Commit Graph

298 Commits

Author SHA1 Message Date
henryruhs
54c5987ef0 Implement predict video 2023-05-31 19:10:41 +02:00
henryruhs
23753b523e Implement predict video 2023-05-31 19:10:41 +02:00
henryruhs
96a61397ef Implement predict video 2023-05-31 19:10:41 +02:00
henryruhs
4e2be506ce Fix GPU for NVIDIA 2023-05-31 19:10:40 +02:00
Pikachu~~~
c16049517d
Update requirements.txt 2023-06-01 00:55:36 +08:00
Pikachu~~~
667c7da540
Add a processer bar with remain time for CLI 2023-06-01 00:54:52 +08:00
Pikachu~~~
cfd29513b4
Add a processer bar with remain time for CLI 2023-06-01 00:50:58 +08:00
Somdev Sangwan
38fb60efca
fix paths fr 2023-05-31 21:57:07 +05:30
CeeBeeEh
7f4e96b086
Merge branch 's0md3v:main' into main 2023-05-31 12:22:34 -04:00
Somdev Sangwan
51664c6d71
Fix path issue 2023-05-31 21:38:22 +05:30
chris
841b2d06c3 Merge remote-tracking branch 'origin/main'
# Conflicts:
#	core/globals.py
#	core/processor.py
#	run.py
2023-05-31 11:35:39 -04:00
chris
6fbc2d32b3 - allow processing all faces within a frame 2023-05-31 11:28:13 -04:00
Somdev Sangwan
f9a2e1d17d
Update run.py 2023-05-31 20:52:43 +05:30
Somdev Sangwan
92e26d0326
more confidence 2023-05-31 20:32:48 +05:30
Somdev Sangwan
1326df0f45
fix output dir 2023-05-31 20:31:44 +05:30
Somdev Sangwan
4e0b456a4a
fix path in windows 2023-05-31 19:37:28 +05:30
Somdev Sangwan
bdfd239231
reorder imports 2023-05-31 19:33:19 +05:30
Somdev Sangwan
80da02775a
fix torch, improve face detection 2023-05-31 19:12:13 +05:30
Somdev Sangwan
b545c69167
bruh moment 12 2023-05-31 18:49:58 +05:30
Somdev Sangwan
a9571f6f74 use exact fps 2023-05-31 18:47:20 +05:30
Somdev Sangwan
ab2fc1c221
Update processor.py 2023-05-31 18:24:11 +05:30
Somdev Sangwan
c0f486ed83
Update requirements.txt 2023-05-31 18:22:35 +05:30
Somdev Sangwan
b88eff0e9b
Update requirements.txt 2023-05-31 18:01:48 +05:30
Somdev Sangwan
e44dd7e567
Update requirements.txt 2023-05-31 17:50:04 +05:30
Somdev Sangwan
2ed83803bd
add protobuf version 2023-05-31 17:37:35 +05:30
Somdev Sangwan
cb3d045bb9
continue when face not found 2023-05-31 16:40:19 +05:30
Somdev Sangwan
7c3f75ae88 better face detection 2023-05-31 15:20:26 +05:30
Henry Ruhs
2c9631e156
Merge pull request #167 from s0md3v/feat/lint-within-pipeline
Introduce lint pipeline
2023-05-31 11:15:17 +02:00
Somdev Sangwan
ef28e3ad7f
fix target_path undefined 2023-05-31 14:43:21 +05:30
Somdev Sangwan
7cbffe4a6f
Update requirements.txt 2023-05-31 14:37:59 +05:30
henryruhs
a8317003a8 Add chmod +x 2023-05-31 11:00:35 +02:00
henryruhs
7abb42724e Introduce lint pipeline and fix some minor issues flake8 complained about 2023-05-31 10:32:40 +02:00
Somdev Sangwan
4867f356ef
fix gpu issues for some envs 2023-05-31 12:43:43 +05:30
Somdev Sangwan
1cc34ce8e0 bruh moment 12 2023-05-31 12:42:34 +05:30
Somdev Sangwan
077b931cef
Rename globals.py to core.globals.py 2023-05-31 12:41:41 +05:30
Somdev Sangwan
278735d6a6
Merge pull request #157 from SaqibAMA/main
Fix: Added missing `rreplace` import from core.utils.
2023-05-31 11:40:55 +05:30
Saqib Ali
ccd63b1dce
Fix: Added missing rreplace import from core.utils.
An import for `rreplace` was missing from core.utils in the run.py file which made the program crash. It has been fixed and added.
2023-05-31 10:53:13 +05:00
Somdev Sangwan
7e68260f71
bruh moment 11 2023-05-31 10:38:16 +05:30
Somdev Sangwan
6a09ee0c4b
handle non-nvidia gpus better 2023-05-31 10:36:18 +05:30
Somdev Sangwan
1aacdb0d62
onnxruntime-gpu -> onnxruntime 2023-05-31 10:27:18 +05:30
Somdev Sangwan
450f8fade7
clarity 2023-05-31 10:16:56 +05:30
Somdev Sangwan
6416d4ebe8
Merge pull request #155 from ShivamKumar2002/patch-1
Fix GPU not being used even when it available
2023-05-31 09:39:14 +05:30
Somdev Sangwan
c2ce47dec5
Merge pull request #151 from Moeblack/TensorRTExcutionProvider-error
fix TensorRTExcutionProvider errors
2023-05-31 09:37:44 +05:30
Shivam Kumar
9c1ea4fa01
Fix GPU not being used even when it available
On lots of systems onnxruntime doesn't detect GPU unless pytorch is imported before it. So despite having CUDA and CUDNN setup correctly, it is only using CPUExecutionProvider. By importing pytorch first, the issue is fixed. So let's use this until any official solution is available.

See: https://stackoverflow.com/questions/75294639/onnxruntime-inference-with-cudnn-on-gpu-only-working-if-pytorch-imported-first
2023-05-31 09:20:41 +05:30
Pikachu~~~
4183e71cf8
fix TensorRTExcutionProvider errors 2023-05-31 10:06:42 +08:00
Henry Ruhs
6258f3c084
Update processor.py 2023-05-31 02:44:58 +02:00
Somdev Sangwan
251fde9596
Merge pull request #91 from henryruhs/main
Introduce pre_check() and limit_resources()
2023-05-31 02:27:34 +05:30
henryruhs
411818d0f9 Need to return at least 2 cores for default 2023-05-30 22:44:48 +02:00
henryruhs
3e35bea1df Adjust max-cores with lower default 2023-05-30 22:16:38 +02:00
henryruhs
5feb6b0d71 Resolve absolute model path 2023-05-30 21:52:39 +02:00