Skip to content

P
PaddleDetection

s920243400 / PaddleDetection
与 Fork 源项目一致

Fork自 PaddlePaddle / PaddleDetection

通知 2
Star 0
Fork 0
P
PaddleDetection
未验证 提交 9630ebb4 编写于 作者: W wangguanzhong 提交者： GitHub
浏览文件
操作

fix weights path (#273)

上级 eb7b0ddd
隐藏空白更改
内联 并排
Showing with 26 addition and 26 deletion
configs/cascade_mask_rcnn_r50_fpn_1x.yml
浏览文件 @ 9630ebb4
...@@ -6,7 +6,7 @@ log_smooth_window: 20 ...@@ -6,7 +6,7 @@ log_smooth_window: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar
metric: COCO metric: COCO
weights: output/cascade_mask_rcnn_r50_fpn_1x/model_final/ weights: output/cascade_mask_rcnn_r50_fpn_1x/model_final
num_classes: 81 num_classes: 81
CascadeMaskRCNN: CascadeMaskRCNN:
......
configs/dcn/cascade_mask_rcnn_dcnv2_se154_vd_fpn_gn_s1x.yml
浏览文件 @ 9630ebb4
...@@ -6,7 +6,7 @@ log_iter: 20 ...@@ -6,7 +6,7 @@ log_iter: 20
log_smooth_window: 20 log_smooth_window: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/SENet154_vd_caffe_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/SENet154_vd_caffe_pretrained.tar
weights: output/cascade_mask_rcnn_dcn_se154_vd_fpn_gn_s1x/model_final/ weights: output/cascade_mask_rcnn_dcn_se154_vd_fpn_gn_s1x/model_final
metric: COCO metric: COCO
num_classes: 81 num_classes: 81
......
configs/dcn/cascade_mask_rcnn_dcnv2_se154_vd_fpn_gn_s1x_ms_test.yml
浏览文件 @ 9630ebb4
...@@ -6,7 +6,7 @@ log_iter: 20 ...@@ -6,7 +6,7 @@ log_iter: 20
log_smooth_window: 20 log_smooth_window: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/SENet154_vd_caffe_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/SENet154_vd_caffe_pretrained.tar
weights: output/cascade_mask_rcnn_dcn_se154_vd_fpn_gn_s1x/model_final/ weights: output/cascade_mask_rcnn_dcn_se154_vd_fpn_gn_s1x/model_final
metric: COCO metric: COCO
num_classes: 81 num_classes: 81
......
configs/dcn/mask_rcnn_dcn_r50_fpn_1x.yml
浏览文件 @ 9630ebb4
...@@ -7,7 +7,7 @@ log_iter: 20 ...@@ -7,7 +7,7 @@ log_iter: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar
metric: COCO metric: COCO
weights: output/mask_rcnn_dcn_r50_fpn_1x/model_final/ weights: output/mask_rcnn_dcn_r50_fpn_1x/model_final
num_classes: 81 num_classes: 81
MaskRCNN: MaskRCNN:
......
configs/dcn/mask_rcnn_dcn_r50_vd_fpn_2x.yml
浏览文件 @ 9630ebb4
...@@ -7,7 +7,7 @@ log_iter: 20 ...@@ -7,7 +7,7 @@ log_iter: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_pretrained.tar
metric: COCO metric: COCO
weights: output/mask_rcnn_dcn_r50_vd_fpn_2x/model_final/ weights: output/mask_rcnn_dcn_r50_vd_fpn_2x/model_final
num_classes: 81 num_classes: 81
MaskRCNN: MaskRCNN:
......
configs/face_detection/blazeface.yml
浏览文件 @ 9630ebb4
...@@ -7,7 +7,7 @@ log_smooth_window: 20 ...@@ -7,7 +7,7 @@ log_smooth_window: 20
log_iter: 20 log_iter: 20
metric: WIDERFACE metric: WIDERFACE
save_dir: output save_dir: output
weights: output/blazeface/model_final/ weights: output/blazeface/model_final
# 1(label_class) + 1(background) # 1(label_class) + 1(background)
num_classes: 2 num_classes: 2
......
configs/face_detection/blazeface_nas.yml
浏览文件 @ 9630ebb4
...@@ -7,7 +7,7 @@ log_smooth_window: 20 ...@@ -7,7 +7,7 @@ log_smooth_window: 20
log_iter: 20 log_iter: 20
metric: WIDERFACE metric: WIDERFACE
save_dir: output save_dir: output
weights: output/blazeface_nas/model_final/ weights: output/blazeface_nas/model_final
# 1(label_class) + 1(background) # 1(label_class) + 1(background)
num_classes: 2 num_classes: 2
......
configs/face_detection/blazeface_nas_v2.yml
浏览文件 @ 9630ebb4
...@@ -7,7 +7,7 @@ log_smooth_window: 20 ...@@ -7,7 +7,7 @@ log_smooth_window: 20
log_iter: 20 log_iter: 20
metric: WIDERFACE metric: WIDERFACE
save_dir: output save_dir: output
weights: output/blazeface_nas/model_final/ weights: output/blazeface_nas/model_final
# 1(label_class) + 1(background) # 1(label_class) + 1(background)
num_classes: 2 num_classes: 2
......
configs/face_detection/faceboxes.yml
浏览文件 @ 9630ebb4
...@@ -7,7 +7,7 @@ log_smooth_window: 20 ...@@ -7,7 +7,7 @@ log_smooth_window: 20
log_iter: 20 log_iter: 20
metric: WIDERFACE metric: WIDERFACE
save_dir: output save_dir: output
weights: output/faceboxes/model_final/ weights: output/faceboxes/model_final
# 1(label_class) + 1(background) # 1(label_class) + 1(background)
num_classes: 2 num_classes: 2
......
configs/face_detection/faceboxes_lite.yml
浏览文件 @ 9630ebb4
...@@ -7,7 +7,7 @@ log_smooth_window: 20 ...@@ -7,7 +7,7 @@ log_smooth_window: 20
log_iter: 20 log_iter: 20
metric: WIDERFACE metric: WIDERFACE
save_dir: output save_dir: output
weights: output/faceboxes_lite/model_final/ weights: output/faceboxes_lite/model_final
# 1(label_class) + 1(background) # 1(label_class) + 1(background)
num_classes: 2 num_classes: 2
......
configs/gcnet/mask_rcnn_r50_vd_fpn_gcb_add_r16_2x.yml
浏览文件 @ 9630ebb4
...@@ -6,7 +6,7 @@ log_smooth_window: 20 ...@@ -6,7 +6,7 @@ log_smooth_window: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_pretrained.tar
metric: COCO metric: COCO
weights: output/mask_rcnn_r50_vd_fpn_gcb_add_r16_2x/model_final/ weights: output/mask_rcnn_r50_vd_fpn_gcb_add_r16_2x/model_final
num_classes: 81 num_classes: 81
MaskRCNN: MaskRCNN:
......
configs/gcnet/mask_rcnn_r50_vd_fpn_gcb_mul_r16_2x.yml
浏览文件 @ 9630ebb4
...@@ -6,7 +6,7 @@ log_smooth_window: 20 ...@@ -6,7 +6,7 @@ log_smooth_window: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_pretrained.tar
metric: COCO metric: COCO
weights: output/mask_rcnn_r50_vd_fpn_gcb_mul_r16_2x/model_final/ weights: output/mask_rcnn_r50_vd_fpn_gcb_mul_r16_2x/model_final
num_classes: 81 num_classes: 81
MaskRCNN: MaskRCNN:
......
configs/gn/cascade_mask_rcnn_r50_fpn_gn_2x.yml
浏览文件 @ 9630ebb4
...@@ -5,7 +5,7 @@ use_gpu: true ...@@ -5,7 +5,7 @@ use_gpu: true
log_smooth_window: 20 log_smooth_window: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar
weights: output/cascade_mask_rcnn_r50_fpn_gn_2x/model_final/ weights: output/cascade_mask_rcnn_r50_fpn_gn_2x/model_final
metric: COCO metric: COCO
num_classes: 81 num_classes: 81
......
configs/gn/mask_rcnn_r50_fpn_gn_2x.yml
浏览文件 @ 9630ebb4
...@@ -5,7 +5,7 @@ use_gpu: true ...@@ -5,7 +5,7 @@ use_gpu: true
log_smooth_window: 20 log_smooth_window: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar
weights: output/mask_rcnn_r50_fpn_gn_2x/model_final/ weights: output/mask_rcnn_r50_fpn_gn_2x/model_final
metric: COCO metric: COCO
num_classes: 81 num_classes: 81
......
configs/mask_rcnn_r101_fpn_1x.yml
浏览文件 @ 9630ebb4
...@@ -6,7 +6,7 @@ log_smooth_window: 20 ...@@ -6,7 +6,7 @@ log_smooth_window: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet101_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet101_pretrained.tar
metric: COCO metric: COCO
weights: output/mask_rcnn_r101_fpn_1x/model_final/ weights: output/mask_rcnn_r101_fpn_1x/model_final
num_classes: 81 num_classes: 81
MaskRCNN: MaskRCNN:
......
configs/mask_rcnn_r50_2x.yml
浏览文件 @ 9630ebb4
...@@ -6,7 +6,7 @@ log_smooth_window: 20 ...@@ -6,7 +6,7 @@ log_smooth_window: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar
metric: COCO metric: COCO
weights: output/mask_rcnn_r50_2x/model_final/ weights: output/mask_rcnn_r50_2x/model_final
num_classes: 81 num_classes: 81
MaskRCNN: MaskRCNN:
......
configs/mask_rcnn_r50_fpn_1x.yml
浏览文件 @ 9630ebb4
...@@ -6,7 +6,7 @@ log_smooth_window: 20 ...@@ -6,7 +6,7 @@ log_smooth_window: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar
metric: COCO metric: COCO
weights: output/mask_rcnn_r50_fpn_1x/model_final/ weights: output/mask_rcnn_r50_fpn_1x/model_final
num_classes: 81 num_classes: 81
MaskRCNN: MaskRCNN:
......
configs/mask_rcnn_r50_fpn_2x.yml
浏览文件 @ 9630ebb4
...@@ -5,7 +5,7 @@ use_gpu: true ...@@ -5,7 +5,7 @@ use_gpu: true
log_smooth_window: 20 log_smooth_window: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar
weights: output/mask_rcnn_r50_fpn_2x/model_final/ weights: output/mask_rcnn_r50_fpn_2x/model_final
metric: COCO metric: COCO
num_classes: 81 num_classes: 81
......
configs/mask_rcnn_r50_vd_fpn_2x.yml
浏览文件 @ 9630ebb4
...@@ -6,7 +6,7 @@ log_smooth_window: 20 ...@@ -6,7 +6,7 @@ log_smooth_window: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_vd_pretrained.tar
metric: COCO metric: COCO
weights: output/mask_rcnn_r50_vd_fpn_2x/model_final/ weights: output/mask_rcnn_r50_vd_fpn_2x/model_final
num_classes: 81 num_classes: 81
MaskRCNN: MaskRCNN:
......
configs/mask_rcnn_se154_vd_fpn_s1x.yml
浏览文件 @ 9630ebb4
...@@ -5,7 +5,7 @@ use_gpu: true ...@@ -5,7 +5,7 @@ use_gpu: true
log_smooth_window: 20 log_smooth_window: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/SENet154_vd_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/SENet154_vd_pretrained.tar
weights: output/mask_rcnn_se154_vd_fpn_s1x/model_final/ weights: output/mask_rcnn_se154_vd_fpn_s1x/model_final
metric: COCO metric: COCO
num_classes: 81 num_classes: 81
......
configs/res2net/mask_rcnn_res2net50_vb_26w_4s_fpn_2x.yml
浏览文件 @ 9630ebb4
...@@ -7,7 +7,7 @@ log_smooth_window: 20 ...@@ -7,7 +7,7 @@ log_smooth_window: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net50_26w_4s_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net50_26w_4s_pretrained.tar
metric: COCO metric: COCO
weights: output/mask_rcnn_res2net50_vb_26w_4s_fpn_2x/model_final/ weights: output/mask_rcnn_res2net50_vb_26w_4s_fpn_2x/model_final
num_classes: 81 num_classes: 81
MaskRCNN: MaskRCNN:
......
configs/res2net/mask_rcnn_res2net50_vd_26w_4s_fpn_2x.yml
浏览文件 @ 9630ebb4
...@@ -7,7 +7,7 @@ log_smooth_window: 20 ...@@ -7,7 +7,7 @@ log_smooth_window: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net50_vd_26w_4s_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net50_vd_26w_4s_pretrained.tar
metric: COCO metric: COCO
weights: output/mask_rcnn_res2net50_vd_26w_4s_fpn_2x/model_final/ weights: output/mask_rcnn_res2net50_vd_26w_4s_fpn_2x/model_final
num_classes: 81 num_classes: 81
MaskRCNN: MaskRCNN:
......
configs/res2net/mask_rcnn_res2net50_vd_26w_4s_fpn_dcnv2_1x.yml
浏览文件 @ 9630ebb4
...@@ -7,7 +7,7 @@ log_smooth_window: 20 ...@@ -7,7 +7,7 @@ log_smooth_window: 20
save_dir: output save_dir: output
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net50_vd_26w_4s_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net50_vd_26w_4s_pretrained.tar
metric: COCO metric: COCO
weights: output/mask_rcnn_res2net50_vd_26w_4s_fpn_dcnv2_1x/model_final/ weights: output/mask_rcnn_res2net50_vd_26w_4s_fpn_dcnv2_1x/model_final
num_classes: 81 num_classes: 81
MaskRCNN: MaskRCNN:
......
configs/ssd/ssd_mobilenet_v1_voc.yml
浏览文件 @ 9630ebb4
...@@ -7,7 +7,7 @@ log_smooth_window: 1 ...@@ -7,7 +7,7 @@ log_smooth_window: 1
metric: VOC metric: VOC
map_type: 11point map_type: 11point
save_dir: output save_dir: output
weights: output/ssd_mobilenet_v1_voc/model_final/ weights: output/ssd_mobilenet_v1_voc/model_final
# 20(label_class) + 1(background) # 20(label_class) + 1(background)
num_classes: 21 num_classes: 21
......
configs/ssd/ssd_vgg16_300_voc.yml
浏览文件 @ 9630ebb4
...@@ -8,7 +8,7 @@ metric: VOC ...@@ -8,7 +8,7 @@ metric: VOC
map_type: 11point map_type: 11point
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/VGG16_caffe_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/VGG16_caffe_pretrained.tar
save_dir: output save_dir: output
weights: output/ssd_vgg16_300_voc/model_final/ weights: output/ssd_vgg16_300_voc/model_final
# 20(label_class) + 1(background) # 20(label_class) + 1(background)
num_classes: 21 num_classes: 21
......
configs/ssd/ssd_vgg16_512_voc.yml
浏览文件 @ 9630ebb4
...@@ -8,7 +8,7 @@ metric: VOC ...@@ -8,7 +8,7 @@ metric: VOC
map_type: 11point map_type: 11point
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/VGG16_caffe_pretrained.tar pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/VGG16_caffe_pretrained.tar
save_dir: output save_dir: output
weights: output/ssd_vgg16_512_voc/model_final/ weights: output/ssd_vgg16_512_voc/model_final
# 20(label_class) + 1(background) # 20(label_class) + 1(background)
num_classes: 21 num_classes: 21
......
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或