark][Training] Refresh SharkTrainer to latest APIs. 2023-01-19 20:30:15 +00:00
Prashant Kumar
9d3af37104 bugfix related to the height width params. 2023-01-20 00:21:44 +05:30
Prashant Kumar
7b3d57cff7 Add height and width as args. 2023-01-19 23:43:29 +05:30
Abhishek Varma
a802270da9 [SD-CLI] Update README.md about variants.json 2023-01-19 22:46:54 +05:30
Abhishek Varma
dd194a8758 [SD-CLI] Reorder loading of opt_params when needed 
-- This commit reorders loading of opt_params when `import_mlir` is not used.

Signed-off-by: Abhishek Varma <abhishek@nod-labs.com>
 2023-01-19 22:02:51 +05:30
Abhishek Varma
6de02de221 [SD-CLI] Make using custom models easier 
-- This commit makes using custom models easier using a combination of
   `import_mlir`, `ckpt_loc` and `hf_model_id`.

Signed-off-by: Abhishek Varma <abhishek@nod-labs.com>
 2023-01-19 22:02:36 +05:30
Abhishek Varma
85259750bf [SD-CLI] Fix variants.json mapping 
-- This commit fixes variants.json's mapping.

Signed-off-by: Abhishek Varma <abhishek@nod-labs.com>
 2023-01-19 22:02:36 +05:30
Prashant Kumar
1249f0007d Remove args.variant and args.version with args.custom_model. 2023-01-19 19:55:12 +05:30
Abhishek Varma
db0514d3fa [SD-CLI] Fix get_model_configuration to use max_length 
-- This commit fixes `get_model_configuration` to use `max_length`.

Signed-off-by: Abhishek Varma <abhishek@nod-labs.com>
 2023-01-19 19:10:04 +05:30
Abhishek Varma
dce42a7fad [SD-CLI] Fix args.max_length range check 
This commit fixes args.max_length range check.

Signed-off-by: Abhishek Varma <abhishek@nod-labs.com>
 2023-01-19 18:26:23 +05:30
Prashant Kumar
ec0b380194 Refactor shark_tank models and custom models. 
The custom models shouldn't depend on shark_tank in anyway.
 2023-01-19 13:56:11 +05:30
xzuyn
2a1d1c1001 make jpeg optimized and progressive (#820) 
* GUI make jpeg optimized and progressive

* CLI make jpeg optimized and progressive
 2023-01-17 16:35:36 -08:00
Abhishek Varma
df7eb80e5b [SD-CLI] Make custom_model take highest priority for generating models if present 
-- This commit makes `custom_model` take highest priority for generating models if present.

Signed-off-by: Abhishek Varma <abhishek@nod-labs.com>
 2023-01-17 22:50:58 +05:30
Abhishek Varma
bc1bb1d188 [SD-CLI] Fix vmfb naming + update README.md for custom_model 
-- This commit introduces a fix for .vmfb naming to strip away any
   non-alphanumeric characters from `custom_model` path.
-- It also updates the README.md to include the `custom_model` arg.

Signed-off-by: Abhishek Varma <abhishek@nod-labs.com>
 2023-01-17 16:27:54 +05:30
Abhishek Varma
3af2dd10ce [SD-CLI] Add CKPT support to update models irrespective of import_mlir flag 
-- This commit adds CKPT support to update models irrespective of `import_mlir` flag.

Signed-off-by: Abhishek Varma <abhishek@nod-labs.com>
 2023-01-17 13:24:27 +05:30
yzhang93
dd22c65855 Add CUDA tuned models for SD variants (#814) 2023-01-16 09:38:27 -08:00
PhaneeshB
48137ced19 add png as default format 2023-01-16 18:37:36 +05:30
Phaneesh Barwaria
6eb47c12d1 add multi-run in single execution (#812) 2023-01-13 11:12:43 -08:00
Prashant Kumar
5a1fc6675a This PR adds --import-mlir for f16 tensors without cuda. 2023-01-13 22:19:53 +05:30
Prashant Kumar
6f80825814 Modify import_with_fx to import with dtype=f16. 2023-01-13 22:19:53 +05:30
PhaneeshB
f0dd48ed2a remaining disk space warning 2023-01-13 19:34:05 +05:30
PhaneeshB
ee0009d4b8 pythonize uname for cpu target triple in windows 2023-01-12 22:39:49 +05:30
PhaneeshB
9d851c3346 small fixes 2023-01-12 22:32:24 +05:30
xzuyn
5d117af8ae Increase JPEG output quality & disable subsampling (#801) 
* Increase JPEG output quality & disable subsampling

Increased to JPEG95 from the default JPEG75 which is way too compressed. Output image size is now ~100kb. Previously was ~20kb.

* Increase JPEG output quality & disable subsampling

Add jpeg quality increase on cli

* line length changes

* line length changes
 2023-01-11 23:06:11 -08:00
yzhang93
bb41c2d15e Add VAE cuda tuned model (#796) 2023-01-11 14:15:03 -08:00
powderluv
eba138ee4a Revert "Change address for connection test (#785)" (#797) 
This reverts commit 187f0fa70c.
 2023-01-11 12:01:37 -08:00
fokin33
dbc0f81211 Add simple telegram bot (#787) 2023-01-11 09:20:23 -06:00
mariecwhite
d0b613d22e Enable Torch-Inductor Part 2 2023-01-10 20:15:29 -08:00
Ean Garvey
72f29b67d5 Add Resnet50 fp16 variant to pytests. (#760) 2023-01-10 16:31:11 -08:00
Quinn Dawkins
9570045cc3 Fix tuned model selection for non-vulkan devices (#792) 2023-01-10 19:04:21 -05:00
Phaneesh Barwaria
e4efdb5cbb add json data for each image (#790) 
Co-authored-by: powderluv <powderluv@users.noreply.github.com>
 2023-01-10 13:13:07 -08:00
calcifer11
187f0fa70c Change address for connection test (#785) 
Some ISP's (like mine) reserves 1.1.1.1 for internal testing, meaning _internet_connected(); needlessly retries for a minute until it fails even though my connection is fine.
Propose 8.8.8.8 instead as this is also publically available and not normally blocked by ISPs.
 2023-01-10 10:51:30 -08:00
mariecwhite
183e447d35 Enable Torch Inductor (#784) 2023-01-10 20:57:58 +11:00
yzhang93
47a119a37f [SD] Add CUDA A100 tuned model (#773) 2023-01-09 15:22:27 -08:00
George Petterson
6ad9b213b9 Add GCN4 
(cherry picked from commit 3be072b3c09c9b38bc2d79ad6e6900eefee49a1c)
 2023-01-09 21:09:50 +05:30
PhaneeshB
e4375e8195 Add support for vulkan target env 2023-01-09 21:09:50 +05:30
mariecwhite
487bf8e29b Enable TF32 in Torch if specified (#768) 2023-01-09 06:48:57 -08:00
Prashant Kumar
4102c124a9 Add the shark upscaler model. (#759) 2023-01-05 14:07:20 -08:00
yzhang93
135bad3280 [SD] Update v1.4 tuned model (#758) 2023-01-05 11:04:30 -08:00
yzhang93
782b449c71 Add script to auto annotate SD models and variants (#751) 
* Add script to auto annotate SD models and variants

* Add model config files

* Add script to auto annotate SD models and variants

* Add model config files

* Move config files to shark_tank
 2023-01-04 15:53:10 -08:00
jinchen62
017dcab685 Add target triple support for TITAN RTX (#756) 2023-01-04 15:39:00 -08:00