Commit Graph

6 Commits (cbf5a80a4466a5b320f6cb696c0879b8ad768025)

Author SHA1 Message Date
Haian Huang(深度眸) 1c833eb195
[Feature] Support TTA (#553)
* support TTA

* update note
2023-02-24 21:46:24 +08:00
Haian Huang(深度眸) 6400fba1af
[Enchance] Optimize and accelerate YOLOX with RTMDet hyps (#542)
* enchance yolox

* update

* update

* fix

* fix

* fix lint
2023-02-17 11:27:03 +08:00
tianlei 4e8bf17c90
[Improve] Beautify the YOLOv7 configuration (#506)
* Beautify the YOLOv7 configuration

* yolov7 config

* Update configs/yolov7/yolov7_l_syncbn_fast_8x16b-300e_coco.py

Co-authored-by: Haian Huang(深度眸) <1286304229@qq.com>

* Update configs/yolov7/yolov7_tiny_syncbn_fast_8x16b-300e_coco.py

Co-authored-by: Haian Huang(深度眸) <1286304229@qq.com>

* Update configs/yolov7/yolov7_w-p6_syncbn_fast_8x16b-300e_coco.py

Co-authored-by: Haian Huang(深度眸) <1286304229@qq.com>

* Update configs/yolov7/yolov7_w-p6_syncbn_fast_8x16b-300e_coco.py

Co-authored-by: Haian Huang(深度眸) <1286304229@qq.com>

* Update configs/yolov7/yolov7_w-p6_syncbn_fast_8x16b-300e_coco.py

Co-authored-by: Haian Huang(深度眸) <1286304229@qq.com>

* Update configs/yolov7/yolov7_w-p6_syncbn_fast_8x16b-300e_coco.py

Co-authored-by: Haian Huang(深度眸) <1286304229@qq.com>

* Beautify the YOLOv7 configuration

---------

Co-authored-by: Haian Huang(深度眸) <1286304229@qq.com>
2023-02-10 10:18:57 +08:00
Haian Huang(深度眸) 884330108d
[BC Breaking] Keep the same input shape order as mmdet 3.0.0rc5 (#436)
* Keep the same input shape order as mmdet

* add desc

* update

* update
2023-01-06 19:00:30 +08:00
CaiLiYuan 9985e0bc08
Update yolov7_l_syncbn_fast_8x16b-300e_coco.py (#333)
change:model->bbox_head->head_module->dict(num_classes=num_classes)
2022-12-02 16:31:27 +08:00
Haian Huang(深度眸) 573ff033e4
[Feature] Support YOLOv7 P5 training (#243)
* support yolov7

* update dev

* update headmodule and convert

* update

* fix pipeline

* fix loss

* fix optimizer parameter groups

* refactor mosaic9

* refactor optim

* refactor loss

* refactor

* refactor

* refactor new

* support yolov7x

* refine

* support tiny

* refactor model

* refactor model

* support yolov7x inference

* support yolov7-tiny inference

* support yolov7-e inference

* refactor

* support yolov7-tiny train

* add docstr

* fix merge error

* fix merge error

* fix merge error

* fix lint

* fix lint

* fix lint

* fix UT

* update

* update
2022-11-21 17:30:19 +08:00