- 20 10月, 2020 1 次提交
-
-
由 wangguanzhong 提交于
* migrate detection ops from Paddle to PaddleDetection * add unittest for detection ops * update api & unittest * add copyright & api name to migrate
-
- 13 10月, 2020 1 次提交
-
-
由 wanghuancoder 提交于
* refine PaddleDetection benchmard print, test=develop * refine PaddleDetection benchmard print, test=develop * refine PaddleDetection benchmard print, test=develop * refine PaddleDetection benchmard print, test=develop
-
- 27 9月, 2020 1 次提交
-
-
由 wangxinxin08 提交于
-
- 27 8月, 2020 1 次提交
-
-
由 wangguanzhong 提交于
-
- 10 8月, 2020 1 次提交
-
-
由 wangguanzhong 提交于
* refactor dygraph & add mask rcnn fpn * fix initialization & minor update * refine architecture of mask_rcnn
-
- 23 7月, 2020 1 次提交
-
-
由 wangguanzhong 提交于
-
- 13 7月, 2020 1 次提交
-
-
由 FDInSky 提交于
add train&eval add reader padding add new config
-
- 01 6月, 2020 1 次提交
-
-
由 FDInSky 提交于
clear directories & files for dygraph branch
-
- 20 5月, 2020 1 次提交
-
-
由 走神的阿圆 提交于
Add image for visualdl board.
-
- 13 5月, 2020 1 次提交
-
-
由 qingqing01 提交于
* Update docs * Update README.md
-
- 09 5月, 2020 1 次提交
-
-
由 wangguanzhong 提交于
* optimize python path
-
- 08 5月, 2020 1 次提交
-
-
由 wangguanzhong 提交于
* add yolov4
-
- 06 5月, 2020 1 次提交
-
-
由 Guanghua Yu 提交于
* update check config * fix slim/xx
-
- 28 4月, 2020 2 次提交
-
-
由 Yang Zhang 提交于
* Revert back to `fluid.clip` for gradient clipping current API is in flux, wait for it to stabilize * Please CI still buggy
-
由 Yang Zhang 提交于
-
- 26 4月, 2020 1 次提交
-
-
由 wangguanzhong 提交于
-
- 22 4月, 2020 1 次提交
-
-
由 qingqing01 提交于
-
- 08 4月, 2020 1 次提交
-
-
由 hysunflower 提交于
-
- 07 4月, 2020 1 次提交
-
-
由 wangguanzhong 提交于
* add devices_num on reader * refine capacity value * refine bufsize
-
- 02 4月, 2020 1 次提交
-
-
由 wangguanzhong 提交于
-
- 17 3月, 2020 1 次提交
-
-
由 xiegegege 提交于
-
- 26 12月, 2019 1 次提交
-
-
由 littletomatodonkey 提交于* Add GIOU loss. * Add DIOU/CIOU loss. * Change loss box type in bbox_head.py. * Remove fuse_elewise_add_act_ops flag in train.py (Error may happen for elem_divbased on the flag).
-
- 23 12月, 2019 1 次提交
-
-
由 Kaipeng Deng 提交于
* split yolov3 loss
-
- 19 12月, 2019 1 次提交
-
-
由 qingqing01 提交于
* Polish Reader to simplify preprocessing logic. # sample_transforms-> make batch -> batch_transforms in Reader. * Clean some code * Imporve yolov3_r50vd_dcn_obj365_pretrained_coco 41.4 to 41.8. * Update all configs.
-
- 30 10月, 2019 1 次提交
-
-
由 wangguanzhong 提交于
* update code & doc
-
- 23 10月, 2019 1 次提交
-
-
由 qingqing01 提交于
-
- 16 10月, 2019 2 次提交
-
-
由 wangguanzhong 提交于
-
由 Yang Zhang 提交于
* Error out if fp16 is enabled and affine channel is used * Add more documentation on mixed precision training
-
- 15 10月, 2019 1 次提交
-
-
由 Yang Zhang 提交于
-
- 14 10月, 2019 2 次提交
-
-
由 wangguanzhong 提交于
* add multi-scale test
-
由 wangguanzhong 提交于
This reverts commit b591cc133ca04dc1ed50631c8d24031bb97db18b.
-
- 12 10月, 2019 2 次提交
-
-
由 Guanghua Yu 提交于
* Add face detection configs and main code. * Integration multi-scale evaluation.
-
由 wangguanzhong 提交于
* unify reader to dataloader * add check_version
-
- 08 10月, 2019 1 次提交
-
-
由 wangchaochaohu 提交于
* refine build strategy for the model optimize test=develop * fix according develop test=develop
-
- 29 9月, 2019 3 次提交
-
-
由 Yang Zhang 提交于
-
由 Yang Zhang 提交于
* Place mixed precision inside PaddleDetection roll back to the monkey patch version as a temporary measure, before it is merged into paddle * Add command flag for `loss_scale` * Fix a stupid indentation error optimizer should be in the mixed precision context * Initial FP16 training * Add mixed precision training to rest of the detection models * Revert "Add support for mixed precision training (#3406)" This reverts commit 3a2c106271885071db7c0d85587540a8f83c24db. * Bug fixes and some tweaks
-
由 wangguanzhong 提交于
* Support finetune by custom dataset * add finetune args * add load finetune * reconstruct load * add transfer learning doc * add fruit demo * add quick start * add data preprocessing FAQ
-
- 25 9月, 2019 3 次提交
-
-
由 wangguanzhong 提交于
-
由 Yang Zhang 提交于
* Disable optimizer fusion for RCNN models waiting for upstream fix * Disable completely
-
由 Yang Zhang 提交于
* Add support for mixed precision training * Disable `fuse_all_reduce_ops` when training in mixed precision waiting for upstream fix
-