Ayush Chaurasia
|
0303ced8ab
|
Model enhancements 2 (#109)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
|
2 years ago |
Ayush Chaurasia
|
384f0ef1c6
|
Model interface enhancement (#106)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
|
2 years ago |
Glenn Jocher
|
38d6df55cb
|
Fix `save_json(predn, batch)` (#105)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
8028e2b1b8
|
Make optimizer static method inside trainer (#103)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
|
2 years ago |
Laughing
|
5c6d11bdb2
|
Allocated updated pycocotools metrics fix (#101)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
|
2 years ago |
Glenn Jocher
|
6f0ba81427
|
Add best.pt val and COCO pycocotools val (#98)
Co-authored-by: ayush chaurasia <ayush.chaurarsia@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Glenn Jocher
|
48cffa176e
|
Shorten module paths with new 'nn' dir (#96)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
4fb04be20b
|
DDP and new dataloader Fix (#95)
Co-authored-by: Laughing-q <1185102784@qq.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
16e3c08883
|
Import YOLOv5 dataloader (#94)
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Glenn Jocher
|
249dfbdc05
|
Fix ClearML Mosaic callback to 'on_train_epoch_end' (#92)
|
2 years ago |
Glenn Jocher
|
98815d560f
|
YOLOv5 updates (#90)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Glenn Jocher
|
ebd3cfb2fd
|
YOLOv8 architecture updates from R&D branch (#88)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
5fbea25f0b
|
Use integrations only in master process (#89)
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
|
2 years ago |
Glenn Jocher
|
cb4f20f3cf
|
Add TensorBoard support (#87)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Glenn Jocher
|
248d54ca03
|
Update metrics names (#85)
|
2 years ago |
Glenn Jocher
|
6432afc5f9
|
Rename `img_size` to `imgsz` (#86)
|
2 years ago |
Ayush Chaurasia
|
ae2443c210
|
Add flops, num_params, inference speed logging and best.pt logging (#84)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
|
2 years ago |
Ayush Chaurasia
|
d198f3f9d1
|
Update docs (#73)
Co-authored-by: Laughing-q <1185102784@qq.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Laughing
|
7690cae2fc
|
add a naive DDP for model interface (#78)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Ayush Chaurasia <ayush.chaurarsia@gmail.com>
|
2 years ago |
Ayush Chaurasia
|
48c95ba083
|
Change class depending on dataset in model interface (#77)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
eb5adf4e0b
|
Model enhancement (#75)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
d85b44f259
|
Update docs (#71)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Laughing
|
d63ee112d4
|
General cleanup (#69)
Co-authored-by: ayush chaurasia <ayush.chaurarsia@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
|
2 years ago |
Ayush Chaurasia
|
7ae45c6cc4
|
[WIP] Model interface (#68)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Laughing-q <1185102784@qq.com>
|
2 years ago |
Ayush Chaurasia
|
e6737f1207
|
Predictor support (#65)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Laughing-q <1185102784@qq.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
|
2 years ago |
Laughing
|
fbeeb5d1e1
|
add resuming (#63)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
5a52e7663a
|
standalone val (#56)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Laughing
|
3a241e4cea
|
update segment training (#57)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: ayush chaurasia <ayush.chaurarsia@gmail.com>
|
2 years ago |
Ayush Chaurasia
|
d0b0fe2592
|
CLI updates (#58)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
c5f5b80c04
|
Deterministic training (#53)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
ayush chaurasia
|
793dde365d
|
Revert "update"
This reverts commit 486bc49414 .
|
2 years ago |
ayush chaurasia
|
486bc49414
|
update
|
2 years ago |
Ayush Chaurasia
|
d7df1770fa
|
Add warmup and accumulation (#52)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
298287298d
|
Add clearml logging (#51)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
4291b9c31c
|
Add EMA and model checkpointing (#49)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
|
2 years ago |
Glenn Jocher
|
27d6545117
|
General console printout updates (#48)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Laughing
|
47f1cb3ef4
|
Fix some cuda training issues of segmentation (#46)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
db1031a1a9
|
Allow setting model attributes before training (#45)
|
2 years ago |
Ayush Chaurasia
|
832ea56eb4
|
update model initialization design, supports custom data/num_classes (#44)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
1f3aad86c1
|
new check_dataset functions (#43)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
d143ac666f
|
[rename] - preprocess-batch -> preprocess, preprocess_preds -> postprocess (#42)
|
2 years ago |
Glenn Jocher
|
4c68b9dcf6
|
Add is_colab() and is_kaggle() (#41)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
f56c9bcc26
|
Segmentation support & other enchancements (#40)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
|
2 years ago |
Glenn Jocher
|
e4f7458d90
|
Hydra *.yml extension deprecated fix (#34)
|
2 years ago |
Glenn Jocher
|
6fe8bead35
|
Classify training cleanup (#33)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
92c60758dd
|
Smart Model loading (#31)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
1054819a59
|
Add initial model interface (#30)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
7b560f7861
|
Model builder (#29)
Co-authored-by: Ayush Chaurasia <ayush.chuararsia@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
2 years ago |
Ayush Chaurasia
|
c5cb76b356
|
Metrics and loss structure (#28)
Co-authored-by: Ayush Chaurasia <ayush.chuararsia@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
|
2 years ago |
Ayush Chaurasia
|
d0b3c9812b
|
Trainer + Dataloaders (#27)
Co-authored-by: Laughing-q <1185102784@qq.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
Co-authored-by: Ayush Chaurasia <ayushchaurasia@Ayushs-MacBook-Pro.local>
Co-authored-by: Ayush Chaurasia <ayush.chuararsia@gmail.com>
|
2 years ago |