├── Illustration ├── COCO.png ├── Cityscapes.png ├── Overview.png └── init ├── LICENSE ├── README.md ├── configs ├── _base_ │ ├── datasets │ │ ├── cityscapes_detection.py │ │ ├── cityscapes_instance.py │ │ ├── coco_detection.py │ │ ├── coco_instance.py │ │ ├── coco_instance_semantic.py │ │ ├── deepfashion.py │ │ ├── lvis_instance.py │ │ ├── voc0712.py │ │ └── wider_face.py │ ├── default_runtime.py │ ├── models │ │ ├── cascade_mask_rcnn_r50_fpn.py │ │ ├── cascade_rcnn_r50_fpn.py │ │ ├── fast_rcnn_r50_fpn.py │ │ ├── faster_rcnn_r50_caffe_c4.py │ │ ├── faster_rcnn_r50_fpn.py │ │ ├── mask_rcnn_r50_caffe_c4.py │ │ ├── mask_rcnn_r50_fpn.py │ │ ├── retinanet_r50_fpn.py │ │ ├── rpn_r50_caffe_c4.py │ │ ├── rpn_r50_fpn.py │ │ └── ssd300.py │ └── schedules │ │ ├── schedule_1x.py │ │ ├── schedule_20e.py │ │ └── schedule_2x.py ├── albu_example │ ├── README.md │ └── mask_rcnn_r50_fpn_albu_1x_coco.py ├── atss │ ├── README.md │ └── atss_r50_fpn_1x_coco.py ├── carafe │ ├── README.md │ ├── faster_rcnn_r50_fpn_carafe_1x_coco.py │ └── mask_rcnn_r50_fpn_carafe_1x_coco.py ├── cascade_rcnn │ ├── README.md │ ├── cascade_mask_rcnn_r101_caffe_fpn_1x_coco.py │ ├── cascade_mask_rcnn_r101_fpn_1x_coco.py │ ├── cascade_mask_rcnn_r101_fpn_20e_coco.py │ ├── cascade_mask_rcnn_r50_caffe_fpn_1x_coco.py │ ├── cascade_mask_rcnn_r50_fpn_1x_coco.py │ ├── cascade_mask_rcnn_r50_fpn_20e_coco.py │ ├── cascade_mask_rcnn_x101_32x4d_fpn_1x_coco.py │ ├── cascade_mask_rcnn_x101_32x4d_fpn_20e_coco.py │ ├── cascade_mask_rcnn_x101_64x4d_fpn_1x_coco.py │ ├── cascade_mask_rcnn_x101_64x4d_fpn_20e_coco.py │ ├── cascade_rcnn_r101_caffe_fpn_1x_coco.py │ ├── cascade_rcnn_r101_fpn_1x_coco.py │ ├── cascade_rcnn_r101_fpn_20e_coco.py │ ├── cascade_rcnn_r50_caffe_fpn_1x_coco.py │ ├── cascade_rcnn_r50_fpn_1x_coco.py │ ├── cascade_rcnn_r50_fpn_20e_coco.py │ ├── cascade_rcnn_x101_32x4d_fpn_1x_coco.py │ ├── cascade_rcnn_x101_32x4d_fpn_20e_coco.py │ ├── cascade_rcnn_x101_64x4d_fpn_1x_coco.py │ └── cascade_rcnn_x101_64x4d_fpn_20e_coco.py ├── cityscapes │ ├── README.md │ ├── faster_rcnn_r50_fpn_1x_cityscapes.py │ ├── mask_rcnn_r50_fpn_1x_cityscapes.py │ ├── mask_rcnn_r50_fpn_1x_cityscapes_nl_1.py │ └── mask_rcnn_r50_fpn_1x_cityscapes_nl_2.py ├── dcn │ ├── README.md │ ├── cascade_mask_rcnn_r101_fpn_dconv_c3-c5_1x_coco.py │ ├── cascade_mask_rcnn_r50_fpn_dconv_c3-c5_1x_coco.py │ ├── cascade_mask_rcnn_x101_32x4d_fpn_dconv_c3-c5_1x_coco.py │ ├── cascade_rcnn_r101_fpn_dconv_c3-c5_1x_coco.py │ ├── cascade_rcnn_r50_fpn_dconv_c3-c5_1x_coco.py │ ├── faster_rcnn_r101_fpn_dconv_c3-c5_1x_coco.py │ ├── faster_rcnn_r50_fpn_dconv_c3-c5_1x_coco.py │ ├── faster_rcnn_r50_fpn_dpool_1x_coco.py │ ├── faster_rcnn_r50_fpn_mdconv_c3-c5_1x_coco.py │ ├── faster_rcnn_r50_fpn_mdconv_c3-c5_group4_1x_coco.py │ ├── faster_rcnn_r50_fpn_mdpool_1x_coco.py │ ├── faster_rcnn_x101_32x4d_fpn_dconv_c3-c5_1x_coco.py │ ├── mask_rcnn_r101_fpn_dconv_c3-c5_1x_coco.py │ ├── mask_rcnn_r50_fpn_dconv_c3-c5_1x_coco.py │ └── mask_rcnn_r50_fpn_mdconv_c3-c5_1x_coco.py ├── deepfashion │ ├── README.md │ └── mask_rcnn_r50_fpn_15e_deepfashion.py ├── detectors │ ├── README.md │ ├── cascade_rcnn_r50_rfp_1x_coco.py │ ├── cascade_rcnn_r50_sac_1x_coco.py │ ├── detectors_cascade_rcnn_r50_1x_coco.py │ ├── detectors_htc_r50_1x_coco.py │ ├── htc_r50_rfp_1x_coco.py │ └── htc_r50_sac_1x_coco.py ├── double_heads │ ├── README.md │ └── dh_faster_rcnn_r50_fpn_1x_coco.py ├── dynamic_rcnn │ ├── README.md │ └── dynamic_rcnn_r50_fpn_1x.py ├── empirical_attention │ ├── README.md │ ├── faster_rcnn_r50_fpn_attention_0010_1x_coco.py │ ├── faster_rcnn_r50_fpn_attention_0010_dcn_1x_coco.py │ ├── faster_rcnn_r50_fpn_attention_1111_1x_coco.py │ └── faster_rcnn_r50_fpn_attention_1111_dcn_1x_coco.py ├── fast_rcnn │ ├── README.md │ ├── fast_rcnn_r101_caffe_fpn_1x_coco.py │ ├── fast_rcnn_r101_fpn_1x_coco.py │ ├── fast_rcnn_r101_fpn_2x_coco.py │ ├── fast_rcnn_r50_caffe_fpn_1x_coco.py │ ├── fast_rcnn_r50_fpn_1x_coco.py │ └── fast_rcnn_r50_fpn_2x_coco.py ├── faster_rcnn │ ├── README.md │ ├── faster_rcnn_r101_caffe_fpn_1x_coco.py │ ├── faster_rcnn_r101_fpn_1x_coco.py │ ├── faster_rcnn_r101_fpn_2x_coco.py │ ├── faster_rcnn_r50_caffe_c4_1x_coco.py │ ├── faster_rcnn_r50_caffe_fpn_1x_coco.py │ ├── faster_rcnn_r50_caffe_fpn_mstrain_1x_coco.py │ ├── faster_rcnn_r50_caffe_fpn_mstrain_2x_coco.py │ ├── faster_rcnn_r50_caffe_fpn_mstrain_3x_coco.py │ ├── faster_rcnn_r50_fpn_1x_coco-person-bicycle-car.py │ ├── faster_rcnn_r50_fpn_1x_coco-person.py │ ├── faster_rcnn_r50_fpn_1x_coco.py │ ├── faster_rcnn_r50_fpn_2x_coco.py │ ├── faster_rcnn_r50_fpn_bounded_iou_1x_coco.py │ ├── faster_rcnn_r50_fpn_giou_1x_coco.py │ ├── faster_rcnn_r50_fpn_iou_1x_coco.py │ ├── faster_rcnn_r50_fpn_ohem_1x_coco.py │ ├── faster_rcnn_r50_fpn_soft_nms_1x_coco.py │ ├── faster_rcnn_x101_32x4d_fpn_1x_coco.py │ ├── faster_rcnn_x101_32x4d_fpn_2x_coco.py │ ├── faster_rcnn_x101_64x4d_fpn_1x_coco.py │ └── faster_rcnn_x101_64x4d_fpn_2x_coco.py ├── fcos │ ├── README.md │ ├── fcos_center-normbbox-centeronreg-giou_r50_caffe_fpn_gn-head_4x4_1x_coco.py │ ├── fcos_center-normbbox-centeronreg-giou_r50_caffe_fpn_gn-head_dcn_4x4_1x_coco.py │ ├── fcos_center_r50_caffe_fpn_gn-head_4x4_1x_coco.py │ ├── fcos_r101_caffe_fpn_gn-head_4x4_1x_coco.py │ ├── fcos_r101_caffe_fpn_gn-head_4x4_2x_coco.py │ ├── fcos_r101_caffe_fpn_gn-head_mstrain_640-800_4x4_2x_coco.py │ ├── fcos_r50_caffe_fpn_4x4_1x_coco.py │ ├── fcos_r50_caffe_fpn_gn-head_4x4_1x_coco.py │ ├── fcos_r50_caffe_fpn_gn-head_4x4_2x_coco.py │ ├── fcos_r50_caffe_fpn_gn-head_mstrain_640-800_4x4_2x_coco.py │ └── fcos_x101_64x4d_fpn_gn-head_mstrain_640-800_4x2_2x_coco.py ├── foveabox │ ├── README.md │ ├── fovea_align_r101_fpn_gn-head_4x4_2x_coco.py │ ├── fovea_align_r101_fpn_gn-head_mstrain_640-800_4x4_2x_coco.py │ ├── fovea_align_r50_fpn_gn-head_4x4_2x_coco.py │ ├── fovea_align_r50_fpn_gn-head_mstrain_640-800_4x4_2x_coco.py │ ├── fovea_r101_fpn_4x4_1x_coco.py │ ├── fovea_r101_fpn_4x4_2x_coco.py │ ├── fovea_r50_fpn_4x4_1x_coco.py │ └── fovea_r50_fpn_4x4_2x_coco.py ├── fp16 │ ├── README.md │ ├── faster_rcnn_r50_fpn_fp16_1x_coco.py │ ├── mask_rcnn_r50_fpn_fp16_1x_coco.py │ └── retinanet_r50_fpn_fp16_1x_coco.py ├── free_anchor │ ├── README.md │ ├── retinanet_free_anchor_r101_fpn_1x_coco.py │ ├── retinanet_free_anchor_r50_fpn_1x_coco.py │ └── retinanet_free_anchor_x101_32x4d_fpn_1x_coco.py ├── fsaf │ ├── README.md │ ├── fsaf_r101_fpn_1x_coco.py │ ├── fsaf_r50_fpn_1x_coco.py │ └── fsaf_x101_64x4d_fpn_1x_coco.py ├── gcnet │ ├── README.md │ ├── cascade_mask_rcnn_x101_32x4d_fpn_syncbn-backbone_1x_coco.py │ ├── cascade_mask_rcnn_x101_32x4d_fpn_syncbn-backbone_dconv_c3-c5_1x_coco.py │ ├── cascade_mask_rcnn_x101_32x4d_fpn_syncbn-backbone_dconv_c3-c5_r16_gcb_c3-c5_1x_coco.py │ ├── cascade_mask_rcnn_x101_32x4d_fpn_syncbn-backbone_dconv_c3-c5_r4_gcb_c3-c5_1x_coco.py │ ├── cascade_mask_rcnn_x101_32x4d_fpn_syncbn-backbone_r16_gcb_c3-c5_1x_coco.py │ ├── cascade_mask_rcnn_x101_32x4d_fpn_syncbn-backbone_r4_gcb_c3-c5_1x_coco.py │ ├── mask_rcnn_r101_fpn_r16_gcb_c3-c5_1x_coco.py │ ├── mask_rcnn_r101_fpn_r4_gcb_c3-c5_1x_coco.py │ ├── mask_rcnn_r101_fpn_syncbn-backbone_1x_coco.py │ ├── mask_rcnn_r101_fpn_syncbn-backbone_r16_gcb_c3-c5_1x_coco.py │ ├── mask_rcnn_r101_fpn_syncbn-backbone_r4_gcb_c3-c5_1x_coco.py │ ├── mask_rcnn_r50_fpn_r16_gcb_c3-c5_1x_coco.py │ ├── mask_rcnn_r50_fpn_r4_gcb_c3-c5_1x_coco.py │ ├── mask_rcnn_r50_fpn_syncbn-backbone_1x_coco.py │ ├── mask_rcnn_r50_fpn_syncbn-backbone_r16_gcb_c3-c5_1x_coco.py │ ├── mask_rcnn_r50_fpn_syncbn-backbone_r4_gcb_c3-c5_1x_coco.py │ ├── mask_rcnn_x101_32x4d_fpn_syncbn-backbone_1x_coco.py │ ├── mask_rcnn_x101_32x4d_fpn_syncbn-backbone_r16_gcb_c3-c5_1x_coco.py │ └── mask_rcnn_x101_32x4d_fpn_syncbn-backbone_r4_gcb_c3-c5_1x_coco.py ├── gfl │ ├── README.md │ ├── gfl_r101_fpn_dconv_c3-c5_mstrain_2x_coco.py │ ├── gfl_r101_fpn_mstrain_2x_coco.py │ ├── gfl_r50_fpn_1x_coco.py │ ├── gfl_r50_fpn_mstrain_2x_coco.py │ ├── gfl_x101_32x4d_fpn_dconv_c4-c5_mstrain_2x_coco.py │ └── gfl_x101_32x4d_fpn_mstrain_2x_coco.py ├── ghm │ ├── README.md │ ├── retinanet_ghm_r101_fpn_1x_coco.py │ ├── retinanet_ghm_r50_fpn_1x_coco.py │ ├── retinanet_ghm_x101_32x4d_fpn_1x_coco.py │ └── retinanet_ghm_x101_64x4d_fpn_1x_coco.py ├── gn+ws │ ├── README.md │ ├── faster_rcnn_r101_fpn_gn_ws-all_1x_coco.py │ ├── faster_rcnn_r50_fpn_gn_ws-all_1x_coco.py │ ├── faster_rcnn_x101_32x4d_fpn_gn_ws-all_1x_coco.py │ ├── faster_rcnn_x50_32x4d_fpn_gn_ws-all_1x_coco.py │ ├── mask_rcnn_r101_fpn_gn_ws-all_20_23_24e_coco.py │ ├── mask_rcnn_r101_fpn_gn_ws-all_2x_coco.py │ ├── mask_rcnn_r50_fpn_gn_ws-all_20_23_24e_coco.py │ ├── mask_rcnn_r50_fpn_gn_ws-all_2x_coco.py │ ├── mask_rcnn_x101_32x4d_fpn_gn_ws-all_20_23_24e_coco.py │ ├── mask_rcnn_x101_32x4d_fpn_gn_ws-all_2x_coco.py │ ├── mask_rcnn_x50_32x4d_fpn_gn_ws-all_20_23_24e_coco.py │ └── mask_rcnn_x50_32x4d_fpn_gn_ws-all_2x_coco.py ├── gn │ ├── README.md │ ├── mask_rcnn_r101_fpn_gn-all_2x_coco.py │ ├── mask_rcnn_r101_fpn_gn-all_3x_coco.py │ ├── mask_rcnn_r50_fpn_gn-all_2x_coco.py │ ├── mask_rcnn_r50_fpn_gn-all_3x_coco.py │ ├── mask_rcnn_r50_fpn_gn-all_contrib_2x_coco.py │ └── mask_rcnn_r50_fpn_gn-all_contrib_3x_coco.py ├── grid_rcnn │ ├── README.md │ ├── grid_rcnn_r101_fpn_gn-head_2x_coco.py │ ├── grid_rcnn_r50_fpn_gn-head_1x_coco.py │ ├── grid_rcnn_r50_fpn_gn-head_2x_coco.py │ ├── grid_rcnn_x101_32x4d_fpn_gn-head_2x_coco.py │ └── grid_rcnn_x101_64x4d_fpn_gn-head_2x_coco.py ├── groie │ ├── README.md │ ├── faster_rcnn_r50_fpn_groie_1x_coco.py │ ├── grid_rcnn_r50_fpn_gn-head_groie_1x_coco.py │ ├── mask_rcnn_r101_fpn_syncbn-backbone_r4_gcb_c3-c5_groie_1x_coco.py │ ├── mask_rcnn_r50_fpn_groie_1x_coco.py │ └── mask_rcnn_r50_fpn_syncbn-backbone_r4_gcb_c3-c5_groie_1x_coco.py ├── guided_anchoring │ ├── README.md │ ├── ga_fast_r50_caffe_fpn_1x_coco.py │ ├── ga_faster_r101_caffe_fpn_1x_coco.py │ ├── ga_faster_r50_caffe_fpn_1x_coco.py │ ├── ga_faster_r50_fpn_1x_coco.py │ ├── ga_faster_x101_32x4d_fpn_1x_coco.py │ ├── ga_faster_x101_64x4d_fpn_1x_coco.py │ ├── ga_retinanet_r101_caffe_fpn_1x_coco.py │ ├── ga_retinanet_r101_caffe_fpn_mstrain_2x.py │ ├── ga_retinanet_r50_caffe_fpn_1x_coco.py │ ├── ga_retinanet_r50_fpn_1x_coco.py │ ├── ga_retinanet_x101_32x4d_fpn_1x_coco.py │ ├── ga_retinanet_x101_64x4d_fpn_1x_coco.py │ ├── ga_rpn_r101_caffe_fpn_1x_coco.py │ ├── ga_rpn_r50_caffe_fpn_1x_coco.py │ ├── ga_rpn_r50_fpn_1x_coco.py │ ├── ga_rpn_x101_32x4d_fpn_1x_coco.py │ └── ga_rpn_x101_64x4d_fpn_1x_coco.py ├── hrnet │ ├── README.md │ ├── cascade_mask_rcnn_hrnetv2p_w18_20e_coco.py │ ├── cascade_mask_rcnn_hrnetv2p_w32_20e_coco.py │ ├── cascade_mask_rcnn_hrnetv2p_w40_20e_coco.py │ ├── cascade_rcnn_hrnetv2p_w18_20e_coco.py │ ├── cascade_rcnn_hrnetv2p_w32_20e_coco.py │ ├── cascade_rcnn_hrnetv2p_w40_20e_coco.py │ ├── faster_rcnn_hrnetv2p_w18_1x_coco.py │ ├── faster_rcnn_hrnetv2p_w18_2x_coco.py │ ├── faster_rcnn_hrnetv2p_w32_1x_coco.py │ ├── faster_rcnn_hrnetv2p_w32_2x_coco.py │ ├── faster_rcnn_hrnetv2p_w40_1x_coco.py │ ├── faster_rcnn_hrnetv2p_w40_2x_coco.py │ ├── fcos_hrnetv2p_w18_gn-head_4x4_1x_coco.py │ ├── fcos_hrnetv2p_w18_gn-head_4x4_2x_coco.py │ ├── fcos_hrnetv2p_w18_gn-head_mstrain_640-800_4x4_2x_coco.py │ ├── fcos_hrnetv2p_w32_gn-head_4x4_1x_coco.py │ ├── fcos_hrnetv2p_w32_gn-head_4x4_2x_coco.py │ ├── fcos_hrnetv2p_w32_gn-head_mstrain_640-800_4x4_2x_coco.py │ ├── fcos_hrnetv2p_w40_gn-head_mstrain_640-800_4x4_2x_coco.py │ ├── htc_hrnetv2p_w18_20e_coco.py │ ├── htc_hrnetv2p_w32_20e_coco.py │ ├── htc_hrnetv2p_w40_20e_coco.py │ ├── htc_hrnetv2p_w40_28e_coco.py │ ├── htc_x101_64x4d_fpn_16x1_28e_coco.py │ ├── mask_rcnn_hrnetv2p_w18_1x_coco.py │ ├── mask_rcnn_hrnetv2p_w18_2x_coco.py │ ├── mask_rcnn_hrnetv2p_w32_1x_coco.py │ ├── mask_rcnn_hrnetv2p_w32_2x_coco.py │ ├── mask_rcnn_hrnetv2p_w40_1x_coco.py │ └── mask_rcnn_hrnetv2p_w40_2x_coco.py ├── htc │ ├── README.md │ ├── htc_r101_fpn_20e_coco.py │ ├── htc_r50_fpn_1x_coco.py │ ├── htc_r50_fpn_20e_coco.py │ ├── htc_without_semantic_r50_fpn_1x_coco.py │ ├── htc_x101_32x4d_fpn_16x1_20e_coco.py │ ├── htc_x101_64x4d_fpn_16x1_20e_coco.py │ └── htc_x101_64x4d_fpn_dconv_c3-c5_mstrain_400_1400_16x1_20e_coco.py ├── instaboost │ ├── README.md │ ├── cascade_mask_rcnn_r101_fpn_instaboost_4x_coco.py │ ├── cascade_mask_rcnn_r50_fpn_instaboost_4x_coco.py │ ├── cascade_mask_rcnn_x101_64x4d_fpn_instaboost_4x_coco.py │ ├── mask_rcnn_r101_fpn_instaboost_4x_coco.py │ ├── mask_rcnn_r50_fpn_instaboost_4x_coco.py │ └── mask_rcnn_x101_64x4d_fpn_instaboost_4x_coco.py ├── legacy_1.x │ ├── README.md │ ├── cascade_mask_rcnn_r50_fpn_1x_coco_v1.py │ ├── faster_rcnn_r50_fpn_1x_coco_v1.py │ ├── mask_rcnn_r50_fpn_1x_coco_v1.py │ ├── retinanet_r50_caffe_fpn_1x_coco_v1.py │ ├── retinanet_r50_fpn_1x_coco_v1.py │ └── ssd300_coco_v1.py ├── libra_rcnn │ ├── README.md │ ├── libra_fast_rcnn_r50_fpn_1x_coco.py │ ├── libra_faster_rcnn_r101_fpn_1x_coco.py │ ├── libra_faster_rcnn_r50_fpn_1x_coco.py │ ├── libra_faster_rcnn_x101_64x4d_fpn_1x_coco.py │ └── libra_retinanet_r50_fpn_1x_coco.py ├── lvis │ ├── README.md │ ├── mask_rcnn_r101_fpn_sample1e-3_mstrain_2x_lvis.py │ ├── mask_rcnn_r50_fpn_sample1e-3_mstrain_2x_lvis.py │ ├── mask_rcnn_x101_32x4d_fpn_sample1e-3_mstrain_2x_lvis.py │ └── mask_rcnn_x101_64x4d_fpn_sample1e-3_mstrain_2x_lvis.py ├── mask_rcnn │ ├── README.md │ ├── mask_rcnn_r101_caffe_fpn_1x_coco.py │ ├── mask_rcnn_r101_fpn_1x_coco.py │ ├── mask_rcnn_r101_fpn_2x_coco.py │ ├── mask_rcnn_r50_caffe_c4_1x_coco.py │ ├── mask_rcnn_r50_caffe_fpn_1x_coco.py │ ├── mask_rcnn_r50_caffe_fpn_mstrain-poly_1x_coco.py │ ├── mask_rcnn_r50_caffe_fpn_mstrain-poly_2x_coco.py │ ├── mask_rcnn_r50_caffe_fpn_mstrain-poly_3x_coco.py │ ├── mask_rcnn_r50_caffe_fpn_mstrain_1x_coco.py │ ├── mask_rcnn_r50_caffe_fpn_poly_1x_coco_v1.py │ ├── mask_rcnn_r50_fpn_1x_coco.py │ ├── mask_rcnn_r50_fpn_2x_coco.py │ ├── mask_rcnn_r50_fpn_poly_1x_coco.py │ ├── mask_rcnn_x101_32x4d_fpn_1x_coco.py │ ├── mask_rcnn_x101_32x4d_fpn_2x_coco.py │ ├── mask_rcnn_x101_32x8d_fpn_1x_coco.py │ ├── mask_rcnn_x101_32x8d_fpn_mstrain-poly_1x_coco.py │ ├── mask_rcnn_x101_32x8d_fpn_mstrain-poly_3x_coco.py │ ├── mask_rcnn_x101_64x4d_fpn_1x_coco.py │ └── mask_rcnn_x101_64x4d_fpn_2x_coco.py ├── ms_rcnn │ ├── README.md │ ├── ms_rcnn_r101_caffe_fpn_1x_coco.py │ ├── ms_rcnn_r101_caffe_fpn_2x_coco.py │ ├── ms_rcnn_r50_caffe_fpn_1x_coco.py │ ├── ms_rcnn_r50_caffe_fpn_2x_coco.py │ ├── ms_rcnn_r50_fpn_1x_coco.py │ ├── ms_rcnn_x101_32x4d_fpn_1x_coco.py │ ├── ms_rcnn_x101_64x4d_fpn_1x_coco.py │ └── ms_rcnn_x101_64x4d_fpn_2x_coco.py ├── nas_fcos │ ├── README.md │ ├── nas_fcos_fcoshead_r50_caffe_fpn_gn-head_4x4_1x_coco.py │ └── nas_fcos_nashead_r50_caffe_fpn_gn-head_4x4_1x_coco.py ├── nas_fpn │ ├── README.md │ ├── retinanet_r50_fpn_crop640_50e_coco.py │ └── retinanet_r50_nasfpn_crop640_50e_coco.py ├── pafpn │ ├── README.md │ └── faster_rcnn_r50_pafpn_1x_coco.py ├── pascal_voc │ ├── README.md │ ├── faster_rcnn_r50_fpn_1x_voc0712.py │ ├── retinanet_r50_fpn_1x_voc0712.py │ ├── ssd300_voc0712.py │ └── ssd512_voc0712.py ├── pisa │ ├── README.md │ ├── pisa_faster_rcnn_r50_fpn_1x_coco.py │ ├── pisa_faster_rcnn_x101_32x4d_fpn_1x_coco.py │ ├── pisa_mask_rcnn_r50_fpn_1x_coco.py │ ├── pisa_mask_rcnn_x101_32x4d_fpn_1x_coco.py │ ├── pisa_retinanet_r50_fpn_1x_coco.py │ ├── pisa_retinanet_x101_32x4d_fpn_1x_coco.py │ ├── pisa_ssd300_coco.py │ └── pisa_ssd512_coco.py ├── point_rend │ ├── README.md │ ├── point_rend_r50_caffe_fpn_mstrain_1x_coco.py │ └── point_rend_r50_caffe_fpn_mstrain_3x_coco.py ├── regnet │ ├── README.md │ ├── faster_rcnn_regnetx-3.2GF_fpn_1x_coco.py │ ├── faster_rcnn_regnetx-3.2GF_fpn_2x_coco.py │ ├── faster_rcnn_regnetx-3.2GF_fpn_mstrain_3x_coco.py │ ├── mask_rcnn_regnetx-12GF_fpn_1x_coco.py │ ├── mask_rcnn_regnetx-3.2GF_fpn_1x_coco.py │ ├── mask_rcnn_regnetx-3.2GF_fpn_mdconv_c3-c5_1x_coco.py │ ├── mask_rcnn_regnetx-3.2GF_fpn_mstrain_3x_coco.py │ ├── mask_rcnn_regnetx-4GF_fpn_1x_coco.py │ ├── mask_rcnn_regnetx-6.4GF_fpn_1x_coco.py │ ├── mask_rcnn_regnetx-8GF_fpn_1x_coco.py │ ├── retinanet_regnetx-1.6GF_fpn_1x_coco.py │ ├── retinanet_regnetx-3.2GF_fpn_1x_coco.py │ └── retinanet_regnetx-800MF_fpn_1x_coco.py ├── reppoints │ ├── README.md │ ├── bbox_r50_grid_center_fpn_gn-neck+head_1x_coco.py │ ├── bbox_r50_grid_fpn_gn-neck+head_1x_coco.py │ ├── reppoints.png │ ├── reppoints_minmax_r50_fpn_gn-neck+head_1x_coco.py │ ├── reppoints_moment_r101_fpn_dconv_c3-c5_gn-neck+head_2x_coco.py │ ├── reppoints_moment_r101_fpn_gn-neck+head_2x_coco.py │ ├── reppoints_moment_r50_fpn_1x_coco.py │ ├── reppoints_moment_r50_fpn_gn-neck+head_1x_coco.py │ ├── reppoints_moment_r50_fpn_gn-neck+head_2x_coco.py │ ├── reppoints_moment_x101_fpn_dconv_c3-c5_gn-neck+head_2x_coco.py │ └── reppoints_partial_minmax_r50_fpn_gn-neck+head_1x_coco.py ├── res2net │ ├── README.md │ ├── cascade_mask_rcnn_r2_101_fpn_20e_coco.py │ ├── cascade_rcnn_r2_101_fpn_20e_coco.py │ ├── faster_rcnn_r2_101_fpn_2x_coco.py │ ├── htc_r2_101_fpn_20e_coco.py │ └── mask_rcnn_r2_101_fpn_2x_coco.py ├── retinanet │ ├── README.md │ ├── retinanet_r101_caffe_fpn_1x_coco.py │ ├── retinanet_r101_fpn_1x_coco.py │ ├── retinanet_r101_fpn_2x_coco.py │ ├── retinanet_r50_caffe_fpn_1x_coco.py │ ├── retinanet_r50_caffe_fpn_mstrain_1x_coco.py │ ├── retinanet_r50_caffe_fpn_mstrain_2x_coco.py │ ├── retinanet_r50_caffe_fpn_mstrain_3x_coco.py │ ├── retinanet_r50_fpn_1x_coco.py │ ├── retinanet_r50_fpn_2x_coco.py │ ├── retinanet_x101_32x4d_fpn_1x_coco.py │ ├── retinanet_x101_32x4d_fpn_2x_coco.py │ ├── retinanet_x101_64x4d_fpn_1x_coco.py │ └── retinanet_x101_64x4d_fpn_2x_coco.py ├── rpn │ ├── README.md │ ├── rpn_r101_caffe_fpn_1x_coco.py │ ├── rpn_r101_fpn_1x_coco.py │ ├── rpn_r101_fpn_2x_coco.py │ ├── rpn_r50_caffe_c4_1x_coco.py │ ├── rpn_r50_caffe_fpn_1x_coco.py │ ├── rpn_r50_fpn_1x_coco.py │ ├── rpn_r50_fpn_2x_coco.py │ ├── rpn_x101_32x4d_fpn_1x_coco.py │ ├── rpn_x101_32x4d_fpn_2x_coco.py │ ├── rpn_x101_64x4d_fpn_1x_coco.py │ └── rpn_x101_64x4d_fpn_2x_coco.py ├── scratch │ ├── README.md │ ├── faster_rcnn_r50_fpn_gn-all_scratch_6x_coco.py │ └── mask_rcnn_r50_fpn_gn-all_scratch_6x_coco.py ├── ssd │ ├── README.md │ ├── ssd300_coco.py │ └── ssd512_coco.py └── wider_face │ ├── README.md │ └── ssd300_wider_face.py ├── demo ├── coco_test_12510.jpg ├── corruptions_sev_3.png ├── data_pipeline.png ├── demo.jpg ├── image_demo.py ├── inference_demo.ipynb ├── loss_curve.png ├── mmdet_inference_colab.ipynb └── webcam_demo.py ├── docker └── Dockerfile ├── docs ├── Makefile ├── api.rst ├── changelog.md ├── compatibility.md ├── conf.py ├── config.md ├── getting_started.md ├── index.rst ├── install.md ├── make.bat ├── model_zoo.md ├── projects.md ├── robustness_benchmarking.md └── tutorials │ ├── data_pipeline.md │ ├── finetune.md │ ├── new_dataset.md │ └── new_modules.md ├── mmdet ├── VERSION ├── __init__.py ├── apis │ ├── __init__.py │ ├── inference.py │ ├── test.py │ └── train.py ├── core │ ├── __init__.py │ ├── anchor │ │ ├── __init__.py │ │ ├── anchor_generator.py │ │ ├── builder.py │ │ ├── point_generator.py │ │ └── utils.py │ ├── bbox │ │ ├── __init__.py │ │ ├── assigners │ │ │ ├── __init__.py │ │ │ ├── approx_max_iou_assigner.py │ │ │ ├── assign_result.py │ │ │ ├── atss_assigner.py │ │ │ ├── base_assigner.py │ │ │ ├── center_region_assigner.py │ │ │ ├── max_iou_assigner.py │ │ │ └── point_assigner.py │ │ ├── builder.py │ │ ├── coder │ │ │ ├── __init__.py │ │ │ ├── base_bbox_coder.py │ │ │ ├── delta_xywh_bbox_coder.py │ │ │ ├── legacy_delta_xywh_bbox_coder.py │ │ │ ├── pseudo_bbox_coder.py │ │ │ └── tblr_bbox_coder.py │ │ ├── demodata.py │ │ ├── iou_calculators │ │ │ ├── __init__.py │ │ │ ├── builder.py │ │ │ └── iou2d_calculator.py │ │ ├── samplers │ │ │ ├── __init__.py │ │ │ ├── base_sampler.py │ │ │ ├── combined_sampler.py │ │ │ ├── instance_balanced_pos_sampler.py │ │ │ ├── iou_balanced_neg_sampler.py │ │ │ ├── ohem_sampler.py │ │ │ ├── pseudo_sampler.py │ │ │ ├── random_sampler.py │ │ │ ├── sampling_result.py │ │ │ └── score_hlr_sampler.py │ │ └── transforms.py │ ├── evaluation │ │ ├── __init__.py │ │ ├── bbox_overlaps.py │ │ ├── class_names.py │ │ ├── eval_hooks.py │ │ ├── mean_ap.py │ │ └── recall.py │ ├── fp16 │ │ ├── __init__.py │ │ ├── decorators.py │ │ ├── hooks.py │ │ └── utils.py │ ├── mask │ │ ├── __init__.py │ │ ├── mask_target.py │ │ ├── structures.py │ │ └── utils.py │ ├── post_processing │ │ ├── __init__.py │ │ ├── bbox_nms.py │ │ └── merge_augs.py │ └── utils │ │ ├── __init__.py │ │ ├── dist_utils.py │ │ └── misc.py ├── datasets │ ├── __init__.py │ ├── builder.py │ ├── cityscapes.py │ ├── coco.py │ ├── custom.py │ ├── dataset_wrappers.py │ ├── deepfashion.py │ ├── lvis.py │ ├── pipelines │ │ ├── __init__.py │ │ ├── auto_augment.py │ │ ├── compose.py │ │ ├── formating.py │ │ ├── instaboost.py │ │ ├── loading.py │ │ ├── test_time_aug.py │ │ └── transforms.py │ ├── samplers │ │ ├── __init__.py │ │ ├── distributed_sampler.py │ │ └── group_sampler.py │ ├── voc.py │ ├── wider_face.py │ └── xml_style.py ├── models │ ├── __init__.py │ ├── backbones │ │ ├── __init__.py │ │ ├── detectors_resnet.py │ │ ├── detectors_resnext.py │ │ ├── hourglass.py │ │ ├── hrnet.py │ │ ├── regnet.py │ │ ├── res2net.py │ │ ├── resnet.py │ │ ├── resnext.py │ │ └── ssd_vgg.py │ ├── builder.py │ ├── dense_heads │ │ ├── __init__.py │ │ ├── anchor_free_head.py │ │ ├── anchor_head.py │ │ ├── atss_head.py │ │ ├── base_dense_head.py │ │ ├── fcos_head.py │ │ ├── fovea_head.py │ │ ├── free_anchor_retina_head.py │ │ ├── fsaf_head.py │ │ ├── ga_retina_head.py │ │ ├── ga_rpn_head.py │ │ ├── gfl_head.py │ │ ├── guided_anchor_head.py │ │ ├── nasfcos_head.py │ │ ├── pisa_retinanet_head.py │ │ ├── pisa_ssd_head.py │ │ ├── reppoints_head.py │ │ ├── retina_head.py │ │ ├── retina_sepbn_head.py │ │ ├── rpn_head.py │ │ ├── rpn_test_mixin.py │ │ └── ssd_head.py │ ├── detectors │ │ ├── __init__.py │ │ ├── atss.py │ │ ├── base.py │ │ ├── cascade_rcnn.py │ │ ├── fast_rcnn.py │ │ ├── faster_rcnn.py │ │ ├── fcos.py │ │ ├── fovea.py │ │ ├── fsaf.py │ │ ├── gfl.py │ │ ├── grid_rcnn.py │ │ ├── htc.py │ │ ├── mask_rcnn.py │ │ ├── mask_scoring_rcnn.py │ │ ├── nasfcos.py │ │ ├── point_rend.py │ │ ├── reppoints_detector.py │ │ ├── retinanet.py │ │ ├── rpn.py │ │ ├── single_stage.py │ │ └── two_stage.py │ ├── losses │ │ ├── __init__.py │ │ ├── accuracy.py │ │ ├── ae_loss.py │ │ ├── balanced_l1_loss.py │ │ ├── cross_entropy_loss.py │ │ ├── focal_loss.py │ │ ├── gaussian_focal_loss.py │ │ ├── generalized_cross_entropy_loss.py │ │ ├── gfocal_loss.py │ │ ├── ghm_loss.py │ │ ├── iou_loss.py │ │ ├── mse_loss.py │ │ ├── new_combination_loss.py │ │ ├── pisa_loss.py │ │ ├── smooth_l1_loss.py │ │ ├── symmetric_cross_entropy_loss.py │ │ └── utils.py │ ├── necks │ │ ├── __init__.py │ │ ├── bfp.py │ │ ├── fpn.py │ │ ├── fpn_carafe.py │ │ ├── hrfpn.py │ │ ├── nas_fpn.py │ │ ├── nasfcos_fpn.py │ │ ├── pafpn.py │ │ └── rfp.py │ ├── roi_heads │ │ ├── __init__.py │ │ ├── base_roi_head.py │ │ ├── bbox_heads │ │ │ ├── __init__.py │ │ │ ├── bbox_head.py │ │ │ ├── convfc_bbox_head.py │ │ │ └── double_bbox_head.py │ │ ├── cascade_roi_head.py │ │ ├── double_roi_head.py │ │ ├── dynamic_roi_head.py │ │ ├── grid_roi_head.py │ │ ├── htc_roi_head.py │ │ ├── mask_heads │ │ │ ├── __init__.py │ │ │ ├── coarse_mask_head.py │ │ │ ├── fcn_mask_head.py │ │ │ ├── fused_semantic_head.py │ │ │ ├── grid_head.py │ │ │ ├── htc_mask_head.py │ │ │ ├── mask_point_head.py │ │ │ └── maskiou_head.py │ │ ├── mask_scoring_roi_head.py │ │ ├── pisa_roi_head.py │ │ ├── point_rend_roi_head.py │ │ ├── roi_extractors │ │ │ ├── __init__.py │ │ │ ├── base_roi_extractor.py │ │ │ ├── generic_roi_extractor.py │ │ │ └── single_level_roi_extractor.py │ │ ├── shared_heads │ │ │ ├── __init__.py │ │ │ └── res_layer.py │ │ ├── standard_roi_head.py │ │ └── test_mixins.py │ └── utils │ │ ├── __init__.py │ │ └── res_layer.py ├── ops │ ├── __init__.py │ ├── carafe │ │ ├── __init__.py │ │ ├── carafe.py │ │ ├── grad_check.py │ │ ├── setup.py │ │ └── src │ │ │ ├── carafe_ext.cpp │ │ │ ├── carafe_naive_ext.cpp │ │ │ └── cuda │ │ │ ├── carafe_cuda.cpp │ │ │ ├── carafe_cuda_kernel.cu │ │ │ ├── carafe_naive_cuda.cpp │ │ │ └── carafe_naive_cuda_kernel.cu │ ├── context_block.py │ ├── conv_ws.py │ ├── corner_pool │ │ ├── __init__.py │ │ ├── corner_pool.py │ │ └── src │ │ │ └── corner_pool.cpp │ ├── dcn │ │ ├── __init__.py │ │ ├── deform_conv.py │ │ ├── deform_pool.py │ │ └── src │ │ │ ├── cuda │ │ │ ├── deform_conv_cuda.cpp │ │ │ ├── deform_conv_cuda_kernel.cu │ │ │ ├── deform_pool_cuda.cpp │ │ │ └── deform_pool_cuda_kernel.cu │ │ │ ├── deform_conv_ext.cpp │ │ │ └── deform_pool_ext.cpp │ ├── generalized_attention.py │ ├── masked_conv │ │ ├── __init__.py │ │ ├── masked_conv.py │ │ └── src │ │ │ ├── cuda │ │ │ ├── masked_conv2d_cuda.cpp │ │ │ └── masked_conv2d_kernel.cu │ │ │ └── masked_conv2d_ext.cpp │ ├── merge_cells.py │ ├── nms │ │ ├── __init__.py │ │ ├── nms_wrapper.py │ │ └── src │ │ │ ├── cpu │ │ │ └── nms_cpu.cpp │ │ │ ├── cuda │ │ │ ├── nms_cuda.cpp │ │ │ └── nms_kernel.cu │ │ │ └── nms_ext.cpp │ ├── non_local.py │ ├── plugin.py │ ├── point_sample.py │ ├── roi_align │ │ ├── __init__.py │ │ ├── gradcheck.py │ │ ├── roi_align.py │ │ └── src │ │ │ ├── cpu │ │ │ └── roi_align_v2.cpp │ │ │ ├── cuda │ │ │ ├── roi_align_kernel.cu │ │ │ └── roi_align_kernel_v2.cu │ │ │ └── roi_align_ext.cpp │ ├── roi_pool │ │ ├── __init__.py │ │ ├── gradcheck.py │ │ ├── roi_pool.py │ │ └── src │ │ │ ├── cuda │ │ │ └── roi_pool_kernel.cu │ │ │ └── roi_pool_ext.cpp │ ├── saconv.py │ ├── sigmoid_focal_loss │ │ ├── __init__.py │ │ ├── sigmoid_focal_loss.py │ │ └── src │ │ │ ├── cuda │ │ │ └── sigmoid_focal_loss_cuda.cu │ │ │ └── sigmoid_focal_loss_ext.cpp │ ├── utils │ │ ├── __init__.py │ │ └── src │ │ │ └── compiling_info.cpp │ └── wrappers.py └── utils │ ├── __init__.py │ ├── collect_env.py │ ├── contextmanagers.py │ ├── logger.py │ ├── profiling.py │ └── util_mixins.py ├── noisy_labels_AN_COCO.py ├── noisy_labels_AN_Cityscapes.py ├── noisy_labels_AN_VOC.py ├── noisy_labels_SN_COCO.py ├── noisy_labels_SN_VOC.py ├── noisy_labels_SN_cityscapes.py ├── pytest.ini ├── requirements.txt ├── requirements ├── build.txt ├── docs.txt ├── optional.txt ├── readthedocs.txt ├── runtime.txt └── tests.txt ├── setup.py ├── tests ├── async_benchmark.py ├── data │ ├── color.jpg │ └── gray.jpg ├── test_anchor.py ├── test_assigner.py ├── test_async.py ├── test_config.py ├── test_data │ ├── test_dataset.py │ ├── test_formatting.py │ ├── test_loading.py │ ├── test_models_aug_test.py │ ├── test_sampler.py │ └── test_transform.py ├── test_fp16.py ├── test_masks.py ├── test_models │ ├── test_backbones.py │ ├── test_forward.py │ ├── test_heads.py │ ├── test_losses.py │ ├── test_necks.py │ ├── test_pisa_heads.py │ └── test_roi_extractor.py └── test_ops │ ├── test_corner_pool.py │ ├── test_merge_cells.py │ ├── test_nms.py │ ├── test_soft_nms.py │ └── test_wrappers.py └── tools ├── analyze_logs.py ├── benchmark.py ├── browse_dataset.py ├── coco_error_analysis.py ├── convert_datasets ├── cityscapes.py └── pascal_voc.py ├── detectron2pytorch.py ├── dist_test.sh ├── dist_train.sh ├── fuse_conv_bn.py ├── get_flops.py ├── print_config.py ├── publish_model.py ├── pytorch2onnx.py ├── regnet2mmdet.py ├── robustness_eval.py ├── slurm_test.sh ├── slurm_train.sh ├── test.py ├── test_robustness.py ├── train.py └── upgrade_model_version.py /Illustration/COCO.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/longrongyang/LNCIS/6b0ad08b79e0b372ae90cba7a31db00d23f43b3d/Illustration/COCO.png -------------------------------------------------------------------------------- /Illustration/Cityscapes.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/longrongyang/LNCIS/6b0ad08b79e0b372ae90cba7a31db00d23f43b3d/Illustration/Cityscapes.png -------------------------------------------------------------------------------- /Illustration/Overview.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/longrongyang/LNCIS/6b0ad08b79e0b372ae90cba7a31db00d23f43b3d/Illustration/Overview.png -------------------------------------------------------------------------------- /Illustration/init: -------------------------------------------------------------------------------- 1 | 2 | -------------------------------------------------------------------------------- /configs/_base_/datasets/lvis_instance.py: -------------------------------------------------------------------------------- 1 | _base_ = 'coco_instance.py' 2 | dataset_type = 'LVISDataset' 3 | data_root = 'data/lvis/' 4 | data = dict( 5 | samples_per_gpu=2, 6 | workers_per_gpu=2, 7 | train=dict( 8 | type='ClassBalancedDataset', 9 | oversample_thr=1e-3, 10 | dataset=dict( 11 | type=dataset_type, 12 | ann_file=data_root + 'annotations/lvis_v0.5_train.json', 13 | img_prefix=data_root + 'train2017/')), 14 | val=dict( 15 | type=dataset_type, 16 | ann_file=data_root + 'annotations/lvis_v0.5_val.json', 17 | img_prefix=data_root + 'val2017/'), 18 | test=dict( 19 | type=dataset_type, 20 | ann_file=data_root + 'annotations/lvis_v0.5_val.json', 21 | img_prefix=data_root + 'val2017/')) 22 | evaluation = dict(metric=['bbox', 'segm']) 23 | -------------------------------------------------------------------------------- /configs/_base_/default_runtime.py: -------------------------------------------------------------------------------- 1 | checkpoint_config = dict(interval=1) 2 | # yapf:disable 3 | log_config = dict( 4 | interval=50, 5 | hooks=[ 6 | dict(type='TextLoggerHook'), 7 | # dict(type='TensorboardLoggerHook') 8 | ]) 9 | # yapf:enable 10 | dist_params = dict(backend='nccl') 11 | log_level = 'INFO' 12 | load_from = None 13 | resume_from = None 14 | workflow = [('train', 1)] 15 | -------------------------------------------------------------------------------- /configs/_base_/schedules/schedule_1x.py: -------------------------------------------------------------------------------- 1 | # optimizer 2 | optimizer = dict(type='SGD', lr=0.02, momentum=0.9, weight_decay=0.0001) 3 | optimizer_config = dict(grad_clip=None) 4 | # learning policy 5 | lr_config = dict( 6 | policy='step', 7 | warmup='linear', 8 | warmup_iters=500, 9 | warmup_ratio=0.001, 10 | step=[8, 11]) 11 | total_epochs = 12 12 | -------------------------------------------------------------------------------- /configs/_base_/schedules/schedule_20e.py: -------------------------------------------------------------------------------- 1 | # optimizer 2 | optimizer = dict(type='SGD', lr=0.02, momentum=0.9, weight_decay=0.0001) 3 | optimizer_config = dict(grad_clip=None) 4 | # learning policy 5 | lr_config = dict( 6 | policy='step', 7 | warmup='linear', 8 | warmup_iters=500, 9 | warmup_ratio=0.001, 10 | step=[16, 19]) 11 | total_epochs = 20 12 | -------------------------------------------------------------------------------- /configs/_base_/schedules/schedule_2x.py: -------------------------------------------------------------------------------- 1 | # optimizer 2 | optimizer = dict(type='SGD', lr=0.02, momentum=0.9, weight_decay=0.0001) 3 | optimizer_config = dict(grad_clip=None) 4 | # learning policy 5 | lr_config = dict( 6 | policy='step', 7 | warmup='linear', 8 | warmup_iters=500, 9 | warmup_ratio=0.001, 10 | step=[16, 22]) 11 | total_epochs = 24 12 | -------------------------------------------------------------------------------- /configs/albu_example/README.md: -------------------------------------------------------------------------------- 1 | ## Results and Models 2 | 3 | | Backbone | Style | Lr schd | Mem (GB) | Inf time (fps) | box AP | mask AP | Download | 4 | |:---------:|:-------:|:-------:|:--------:|:--------------:|:------:|:-------:|:--------:| 5 | | R-50 | pytorch | 1x | 4.4 | 16.6 | 38.0 | 34.5 |[model](https://open-mmlab.s3.ap-northeast-2.amazonaws.com/mmdetection/v2.0/albu_example/mask_rcnn_r50_fpn_albu_1x_coco/mask_rcnn_r50_fpn_albu_1x_coco_20200208-ab203bcd.pth) | [log](https://open-mmlab.s3.ap-northeast-2.amazonaws.com/mmdetection/v2.0/albu_example/mask_rcnn_r50_fpn_albu_1x_coco/mask_rcnn_r50_fpn_albu_1x_coco_20200208_225520.log.json) | 6 | -------------------------------------------------------------------------------- /configs/atss/README.md: -------------------------------------------------------------------------------- 1 | # Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selection 2 | 3 | 4 | ## Introduction 5 | 6 | ``` 7 | @article{zhang2019bridging, 8 | title = {Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selection}, 9 | author = {Zhang, Shifeng and Chi, Cheng and Yao, Yongqiang and Lei, Zhen and Li, Stan Z.}, 10 | journal = {arXiv preprint arXiv:1912.02424}, 11 | year = {2019} 12 | } 13 | ``` 14 | 15 | 16 | ## Results and Models 17 | 18 | | Backbone | Style | Lr schd | Mem (GB) | Inf time (fps) | box AP | Download | 19 | |:---------:|:-------:|:-------:|:--------:|:--------------:|:------:|:--------:| 20 | | R-50 | pytorch | 1x | 3.7 | 19.7 | 39.4 | [model](https://open-mmlab.s3.ap-northeast-2.amazonaws.com/mmdetection/v2.0/atss/atss_r50_fpn_1x_coco/atss_r50_fpn_1x_coco_20200209-985f7bd0.pth) | [log](https://open-mmlab.s3.ap-northeast-2.amazonaws.com/mmdetection/v2.0/atss/atss_r50_fpn_1x_coco/atss_r50_fpn_1x_coco_20200209_102539.log.json) | 21 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_mask_rcnn_r101_caffe_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_mask_rcnn_r50_caffe_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://detectron2/resnet101_caffe', 4 | backbone=dict(depth=101)) 5 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_mask_rcnn_r101_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_mask_rcnn_r50_fpn_1x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_mask_rcnn_r101_fpn_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_mask_rcnn_r50_fpn_20e_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_mask_rcnn_r50_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/cascade_mask_rcnn_r50_fpn.py', 3 | '../_base_/datasets/coco_instance.py', 4 | '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py' 5 | ] 6 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_mask_rcnn_r50_fpn_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/cascade_mask_rcnn_r50_fpn.py', 3 | '../_base_/datasets/coco_instance.py', 4 | '../_base_/schedules/schedule_20e.py', '../_base_/default_runtime.py' 5 | ] 6 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_mask_rcnn_x101_32x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_mask_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_mask_rcnn_x101_32x4d_fpn_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_mask_rcnn_r50_fpn_20e_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_mask_rcnn_x101_64x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_mask_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_mask_rcnn_x101_64x4d_fpn_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_mask_rcnn_r50_fpn_20e_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_rcnn_r101_caffe_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_rcnn_r50_caffe_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://detectron2/resnet101_caffe', 4 | backbone=dict(depth=101)) 5 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_rcnn_r101_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_rcnn_r50_fpn_1x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_rcnn_r101_fpn_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_rcnn_r50_fpn_20e_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_rcnn_r50_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/cascade_rcnn_r50_fpn.py', 3 | '../_base_/datasets/coco_detection.py', 4 | '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py' 5 | ] 6 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_rcnn_r50_fpn_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_rcnn_r50_fpn_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 19]) 4 | total_epochs = 20 5 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_rcnn_x101_32x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_rcnn_x101_32x4d_fpn_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_rcnn_r50_fpn_20e_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_rcnn_x101_64x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | type='CascadeRCNN', 4 | pretrained='open-mmlab://resnext101_64x4d', 5 | backbone=dict( 6 | type='ResNeXt', 7 | depth=101, 8 | groups=64, 9 | base_width=4, 10 | num_stages=4, 11 | out_indices=(0, 1, 2, 3), 12 | frozen_stages=1, 13 | norm_cfg=dict(type='BN', requires_grad=True), 14 | style='pytorch')) 15 | -------------------------------------------------------------------------------- /configs/cascade_rcnn/cascade_rcnn_x101_64x4d_fpn_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_rcnn_r50_fpn_20e_coco.py' 2 | model = dict( 3 | type='CascadeRCNN', 4 | pretrained='open-mmlab://resnext101_64x4d', 5 | backbone=dict( 6 | type='ResNeXt', 7 | depth=101, 8 | groups=64, 9 | base_width=4, 10 | num_stages=4, 11 | out_indices=(0, 1, 2, 3), 12 | frozen_stages=1, 13 | norm_cfg=dict(type='BN', requires_grad=True), 14 | style='pytorch')) 15 | -------------------------------------------------------------------------------- /configs/dcn/cascade_mask_rcnn_r101_fpn_dconv_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../cascade_rcnn/cascade_mask_rcnn_r101_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | dcn=dict(type='DCN', deformable_groups=1, fallback_on_stride=False), 5 | stage_with_dcn=(False, True, True, True))) 6 | -------------------------------------------------------------------------------- /configs/dcn/cascade_mask_rcnn_r50_fpn_dconv_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../cascade_rcnn/cascade_mask_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | dcn=dict(type='DCN', deformable_groups=1, fallback_on_stride=False), 5 | stage_with_dcn=(False, True, True, True))) 6 | -------------------------------------------------------------------------------- /configs/dcn/cascade_mask_rcnn_x101_32x4d_fpn_dconv_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../cascade_rcnn/cascade_mask_rcnn_x101_32x4d_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | dcn=dict(type='DCN', deformable_groups=1, fallback_on_stride=False), 5 | stage_with_dcn=(False, True, True, True))) 6 | -------------------------------------------------------------------------------- /configs/dcn/cascade_rcnn_r101_fpn_dconv_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../cascade_rcnn/cascade_rcnn_r101_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | dcn=dict(type='DCN', deformable_groups=1, fallback_on_stride=False), 5 | stage_with_dcn=(False, True, True, True))) 6 | -------------------------------------------------------------------------------- /configs/dcn/cascade_rcnn_r50_fpn_dconv_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../cascade_rcnn/cascade_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | dcn=dict(type='DCN', deformable_groups=1, fallback_on_stride=False), 5 | stage_with_dcn=(False, True, True, True))) 6 | -------------------------------------------------------------------------------- /configs/dcn/faster_rcnn_r101_fpn_dconv_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r101_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | dcn=dict(type='DCN', deformable_groups=1, fallback_on_stride=False), 5 | stage_with_dcn=(False, True, True, True))) 6 | -------------------------------------------------------------------------------- /configs/dcn/faster_rcnn_r50_fpn_dconv_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | dcn=dict(type='DCN', deformable_groups=1, fallback_on_stride=False), 5 | stage_with_dcn=(False, True, True, True))) 6 | -------------------------------------------------------------------------------- /configs/dcn/faster_rcnn_r50_fpn_dpool_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | roi_head=dict( 4 | bbox_roi_extractor=dict( 5 | type='SingleRoIExtractor', 6 | roi_layer=dict( 7 | _delete_=True, 8 | type='DeformRoIPoolingPack', 9 | out_size=7, 10 | out_channels=256, 11 | no_trans=False, 12 | group_size=1, 13 | trans_std=0.1), 14 | out_channels=256, 15 | featmap_strides=[4, 8, 16, 32]))) 16 | -------------------------------------------------------------------------------- /configs/dcn/faster_rcnn_r50_fpn_mdconv_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | dcn=dict(type='DCNv2', deformable_groups=1, fallback_on_stride=False), 5 | stage_with_dcn=(False, True, True, True))) 6 | -------------------------------------------------------------------------------- /configs/dcn/faster_rcnn_r50_fpn_mdconv_c3-c5_group4_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | dcn=dict(type='DCNv2', deformable_groups=4, fallback_on_stride=False), 5 | stage_with_dcn=(False, True, True, True))) 6 | -------------------------------------------------------------------------------- /configs/dcn/faster_rcnn_r50_fpn_mdpool_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | roi_head=dict( 4 | bbox_roi_extractor=dict( 5 | type='SingleRoIExtractor', 6 | roi_layer=dict( 7 | _delete_=True, 8 | type='ModulatedDeformRoIPoolingPack', 9 | out_size=7, 10 | out_channels=256, 11 | no_trans=False, 12 | group_size=1, 13 | trans_std=0.1), 14 | out_channels=256, 15 | featmap_strides=[4, 8, 16, 32]))) 16 | -------------------------------------------------------------------------------- /configs/dcn/faster_rcnn_x101_32x4d_fpn_dconv_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch', 14 | dcn=dict(type='DCN', deformable_groups=1, fallback_on_stride=False), 15 | stage_with_dcn=(False, True, True, True))) 16 | -------------------------------------------------------------------------------- /configs/dcn/mask_rcnn_r101_fpn_dconv_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r101_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | dcn=dict(type='DCN', deformable_groups=1, fallback_on_stride=False), 5 | stage_with_dcn=(False, True, True, True))) 6 | -------------------------------------------------------------------------------- /configs/dcn/mask_rcnn_r50_fpn_dconv_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | dcn=dict(type='DCN', deformable_groups=1, fallback_on_stride=False), 5 | stage_with_dcn=(False, True, True, True))) 6 | -------------------------------------------------------------------------------- /configs/dcn/mask_rcnn_r50_fpn_mdconv_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | dcn=dict(type='DCNv2', deformable_groups=1, fallback_on_stride=False), 5 | stage_with_dcn=(False, True, True, True))) 6 | -------------------------------------------------------------------------------- /configs/deepfashion/mask_rcnn_r50_fpn_15e_deepfashion.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/mask_rcnn_r50_fpn.py', 3 | '../_base_/datasets/deepfashion.py', '../_base_/schedules/schedule_1x.py', 4 | '../_base_/default_runtime.py' 5 | ] 6 | model = dict( 7 | roi_head=dict( 8 | bbox_head=dict(num_classes=15), mask_head=dict(num_classes=15))) 9 | # runtime settings 10 | total_epochs = 15 11 | -------------------------------------------------------------------------------- /configs/detectors/cascade_rcnn_r50_rfp_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/cascade_rcnn_r50_fpn.py', 3 | '../_base_/datasets/coco_detection.py', 4 | '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py' 5 | ] 6 | 7 | model = dict( 8 | backbone=dict( 9 | type='DetectoRS_ResNet', 10 | conv_cfg=dict(type='ConvAWS'), 11 | output_img=True), 12 | neck=dict( 13 | type='RFP', 14 | rfp_steps=2, 15 | aspp_out_channels=64, 16 | aspp_dilations=(1, 3, 6, 1), 17 | rfp_backbone=dict( 18 | rfp_inplanes=256, 19 | type='DetectoRS_ResNet', 20 | depth=50, 21 | num_stages=4, 22 | out_indices=(0, 1, 2, 3), 23 | frozen_stages=1, 24 | norm_cfg=dict(type='BN', requires_grad=True), 25 | norm_eval=True, 26 | conv_cfg=dict(type='ConvAWS'), 27 | pretrained='torchvision://resnet50', 28 | style='pytorch'))) 29 | -------------------------------------------------------------------------------- /configs/detectors/cascade_rcnn_r50_sac_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/cascade_rcnn_r50_fpn.py', 3 | '../_base_/datasets/coco_detection.py', 4 | '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py' 5 | ] 6 | 7 | model = dict( 8 | backbone=dict( 9 | type='DetectoRS_ResNet', 10 | conv_cfg=dict(type='ConvAWS'), 11 | sac=dict(type='SAC', use_deform=True), 12 | stage_with_sac=(False, True, True, True))) 13 | -------------------------------------------------------------------------------- /configs/detectors/detectors_cascade_rcnn_r50_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/cascade_rcnn_r50_fpn.py', 3 | '../_base_/datasets/coco_detection.py', 4 | '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py' 5 | ] 6 | 7 | model = dict( 8 | backbone=dict( 9 | type='DetectoRS_ResNet', 10 | conv_cfg=dict(type='ConvAWS'), 11 | sac=dict(type='SAC', use_deform=True), 12 | stage_with_sac=(False, True, True, True), 13 | output_img=True), 14 | neck=dict( 15 | type='RFP', 16 | rfp_steps=2, 17 | aspp_out_channels=64, 18 | aspp_dilations=(1, 3, 6, 1), 19 | rfp_backbone=dict( 20 | rfp_inplanes=256, 21 | type='DetectoRS_ResNet', 22 | depth=50, 23 | num_stages=4, 24 | out_indices=(0, 1, 2, 3), 25 | frozen_stages=1, 26 | norm_cfg=dict(type='BN', requires_grad=True), 27 | norm_eval=True, 28 | conv_cfg=dict(type='ConvAWS'), 29 | sac=dict(type='SAC', use_deform=True), 30 | stage_with_sac=(False, True, True, True), 31 | pretrained='torchvision://resnet50', 32 | style='pytorch'))) 33 | -------------------------------------------------------------------------------- /configs/detectors/detectors_htc_r50_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../htc/htc_r50_fpn_1x_coco.py' 2 | 3 | model = dict( 4 | backbone=dict( 5 | type='DetectoRS_ResNet', 6 | conv_cfg=dict(type='ConvAWS'), 7 | sac=dict(type='SAC', use_deform=True), 8 | stage_with_sac=(False, True, True, True), 9 | output_img=True), 10 | neck=dict( 11 | type='RFP', 12 | rfp_steps=2, 13 | aspp_out_channels=64, 14 | aspp_dilations=(1, 3, 6, 1), 15 | rfp_backbone=dict( 16 | rfp_inplanes=256, 17 | type='DetectoRS_ResNet', 18 | depth=50, 19 | num_stages=4, 20 | out_indices=(0, 1, 2, 3), 21 | frozen_stages=1, 22 | norm_cfg=dict(type='BN', requires_grad=True), 23 | norm_eval=True, 24 | conv_cfg=dict(type='ConvAWS'), 25 | sac=dict(type='SAC', use_deform=True), 26 | stage_with_sac=(False, True, True, True), 27 | pretrained='torchvision://resnet50', 28 | style='pytorch'))) 29 | -------------------------------------------------------------------------------- /configs/detectors/htc_r50_rfp_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../htc/htc_r50_fpn_1x_coco.py' 2 | 3 | model = dict( 4 | backbone=dict( 5 | type='DetectoRS_ResNet', 6 | conv_cfg=dict(type='ConvAWS'), 7 | output_img=True), 8 | neck=dict( 9 | type='RFP', 10 | rfp_steps=2, 11 | aspp_out_channels=64, 12 | aspp_dilations=(1, 3, 6, 1), 13 | rfp_backbone=dict( 14 | rfp_inplanes=256, 15 | type='DetectoRS_ResNet', 16 | depth=50, 17 | num_stages=4, 18 | out_indices=(0, 1, 2, 3), 19 | frozen_stages=1, 20 | norm_cfg=dict(type='BN', requires_grad=True), 21 | norm_eval=True, 22 | conv_cfg=dict(type='ConvAWS'), 23 | pretrained='torchvision://resnet50', 24 | style='pytorch'))) 25 | -------------------------------------------------------------------------------- /configs/detectors/htc_r50_sac_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../htc/htc_r50_fpn_1x_coco.py' 2 | 3 | model = dict( 4 | backbone=dict( 5 | type='DetectoRS_ResNet', 6 | conv_cfg=dict(type='ConvAWS'), 7 | sac=dict(type='SAC', use_deform=True), 8 | stage_with_sac=(False, True, True, True))) 9 | -------------------------------------------------------------------------------- /configs/double_heads/README.md: -------------------------------------------------------------------------------- 1 | # Rethinking Classification and Localization for Object Detection 2 | 3 | ## Introduction 4 | ``` 5 | @article{wu2019rethinking, 6 | title={Rethinking Classification and Localization for Object Detection}, 7 | author={Yue Wu and Yinpeng Chen and Lu Yuan and Zicheng Liu and Lijuan Wang and Hongzhi Li and Yun Fu}, 8 | year={2019}, 9 | eprint={1904.06493}, 10 | archivePrefix={arXiv}, 11 | primaryClass={cs.CV} 12 | } 13 | ``` 14 | 15 | ## Results and models 16 | 17 | | Backbone | Style | Lr schd | Mem (GB) | Inf time (fps) | box AP | Download | 18 | | :-------------: | :-----: | :-----: | :------: | :------------: | :----: | :----------------: | 19 | | R-50-FPN | pytorch | 1x | 6.8 | 9.5 | 40.0 | [model](https://open-mmlab.s3.ap-northeast-2.amazonaws.com/mmdetection/v2.0/double_heads/dh_faster_rcnn_r50_fpn_1x_coco/dh_faster_rcnn_r50_fpn_1x_coco_20200130-586b67df.pth) | [log](https://open-mmlab.s3.ap-northeast-2.amazonaws.com/mmdetection/v2.0/double_heads/dh_faster_rcnn_r50_fpn_1x_coco/dh_faster_rcnn_r50_fpn_1x_coco_20200130_220238.log.json) | 20 | -------------------------------------------------------------------------------- /configs/double_heads/dh_faster_rcnn_r50_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | roi_head=dict( 4 | type='DoubleHeadRoIHead', 5 | reg_roi_scale_factor=1.3, 6 | bbox_head=dict( 7 | _delete_=True, 8 | type='DoubleConvFCBBoxHead', 9 | num_convs=4, 10 | num_fcs=2, 11 | in_channels=256, 12 | conv_out_channels=1024, 13 | fc_out_channels=1024, 14 | roi_feat_size=7, 15 | num_classes=80, 16 | bbox_coder=dict( 17 | type='DeltaXYWHBBoxCoder', 18 | target_means=[0., 0., 0., 0.], 19 | target_stds=[0.1, 0.1, 0.2, 0.2]), 20 | reg_class_agnostic=False, 21 | loss_cls=dict( 22 | type='CrossEntropyLoss', use_sigmoid=False, loss_weight=2.0), 23 | loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=2.0)))) 24 | -------------------------------------------------------------------------------- /configs/dynamic_rcnn/README.md: -------------------------------------------------------------------------------- 1 | # Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Training 2 | 3 | ## Introduction 4 | 5 | ``` 6 | @article{DynamicRCNN, 7 | author = {Hongkai Zhang and Hong Chang and Bingpeng Ma and Naiyan Wang and Xilin Chen}, 8 | title = {Dynamic {R-CNN}: Towards High Quality Object Detection via Dynamic Training}, 9 | journal = {arXiv preprint arXiv:2004.06002}, 10 | year = {2020} 11 | } 12 | ``` 13 | 14 | ## Results and Models 15 | 16 | | Backbone | Style | Lr schd | Mem (GB) | Inf time (fps) | box AP | Download | 17 | |:---------:|:-------:|:-------:|:--------:|:--------------:|:------:|:--------:| 18 | | R-50 | pytorch | 1x | 3.8 | | 38.9 | [model](https://open-mmlab.s3.ap-northeast-2.amazonaws.com/mmdetection/v2.0/dynamic_rcnn/dynamic_rcnn_r50_fpn_1x/dynamic_rcnn_r50_fpn_1x-62a3f276.pth) | [log](https://open-mmlab.s3.ap-northeast-2.amazonaws.com/mmdetection/v2.0/dynamic_rcnn/dynamic_rcnn_r50_fpn_1x/dynamic_rcnn_r50_fpn_1x_20200618_095048.log.json) | 19 | -------------------------------------------------------------------------------- /configs/dynamic_rcnn/dynamic_rcnn_r50_fpn_1x.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | roi_head=dict( 4 | type='DynamicRoIHead', 5 | bbox_head=dict( 6 | type='Shared2FCBBoxHead', 7 | in_channels=256, 8 | fc_out_channels=1024, 9 | roi_feat_size=7, 10 | num_classes=80, 11 | bbox_coder=dict( 12 | type='DeltaXYWHBBoxCoder', 13 | target_means=[0., 0., 0., 0.], 14 | target_stds=[0.1, 0.1, 0.2, 0.2]), 15 | reg_class_agnostic=False, 16 | loss_cls=dict( 17 | type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0), 18 | loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)))) 19 | train_cfg = dict( 20 | rpn_proposal=dict(nms_thr=0.85), 21 | rcnn=dict( 22 | dynamic_rcnn=dict( 23 | iou_topk=75, 24 | beta_topk=10, 25 | update_iter_interval=100, 26 | initial_iou=0.4, 27 | initial_beta=1.0))) 28 | test_cfg = dict(rpn=dict(nms_thr=0.85)) 29 | -------------------------------------------------------------------------------- /configs/empirical_attention/faster_rcnn_r50_fpn_attention_0010_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict(plugins=[ 4 | dict( 5 | cfg=dict( 6 | type='GeneralizedAttention', 7 | spatial_range=-1, 8 | num_heads=8, 9 | attention_type='0010', 10 | kv_stride=2), 11 | stages=(False, False, True, True), 12 | position='after_conv2') 13 | ])) 14 | -------------------------------------------------------------------------------- /configs/empirical_attention/faster_rcnn_r50_fpn_attention_0010_dcn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | plugins=[ 5 | dict( 6 | cfg=dict( 7 | type='GeneralizedAttention', 8 | spatial_range=-1, 9 | num_heads=8, 10 | attention_type='0010', 11 | kv_stride=2), 12 | stages=(False, False, True, True), 13 | position='after_conv2') 14 | ], 15 | dcn=dict(type='DCN', deformable_groups=1, fallback_on_stride=False), 16 | stage_with_dcn=(False, True, True, True))) 17 | -------------------------------------------------------------------------------- /configs/empirical_attention/faster_rcnn_r50_fpn_attention_1111_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict(plugins=[ 4 | dict( 5 | cfg=dict( 6 | type='GeneralizedAttention', 7 | spatial_range=-1, 8 | num_heads=8, 9 | attention_type='1111', 10 | kv_stride=2), 11 | stages=(False, False, True, True), 12 | position='after_conv2') 13 | ])) 14 | -------------------------------------------------------------------------------- /configs/empirical_attention/faster_rcnn_r50_fpn_attention_1111_dcn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | plugins=[ 5 | dict( 6 | cfg=dict( 7 | type='GeneralizedAttention', 8 | spatial_range=-1, 9 | num_heads=8, 10 | attention_type='1111', 11 | kv_stride=2), 12 | stages=(False, False, True, True), 13 | position='after_conv2') 14 | ], 15 | dcn=dict(type='DCN', deformable_groups=1, fallback_on_stride=False), 16 | stage_with_dcn=(False, True, True, True))) 17 | -------------------------------------------------------------------------------- /configs/fast_rcnn/README.md: -------------------------------------------------------------------------------- 1 | # Fast R-CNN 2 | 3 | ## Introduction 4 | ``` 5 | @inproceedings{girshick2015fast, 6 | title={Fast r-cnn}, 7 | author={Girshick, Ross}, 8 | booktitle={Proceedings of the IEEE international conference on computer vision}, 9 | year={2015} 10 | } 11 | ``` 12 | 13 | ## Results and models 14 | -------------------------------------------------------------------------------- /configs/fast_rcnn/fast_rcnn_r101_caffe_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fast_rcnn_r50_caffe_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://detectron2/resnet101_caffe', 4 | backbone=dict(depth=101)) 5 | -------------------------------------------------------------------------------- /configs/fast_rcnn/fast_rcnn_r101_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fast_rcnn_r50_fpn_1x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/fast_rcnn/fast_rcnn_r101_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fast_rcnn_r50_fpn_2x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/fast_rcnn/fast_rcnn_r50_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fast_rcnn_r50_fpn_1x_coco.py' 2 | 3 | # learning policy 4 | lr_config = dict(step=[16, 22]) 5 | total_epochs = 24 6 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_r101_caffe_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_caffe_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://detectron2/resnet101_caffe', 4 | backbone=dict(depth=101)) 5 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_r101_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_r101_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_fpn_2x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_r50_caffe_fpn_mstrain_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_caffe_fpn_mstrain_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 23]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_r50_caffe_fpn_mstrain_3x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_caffe_fpn_mstrain_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[28, 34]) 4 | total_epochs = 36 5 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_r50_fpn_1x_coco-person-bicycle-car.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_fpn_1x_coco.py' 2 | classes = ('person', 'bicycle', 'car') 3 | data = dict( 4 | train=dict(classes=classes), 5 | val=dict(classes=classes), 6 | test=dict(classes=classes)) 7 | # TODO: Update model url after bumping to V2.0 8 | load_from = 'https://s3.ap-northeast-2.amazonaws.com/open-mmlab/mmdetection/models/faster_rcnn_r50_fpn_1x_20181010-3d1b3351.pth' # noqa 9 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_r50_fpn_1x_coco-person.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_fpn_1x_coco.py' 2 | classes = ('person', ) 3 | data = dict( 4 | train=dict(classes=classes), 5 | val=dict(classes=classes), 6 | test=dict(classes=classes)) 7 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/faster_rcnn_r50_fpn.py', 3 | '../_base_/datasets/coco_detection.py', 4 | '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py' 5 | ] 6 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_r50_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/faster_rcnn_r50_fpn.py', 3 | '../_base_/datasets/coco_detection.py', 4 | '../_base_/schedules/schedule_2x.py', '../_base_/default_runtime.py' 5 | ] 6 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_r50_fpn_bounded_iou_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | roi_head=dict( 4 | bbox_head=dict( 5 | reg_decoded_bbox=True, 6 | loss_bbox=dict(type='BoundedIoULoss', loss_weight=10.0)))) 7 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_r50_fpn_giou_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | roi_head=dict( 4 | bbox_head=dict( 5 | reg_decoded_bbox=True, 6 | loss_bbox=dict(type='GIoULoss', loss_weight=10.0)))) 7 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_r50_fpn_iou_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | roi_head=dict( 4 | bbox_head=dict( 5 | reg_decoded_bbox=True, 6 | loss_bbox=dict(type='IoULoss', loss_weight=10.0)))) 7 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_r50_fpn_ohem_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_fpn_1x_coco.py' 2 | train_cfg = dict(rcnn=dict(sampler=dict(type='OHEMSampler'))) 3 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_r50_fpn_soft_nms_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/faster_rcnn_r50_fpn.py', 3 | '../_base_/datasets/coco_detection.py', 4 | '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py' 5 | ] 6 | 7 | test_cfg = dict( 8 | rcnn=dict( 9 | score_thr=0.05, 10 | nms=dict(type='soft_nms', iou_thr=0.5), 11 | max_per_img=100)) 12 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_x101_32x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_x101_32x4d_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_fpn_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_x101_64x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/faster_rcnn/faster_rcnn_x101_64x4d_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_fpn_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/fcos/fcos_center_r50_caffe_fpn_gn-head_4x4_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fcos_r50_caffe_fpn_gn-head_4x4_1x_coco.py' 2 | model = dict(bbox_head=dict(center_sampling=True, center_sample_radius=1.5)) 3 | -------------------------------------------------------------------------------- /configs/fcos/fcos_r101_caffe_fpn_gn-head_4x4_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fcos_r50_caffe_fpn_gn-head_4x4_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://detectron/resnet101_caffe', 4 | backbone=dict(depth=101)) 5 | -------------------------------------------------------------------------------- /configs/fcos/fcos_r101_caffe_fpn_gn-head_4x4_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = ['./fcos_r50_caffe_fpn_gn-head_4x4_2x_coco.py'] 2 | model = dict( 3 | pretrained='open-mmlab://detectron/resnet101_caffe', 4 | backbone=dict(depth=101)) 5 | -------------------------------------------------------------------------------- /configs/fcos/fcos_r50_caffe_fpn_gn-head_4x4_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fcos_r50_caffe_fpn_gn-head_4x4_1x_coco.py' 2 | 3 | # learning policy 4 | lr_config = dict(step=[16, 22]) 5 | total_epochs = 24 6 | -------------------------------------------------------------------------------- /configs/foveabox/fovea_align_r101_fpn_gn-head_4x4_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fovea_r50_fpn_4x4_1x_coco.py' 2 | model = dict( 3 | pretrained='torchvision://resnet101', 4 | backbone=dict(depth=101), 5 | bbox_head=dict( 6 | with_deform=True, 7 | norm_cfg=dict(type='GN', num_groups=32, requires_grad=True))) 8 | # learning policy 9 | lr_config = dict(step=[16, 22]) 10 | total_epochs = 24 11 | -------------------------------------------------------------------------------- /configs/foveabox/fovea_align_r101_fpn_gn-head_mstrain_640-800_4x4_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fovea_r50_fpn_4x4_1x_coco.py' 2 | model = dict( 3 | pretrained='torchvision://resnet101', 4 | backbone=dict(depth=101), 5 | bbox_head=dict( 6 | with_deform=True, 7 | norm_cfg=dict(type='GN', num_groups=32, requires_grad=True))) 8 | img_norm_cfg = dict( 9 | mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True) 10 | train_pipeline = [ 11 | dict(type='LoadImageFromFile'), 12 | dict(type='LoadAnnotations', with_bbox=True), 13 | dict( 14 | type='Resize', 15 | img_scale=[(1333, 640), (1333, 800)], 16 | multiscale_mode='value', 17 | keep_ratio=True), 18 | dict(type='RandomFlip', flip_ratio=0.5), 19 | dict(type='Normalize', **img_norm_cfg), 20 | dict(type='Pad', size_divisor=32), 21 | dict(type='DefaultFormatBundle'), 22 | dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels']), 23 | ] 24 | data = dict(train=dict(pipeline=train_pipeline)) 25 | # learning policy 26 | lr_config = dict(step=[16, 22]) 27 | total_epochs = 24 28 | -------------------------------------------------------------------------------- /configs/foveabox/fovea_align_r50_fpn_gn-head_4x4_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fovea_r50_fpn_4x4_1x_coco.py' 2 | model = dict( 3 | bbox_head=dict( 4 | with_deform=True, 5 | norm_cfg=dict(type='GN', num_groups=32, requires_grad=True))) 6 | # learning policy 7 | lr_config = dict(step=[16, 22]) 8 | total_epochs = 24 9 | optimizer_config = dict( 10 | _delete_=True, grad_clip=dict(max_norm=35, norm_type=2)) 11 | -------------------------------------------------------------------------------- /configs/foveabox/fovea_align_r50_fpn_gn-head_mstrain_640-800_4x4_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fovea_r50_fpn_4x4_1x_coco.py' 2 | model = dict( 3 | bbox_head=dict( 4 | with_deform=True, 5 | norm_cfg=dict(type='GN', num_groups=32, requires_grad=True))) 6 | img_norm_cfg = dict( 7 | mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True) 8 | train_pipeline = [ 9 | dict(type='LoadImageFromFile'), 10 | dict(type='LoadAnnotations', with_bbox=True), 11 | dict( 12 | type='Resize', 13 | img_scale=[(1333, 640), (1333, 800)], 14 | multiscale_mode='value', 15 | keep_ratio=True), 16 | dict(type='RandomFlip', flip_ratio=0.5), 17 | dict(type='Normalize', **img_norm_cfg), 18 | dict(type='Pad', size_divisor=32), 19 | dict(type='DefaultFormatBundle'), 20 | dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels']), 21 | ] 22 | data = dict(train=dict(pipeline=train_pipeline)) 23 | # learning policy 24 | lr_config = dict(step=[16, 22]) 25 | total_epochs = 24 26 | -------------------------------------------------------------------------------- /configs/foveabox/fovea_r101_fpn_4x4_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fovea_r50_fpn_4x4_1x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/foveabox/fovea_r101_fpn_4x4_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fovea_r50_fpn_4x4_2x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/foveabox/fovea_r50_fpn_4x4_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fovea_r50_fpn_4x4_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 22]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/fp16/faster_rcnn_r50_fpn_fp16_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | # fp16 settings 3 | fp16 = dict(loss_scale=512.) 4 | -------------------------------------------------------------------------------- /configs/fp16/mask_rcnn_r50_fpn_fp16_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py' 2 | # fp16 settings 3 | fp16 = dict(loss_scale=512.) 4 | -------------------------------------------------------------------------------- /configs/fp16/retinanet_r50_fpn_fp16_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../retinanet/retinanet_r50_fpn_1x_coco.py' 2 | # fp16 settings 3 | fp16 = dict(loss_scale=512.) 4 | -------------------------------------------------------------------------------- /configs/free_anchor/retinanet_free_anchor_r101_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_free_anchor_r50_fpn_1x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/free_anchor/retinanet_free_anchor_r50_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../retinanet/retinanet_r50_fpn_1x_coco.py' 2 | model = dict( 3 | bbox_head=dict( 4 | _delete_=True, 5 | type='FreeAnchorRetinaHead', 6 | num_classes=80, 7 | in_channels=256, 8 | stacked_convs=4, 9 | feat_channels=256, 10 | anchor_generator=dict( 11 | type='AnchorGenerator', 12 | octave_base_scale=4, 13 | scales_per_octave=3, 14 | ratios=[0.5, 1.0, 2.0], 15 | strides=[8, 16, 32, 64, 128]), 16 | bbox_coder=dict( 17 | type='DeltaXYWHBBoxCoder', 18 | target_means=[.0, .0, .0, .0], 19 | target_stds=[0.1, 0.1, 0.2, 0.2]), 20 | loss_bbox=dict(type='SmoothL1Loss', beta=0.11, loss_weight=0.75))) 21 | optimizer_config = dict( 22 | _delete_=True, grad_clip=dict(max_norm=35, norm_type=2)) 23 | -------------------------------------------------------------------------------- /configs/free_anchor/retinanet_free_anchor_x101_32x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_free_anchor_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | style='pytorch')) 13 | -------------------------------------------------------------------------------- /configs/fsaf/fsaf_r101_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fsaf_r50_fpn_1x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/fsaf/fsaf_x101_64x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fsaf_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/gcnet/cascade_mask_rcnn_x101_32x4d_fpn_syncbn-backbone_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../cascade_rcnn/cascade_mask_rcnn_x101_32x4d_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | norm_cfg=dict(type='SyncBN', requires_grad=True), norm_eval=False)) 5 | -------------------------------------------------------------------------------- /configs/gcnet/cascade_mask_rcnn_x101_32x4d_fpn_syncbn-backbone_dconv_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../dcn/cascade_mask_rcnn_r50_fpn_dconv_c3-c5_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | norm_cfg=dict(type='SyncBN', requires_grad=True), norm_eval=False)) 5 | -------------------------------------------------------------------------------- /configs/gcnet/cascade_mask_rcnn_x101_32x4d_fpn_syncbn-backbone_dconv_c3-c5_r16_gcb_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../dcn/cascade_rcnn_r50_fpn_dconv_c3-c5_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | norm_cfg=dict(type='SyncBN', requires_grad=True), 5 | norm_eval=False, 6 | plugins=[ 7 | dict( 8 | cfg=dict(type='ContextBlock', ratio=1. / 16), 9 | stages=(False, True, True, True), 10 | position='after_conv3') 11 | ])) 12 | -------------------------------------------------------------------------------- /configs/gcnet/cascade_mask_rcnn_x101_32x4d_fpn_syncbn-backbone_dconv_c3-c5_r4_gcb_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../dcn/cascade_mask_rcnn_r50_fpn_dconv_c3-c5_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | norm_cfg=dict(type='SyncBN', requires_grad=True), 5 | norm_eval=False, 6 | plugins=[ 7 | dict( 8 | cfg=dict(type='ContextBlock', ratio=1. / 4), 9 | stages=(False, True, True, True), 10 | position='after_conv3') 11 | ])) 12 | -------------------------------------------------------------------------------- /configs/gcnet/cascade_mask_rcnn_x101_32x4d_fpn_syncbn-backbone_r16_gcb_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../cascade_rcnn/cascade_mask_rcnn_x101_32x4d_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | norm_cfg=dict(type='SyncBN', requires_grad=True), 5 | norm_eval=False, 6 | plugins=[ 7 | dict( 8 | cfg=dict(type='ContextBlock', ratio=1. / 16), 9 | stages=(False, True, True, True), 10 | position='after_conv3') 11 | ])) 12 | -------------------------------------------------------------------------------- /configs/gcnet/cascade_mask_rcnn_x101_32x4d_fpn_syncbn-backbone_r4_gcb_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../cascade_rcnn/cascade_mask_rcnn_x101_32x4d_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | norm_cfg=dict(type='SyncBN', requires_grad=True), 5 | norm_eval=False, 6 | plugins=[ 7 | dict( 8 | cfg=dict(type='ContextBlock', ratio=1. / 4), 9 | stages=(False, True, True, True), 10 | position='after_conv3') 11 | ])) 12 | -------------------------------------------------------------------------------- /configs/gcnet/mask_rcnn_r101_fpn_r16_gcb_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r101_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict(plugins=[ 4 | dict( 5 | cfg=dict(type='ContextBlock', ratio=1. / 16), 6 | stages=(False, True, True, True), 7 | position='after_conv3') 8 | ])) 9 | -------------------------------------------------------------------------------- /configs/gcnet/mask_rcnn_r101_fpn_r4_gcb_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r101_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict(plugins=[ 4 | dict( 5 | cfg=dict(type='ContextBlock', ratio=1. / 4), 6 | stages=(False, True, True, True), 7 | position='after_conv3') 8 | ])) 9 | -------------------------------------------------------------------------------- /configs/gcnet/mask_rcnn_r101_fpn_syncbn-backbone_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r101_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | norm_cfg=dict(type='SyncBN', requires_grad=True), norm_eval=False)) 5 | -------------------------------------------------------------------------------- /configs/gcnet/mask_rcnn_r101_fpn_syncbn-backbone_r16_gcb_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r101_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | norm_cfg=dict(type='SyncBN', requires_grad=True), 5 | norm_eval=False, 6 | plugins=[ 7 | dict( 8 | cfg=dict(type='ContextBlock', ratio=1. / 16), 9 | stages=(False, True, True, True), 10 | position='after_conv3') 11 | ])) 12 | -------------------------------------------------------------------------------- /configs/gcnet/mask_rcnn_r101_fpn_syncbn-backbone_r4_gcb_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r101_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | norm_cfg=dict(type='SyncBN', requires_grad=True), 5 | norm_eval=False, 6 | plugins=[ 7 | dict( 8 | cfg=dict(type='ContextBlock', ratio=1. / 4), 9 | stages=(False, True, True, True), 10 | position='after_conv3') 11 | ])) 12 | -------------------------------------------------------------------------------- /configs/gcnet/mask_rcnn_r50_fpn_r16_gcb_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict(plugins=[ 4 | dict( 5 | cfg=dict(type='ContextBlock', ratio=1. / 16), 6 | stages=(False, True, True, True), 7 | position='after_conv3') 8 | ])) 9 | -------------------------------------------------------------------------------- /configs/gcnet/mask_rcnn_r50_fpn_r4_gcb_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict(plugins=[ 4 | dict( 5 | cfg=dict(type='ContextBlock', ratio=1. / 4), 6 | stages=(False, True, True, True), 7 | position='after_conv3') 8 | ])) 9 | -------------------------------------------------------------------------------- /configs/gcnet/mask_rcnn_r50_fpn_syncbn-backbone_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | norm_cfg=dict(type='SyncBN', requires_grad=True), norm_eval=False)) 5 | -------------------------------------------------------------------------------- /configs/gcnet/mask_rcnn_r50_fpn_syncbn-backbone_r16_gcb_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | norm_cfg=dict(type='SyncBN', requires_grad=True), 5 | norm_eval=False, 6 | plugins=[ 7 | dict( 8 | cfg=dict(type='ContextBlock', ratio=1. / 16), 9 | stages=(False, True, True, True), 10 | position='after_conv3') 11 | ])) 12 | -------------------------------------------------------------------------------- /configs/gcnet/mask_rcnn_r50_fpn_syncbn-backbone_r4_gcb_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | norm_cfg=dict(type='SyncBN', requires_grad=True), 5 | norm_eval=False, 6 | plugins=[ 7 | dict( 8 | cfg=dict(type='ContextBlock', ratio=1. / 4), 9 | stages=(False, True, True, True), 10 | position='after_conv3') 11 | ])) 12 | -------------------------------------------------------------------------------- /configs/gcnet/mask_rcnn_x101_32x4d_fpn_syncbn-backbone_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_x101_32x4d_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | norm_cfg=dict(type='SyncBN', requires_grad=True), norm_eval=False)) 5 | -------------------------------------------------------------------------------- /configs/gcnet/mask_rcnn_x101_32x4d_fpn_syncbn-backbone_r16_gcb_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_x101_32x4d_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | norm_cfg=dict(type='SyncBN', requires_grad=True), 5 | norm_eval=False, 6 | plugins=[ 7 | dict( 8 | cfg=dict(type='ContextBlock', ratio=1. / 16), 9 | stages=(False, True, True, True), 10 | position='after_conv3') 11 | ])) 12 | -------------------------------------------------------------------------------- /configs/gcnet/mask_rcnn_x101_32x4d_fpn_syncbn-backbone_r4_gcb_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_x101_32x4d_fpn_1x_coco.py' 2 | model = dict( 3 | backbone=dict( 4 | norm_cfg=dict(type='SyncBN', requires_grad=True), 5 | norm_eval=False, 6 | plugins=[ 7 | dict( 8 | cfg=dict(type='ContextBlock', ratio=1. / 4), 9 | stages=(False, True, True, True), 10 | position='after_conv3') 11 | ])) 12 | -------------------------------------------------------------------------------- /configs/gfl/gfl_r101_fpn_dconv_c3-c5_mstrain_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './gfl_r50_fpn_mstrain_2x_coco.py' 2 | model = dict( 3 | pretrained='torchvision://resnet101', 4 | backbone=dict( 5 | type='ResNet', 6 | depth=101, 7 | num_stages=4, 8 | out_indices=(0, 1, 2, 3), 9 | frozen_stages=1, 10 | norm_cfg=dict(type='BN', requires_grad=True), 11 | dcn=dict(type='DCN', deformable_groups=1, fallback_on_stride=False), 12 | stage_with_dcn=(False, True, True, True), 13 | norm_eval=True, 14 | style='pytorch')) 15 | -------------------------------------------------------------------------------- /configs/gfl/gfl_r101_fpn_mstrain_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './gfl_r50_fpn_mstrain_2x_coco.py' 2 | model = dict( 3 | pretrained='torchvision://resnet101', 4 | backbone=dict( 5 | type='ResNet', 6 | depth=101, 7 | num_stages=4, 8 | out_indices=(0, 1, 2, 3), 9 | frozen_stages=1, 10 | norm_cfg=dict(type='BN', requires_grad=True), 11 | norm_eval=True, 12 | style='pytorch')) 13 | -------------------------------------------------------------------------------- /configs/gfl/gfl_r50_fpn_mstrain_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './gfl_r50_fpn_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 22]) 4 | total_epochs = 24 5 | # multi-scale training 6 | img_norm_cfg = dict( 7 | mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True) 8 | train_pipeline = [ 9 | dict(type='LoadImageFromFile'), 10 | dict(type='LoadAnnotations', with_bbox=True), 11 | dict( 12 | type='Resize', 13 | img_scale=[(1333, 480), (1333, 800)], 14 | multiscale_mode='range', 15 | keep_ratio=True), 16 | dict(type='RandomFlip', flip_ratio=0.5), 17 | dict(type='Normalize', **img_norm_cfg), 18 | dict(type='Pad', size_divisor=32), 19 | dict(type='DefaultFormatBundle'), 20 | dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels']), 21 | ] 22 | data = dict(train=dict(pipeline=train_pipeline)) 23 | -------------------------------------------------------------------------------- /configs/gfl/gfl_x101_32x4d_fpn_dconv_c4-c5_mstrain_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './gfl_r50_fpn_mstrain_2x_coco.py' 2 | model = dict( 3 | type='GFL', 4 | pretrained='open-mmlab://resnext101_32x4d', 5 | backbone=dict( 6 | type='ResNeXt', 7 | depth=101, 8 | groups=32, 9 | base_width=4, 10 | num_stages=4, 11 | out_indices=(0, 1, 2, 3), 12 | frozen_stages=1, 13 | norm_cfg=dict(type='BN', requires_grad=True), 14 | dcn=dict(type='DCN', deformable_groups=1, fallback_on_stride=False), 15 | stage_with_dcn=(False, False, True, True), 16 | norm_eval=True, 17 | style='pytorch')) 18 | -------------------------------------------------------------------------------- /configs/gfl/gfl_x101_32x4d_fpn_mstrain_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './gfl_r50_fpn_mstrain_2x_coco.py' 2 | model = dict( 3 | type='GFL', 4 | pretrained='open-mmlab://resnext101_32x4d', 5 | backbone=dict( 6 | type='ResNeXt', 7 | depth=101, 8 | groups=32, 9 | base_width=4, 10 | num_stages=4, 11 | out_indices=(0, 1, 2, 3), 12 | frozen_stages=1, 13 | norm_cfg=dict(type='BN', requires_grad=True), 14 | norm_eval=True, 15 | style='pytorch')) 16 | -------------------------------------------------------------------------------- /configs/ghm/retinanet_ghm_r101_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_ghm_r50_fpn_1x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/ghm/retinanet_ghm_r50_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../retinanet/retinanet_r50_fpn_1x_coco.py' 2 | model = dict( 3 | bbox_head=dict( 4 | loss_cls=dict( 5 | _delete_=True, 6 | type='GHMC', 7 | bins=30, 8 | momentum=0.75, 9 | use_sigmoid=True, 10 | loss_weight=1.0), 11 | loss_bbox=dict( 12 | _delete_=True, 13 | type='GHMR', 14 | mu=0.02, 15 | bins=10, 16 | momentum=0.7, 17 | loss_weight=10.0))) 18 | optimizer_config = dict( 19 | _delete_=True, grad_clip=dict(max_norm=35, norm_type=2)) 20 | -------------------------------------------------------------------------------- /configs/ghm/retinanet_ghm_x101_32x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_ghm_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/ghm/retinanet_ghm_x101_64x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_ghm_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/gn+ws/faster_rcnn_r101_fpn_gn_ws-all_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_fpn_gn_ws-all_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://jhu/resnet101_gn_ws', backbone=dict(depth=101)) 4 | -------------------------------------------------------------------------------- /configs/gn+ws/faster_rcnn_r50_fpn_gn_ws-all_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | conv_cfg = dict(type='ConvWS') 3 | norm_cfg = dict(type='GN', num_groups=32, requires_grad=True) 4 | model = dict( 5 | pretrained='open-mmlab://jhu/resnet50_gn_ws', 6 | backbone=dict(conv_cfg=conv_cfg, norm_cfg=norm_cfg), 7 | neck=dict(conv_cfg=conv_cfg, norm_cfg=norm_cfg), 8 | roi_head=dict( 9 | bbox_head=dict( 10 | type='Shared4Conv1FCBBoxHead', 11 | conv_out_channels=256, 12 | conv_cfg=conv_cfg, 13 | norm_cfg=norm_cfg))) 14 | -------------------------------------------------------------------------------- /configs/gn+ws/faster_rcnn_x101_32x4d_fpn_gn_ws-all_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_fpn_gn_ws-all_1x_coco.py' 2 | conv_cfg = dict(type='ConvWS') 3 | norm_cfg = dict(type='GN', num_groups=32, requires_grad=True) 4 | model = dict( 5 | pretrained='open-mmlab://jhu/resnext101_32x4d_gn_ws', 6 | backbone=dict( 7 | type='ResNeXt', 8 | depth=101, 9 | groups=32, 10 | base_width=4, 11 | num_stages=4, 12 | out_indices=(0, 1, 2, 3), 13 | frozen_stages=1, 14 | style='pytorch', 15 | conv_cfg=conv_cfg, 16 | norm_cfg=norm_cfg)) 17 | -------------------------------------------------------------------------------- /configs/gn+ws/faster_rcnn_x50_32x4d_fpn_gn_ws-all_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_r50_fpn_gn_ws-all_1x_coco.py' 2 | conv_cfg = dict(type='ConvWS') 3 | norm_cfg = dict(type='GN', num_groups=32, requires_grad=True) 4 | model = dict( 5 | pretrained='open-mmlab://jhu/resnext50_32x4d_gn_ws', 6 | backbone=dict( 7 | type='ResNeXt', 8 | depth=50, 9 | groups=32, 10 | base_width=4, 11 | num_stages=4, 12 | out_indices=(0, 1, 2, 3), 13 | frozen_stages=1, 14 | style='pytorch', 15 | conv_cfg=conv_cfg, 16 | norm_cfg=norm_cfg)) 17 | -------------------------------------------------------------------------------- /configs/gn+ws/mask_rcnn_r101_fpn_gn_ws-all_20_23_24e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r101_fpn_gn_ws-all_2x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[20, 23]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/gn+ws/mask_rcnn_r101_fpn_gn_ws-all_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_fpn_gn_ws-all_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://jhu/resnet101_gn_ws', backbone=dict(depth=101)) 4 | -------------------------------------------------------------------------------- /configs/gn+ws/mask_rcnn_r50_fpn_gn_ws-all_20_23_24e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_fpn_gn_ws-all_2x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[20, 23]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/gn+ws/mask_rcnn_r50_fpn_gn_ws-all_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py' 2 | conv_cfg = dict(type='ConvWS') 3 | norm_cfg = dict(type='GN', num_groups=32, requires_grad=True) 4 | model = dict( 5 | pretrained='open-mmlab://jhu/resnet50_gn_ws', 6 | backbone=dict(conv_cfg=conv_cfg, norm_cfg=norm_cfg), 7 | neck=dict(conv_cfg=conv_cfg, norm_cfg=norm_cfg), 8 | roi_head=dict( 9 | bbox_head=dict( 10 | type='Shared4Conv1FCBBoxHead', 11 | conv_out_channels=256, 12 | conv_cfg=conv_cfg, 13 | norm_cfg=norm_cfg), 14 | mask_head=dict(conv_cfg=conv_cfg, norm_cfg=norm_cfg))) 15 | # learning policy 16 | lr_config = dict(step=[16, 22]) 17 | total_epochs = 24 18 | -------------------------------------------------------------------------------- /configs/gn+ws/mask_rcnn_x101_32x4d_fpn_gn_ws-all_20_23_24e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_x101_32x4d_fpn_gn_ws-all_2x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[20, 23]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/gn+ws/mask_rcnn_x101_32x4d_fpn_gn_ws-all_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_fpn_gn_ws-all_2x_coco.py' 2 | # model settings 3 | conv_cfg = dict(type='ConvWS') 4 | norm_cfg = dict(type='GN', num_groups=32, requires_grad=True) 5 | model = dict( 6 | pretrained='open-mmlab://jhu/resnext101_32x4d_gn_ws', 7 | backbone=dict( 8 | type='ResNeXt', 9 | depth=101, 10 | groups=32, 11 | base_width=4, 12 | num_stages=4, 13 | out_indices=(0, 1, 2, 3), 14 | frozen_stages=1, 15 | style='pytorch', 16 | conv_cfg=conv_cfg, 17 | norm_cfg=norm_cfg)) 18 | -------------------------------------------------------------------------------- /configs/gn+ws/mask_rcnn_x50_32x4d_fpn_gn_ws-all_20_23_24e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_x50_32x4d_fpn_gn_ws-all_2x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[20, 23]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/gn+ws/mask_rcnn_x50_32x4d_fpn_gn_ws-all_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_fpn_gn_ws-all_2x_coco.py' 2 | # model settings 3 | conv_cfg = dict(type='ConvWS') 4 | norm_cfg = dict(type='GN', num_groups=32, requires_grad=True) 5 | model = dict( 6 | pretrained='open-mmlab://jhu/resnext50_32x4d_gn_ws', 7 | backbone=dict( 8 | type='ResNeXt', 9 | depth=50, 10 | groups=32, 11 | base_width=4, 12 | num_stages=4, 13 | out_indices=(0, 1, 2, 3), 14 | frozen_stages=1, 15 | style='pytorch', 16 | conv_cfg=conv_cfg, 17 | norm_cfg=norm_cfg)) 18 | -------------------------------------------------------------------------------- /configs/gn/mask_rcnn_r101_fpn_gn-all_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_fpn_gn-all_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://detectron/resnet101_gn', backbone=dict(depth=101)) 4 | -------------------------------------------------------------------------------- /configs/gn/mask_rcnn_r101_fpn_gn-all_3x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r101_fpn_gn-all_2x_coco.py' 2 | 3 | # learning policy 4 | lr_config = dict(step=[28, 34]) 5 | total_epochs = 36 6 | -------------------------------------------------------------------------------- /configs/gn/mask_rcnn_r50_fpn_gn-all_3x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_fpn_gn-all_2x_coco.py' 2 | 3 | # learning policy 4 | lr_config = dict(step=[28, 34]) 5 | total_epochs = 36 6 | -------------------------------------------------------------------------------- /configs/gn/mask_rcnn_r50_fpn_gn-all_contrib_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py' 2 | norm_cfg = dict(type='GN', num_groups=32, requires_grad=True) 3 | model = dict( 4 | pretrained='open-mmlab://contrib/resnet50_gn', 5 | backbone=dict(norm_cfg=norm_cfg), 6 | neck=dict(norm_cfg=norm_cfg), 7 | roi_head=dict( 8 | bbox_head=dict( 9 | type='Shared4Conv1FCBBoxHead', 10 | conv_out_channels=256, 11 | norm_cfg=norm_cfg), 12 | mask_head=dict(norm_cfg=norm_cfg))) 13 | # learning policy 14 | lr_config = dict(step=[16, 22]) 15 | total_epochs = 24 16 | -------------------------------------------------------------------------------- /configs/gn/mask_rcnn_r50_fpn_gn-all_contrib_3x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_fpn_gn-all_contrib_2x_coco.py' 2 | 3 | # learning policy 4 | lr_config = dict(step=[28, 34]) 5 | total_epochs = 36 6 | -------------------------------------------------------------------------------- /configs/grid_rcnn/grid_rcnn_r101_fpn_gn-head_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './grid_rcnn_r50_fpn_gn-head_2x_coco.py' 2 | 3 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 4 | -------------------------------------------------------------------------------- /configs/grid_rcnn/grid_rcnn_r50_fpn_gn-head_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = ['../grid_rcnn/grid_rcnn_r50_fpn_gn-head_2x_coco.py'] 2 | # learning policy 3 | lr_config = dict( 4 | policy='step', 5 | warmup='linear', 6 | warmup_iters=500, 7 | warmup_ratio=0.001, 8 | step=[8, 11]) 9 | checkpoint_config = dict(interval=1) 10 | # runtime settings 11 | total_epochs = 12 12 | -------------------------------------------------------------------------------- /configs/grid_rcnn/grid_rcnn_x101_32x4d_fpn_gn-head_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './grid_rcnn_r50_fpn_gn-head_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | style='pytorch')) 13 | # optimizer 14 | optimizer = dict(type='SGD', lr=0.02, momentum=0.9, weight_decay=0.0001) 15 | optimizer_config = dict(grad_clip=None) 16 | # learning policy 17 | lr_config = dict( 18 | policy='step', 19 | warmup='linear', 20 | warmup_iters=3665, 21 | warmup_ratio=1.0 / 80, 22 | step=[17, 23]) 23 | total_epochs = 25 24 | -------------------------------------------------------------------------------- /configs/grid_rcnn/grid_rcnn_x101_64x4d_fpn_gn-head_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './grid_rcnn_x101_32x4d_fpn_gn-head_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | style='pytorch')) 13 | -------------------------------------------------------------------------------- /configs/groie/faster_rcnn_r50_fpn_groie_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | # model settings 3 | model = dict( 4 | roi_head=dict( 5 | bbox_roi_extractor=dict( 6 | type='GenericRoIExtractor', 7 | aggregation='sum', 8 | roi_layer=dict(type='RoIAlign', out_size=7, sample_num=2), 9 | out_channels=256, 10 | featmap_strides=[4, 8, 16, 32], 11 | pre_cfg=dict( 12 | type='ConvModule', 13 | in_channels=256, 14 | out_channels=256, 15 | kernel_size=5, 16 | padding=2, 17 | inplace=False, 18 | ), 19 | post_cfg=dict( 20 | type='GeneralizedAttention', 21 | in_channels=256, 22 | spatial_range=-1, 23 | num_heads=6, 24 | attention_type='0100', 25 | kv_stride=2)))) 26 | -------------------------------------------------------------------------------- /configs/guided_anchoring/ga_faster_r101_caffe_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './ga_faster_r50_caffe_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://detectron2/resnet101_caffe', 4 | backbone=dict(depth=101)) 5 | -------------------------------------------------------------------------------- /configs/guided_anchoring/ga_faster_x101_32x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './ga_faster_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/guided_anchoring/ga_faster_x101_64x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './ga_faster_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/guided_anchoring/ga_retinanet_r101_caffe_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './ga_retinanet_r50_caffe_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://detectron2/resnet101_caffe', 4 | backbone=dict(depth=101)) 5 | -------------------------------------------------------------------------------- /configs/guided_anchoring/ga_retinanet_x101_32x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './ga_retinanet_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/guided_anchoring/ga_retinanet_x101_64x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './ga_retinanet_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/guided_anchoring/ga_rpn_r101_caffe_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './ga_rpn_r50_caffe_fpn_1x_coco.py' 2 | # model settings 3 | model = dict( 4 | pretrained='open-mmlab://detectron2/resnet101_caffe', 5 | backbone=dict(depth=101)) 6 | -------------------------------------------------------------------------------- /configs/guided_anchoring/ga_rpn_x101_32x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './ga_rpn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/guided_anchoring/ga_rpn_x101_64x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './ga_rpn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/hrnet/cascade_mask_rcnn_hrnetv2p_w18_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_mask_rcnn_hrnetv2p_w32_20e_coco.py' 2 | # model settings 3 | model = dict( 4 | pretrained='open-mmlab://msra/hrnetv2_w18', 5 | backbone=dict( 6 | extra=dict( 7 | stage2=dict(num_channels=(18, 36)), 8 | stage3=dict(num_channels=(18, 36, 72)), 9 | stage4=dict(num_channels=(18, 36, 72, 144)))), 10 | neck=dict(type='HRFPN', in_channels=[18, 36, 72, 144], out_channels=256)) 11 | -------------------------------------------------------------------------------- /configs/hrnet/cascade_mask_rcnn_hrnetv2p_w40_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_mask_rcnn_hrnetv2p_w32_20e_coco.py' 2 | # model settings 3 | model = dict( 4 | pretrained='open-mmlab://msra/hrnetv2_w40', 5 | backbone=dict( 6 | type='HRNet', 7 | extra=dict( 8 | stage2=dict(num_channels=(40, 80)), 9 | stage3=dict(num_channels=(40, 80, 160)), 10 | stage4=dict(num_channels=(40, 80, 160, 320)))), 11 | neck=dict(type='HRFPN', in_channels=[40, 80, 160, 320], out_channels=256)) 12 | -------------------------------------------------------------------------------- /configs/hrnet/cascade_rcnn_hrnetv2p_w18_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_rcnn_hrnetv2p_w32_20e_coco.py' 2 | # model settings 3 | model = dict( 4 | pretrained='open-mmlab://msra/hrnetv2_w18', 5 | backbone=dict( 6 | extra=dict( 7 | stage2=dict(num_channels=(18, 36)), 8 | stage3=dict(num_channels=(18, 36, 72)), 9 | stage4=dict(num_channels=(18, 36, 72, 144)))), 10 | neck=dict(type='HRFPN', in_channels=[18, 36, 72, 144], out_channels=256)) 11 | -------------------------------------------------------------------------------- /configs/hrnet/cascade_rcnn_hrnetv2p_w40_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_rcnn_hrnetv2p_w32_20e_coco.py' 2 | # model settings 3 | model = dict( 4 | pretrained='open-mmlab://msra/hrnetv2_w40', 5 | backbone=dict( 6 | type='HRNet', 7 | extra=dict( 8 | stage2=dict(num_channels=(40, 80)), 9 | stage3=dict(num_channels=(40, 80, 160)), 10 | stage4=dict(num_channels=(40, 80, 160, 320)))), 11 | neck=dict(type='HRFPN', in_channels=[40, 80, 160, 320], out_channels=256)) 12 | -------------------------------------------------------------------------------- /configs/hrnet/faster_rcnn_hrnetv2p_w18_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_hrnetv2p_w32_1x_coco.py' 2 | # model settings 3 | model = dict( 4 | pretrained='open-mmlab://msra/hrnetv2_w18', 5 | backbone=dict( 6 | extra=dict( 7 | stage2=dict(num_channels=(18, 36)), 8 | stage3=dict(num_channels=(18, 36, 72)), 9 | stage4=dict(num_channels=(18, 36, 72, 144)))), 10 | neck=dict(type='HRFPN', in_channels=[18, 36, 72, 144], out_channels=256)) 11 | -------------------------------------------------------------------------------- /configs/hrnet/faster_rcnn_hrnetv2p_w18_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_hrnetv2p_w18_1x_coco.py' 2 | 3 | # learning policy 4 | lr_config = dict(step=[16, 22]) 5 | total_epochs = 24 6 | -------------------------------------------------------------------------------- /configs/hrnet/faster_rcnn_hrnetv2p_w32_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://msra/hrnetv2_w32', 4 | backbone=dict( 5 | _delete_=True, 6 | type='HRNet', 7 | extra=dict( 8 | stage1=dict( 9 | num_modules=1, 10 | num_branches=1, 11 | block='BOTTLENECK', 12 | num_blocks=(4, ), 13 | num_channels=(64, )), 14 | stage2=dict( 15 | num_modules=1, 16 | num_branches=2, 17 | block='BASIC', 18 | num_blocks=(4, 4), 19 | num_channels=(32, 64)), 20 | stage3=dict( 21 | num_modules=4, 22 | num_branches=3, 23 | block='BASIC', 24 | num_blocks=(4, 4, 4), 25 | num_channels=(32, 64, 128)), 26 | stage4=dict( 27 | num_modules=3, 28 | num_branches=4, 29 | block='BASIC', 30 | num_blocks=(4, 4, 4, 4), 31 | num_channels=(32, 64, 128, 256)))), 32 | neck=dict( 33 | _delete_=True, 34 | type='HRFPN', 35 | in_channels=[32, 64, 128, 256], 36 | out_channels=256)) 37 | -------------------------------------------------------------------------------- /configs/hrnet/faster_rcnn_hrnetv2p_w32_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_hrnetv2p_w32_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 22]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/hrnet/faster_rcnn_hrnetv2p_w40_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_hrnetv2p_w32_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://msra/hrnetv2_w40', 4 | backbone=dict( 5 | type='HRNet', 6 | extra=dict( 7 | stage2=dict(num_channels=(40, 80)), 8 | stage3=dict(num_channels=(40, 80, 160)), 9 | stage4=dict(num_channels=(40, 80, 160, 320)))), 10 | neck=dict(type='HRFPN', in_channels=[40, 80, 160, 320], out_channels=256)) 11 | -------------------------------------------------------------------------------- /configs/hrnet/faster_rcnn_hrnetv2p_w40_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_hrnetv2p_w40_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 22]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/hrnet/fcos_hrnetv2p_w18_gn-head_4x4_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fcos_hrnetv2p_w32_gn-head_4x4_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://msra/hrnetv2_w18', 4 | backbone=dict( 5 | extra=dict( 6 | stage2=dict(num_channels=(18, 36)), 7 | stage3=dict(num_channels=(18, 36, 72)), 8 | stage4=dict(num_channels=(18, 36, 72, 144)))), 9 | neck=dict(type='HRFPN', in_channels=[18, 36, 72, 144], out_channels=256)) 10 | -------------------------------------------------------------------------------- /configs/hrnet/fcos_hrnetv2p_w18_gn-head_4x4_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fcos_hrnetv2p_w18_gn-head_4x4_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 22]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/hrnet/fcos_hrnetv2p_w18_gn-head_mstrain_640-800_4x4_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fcos_hrnetv2p_w32_gn-head_mstrain_640-800_4x4_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://msra/hrnetv2_w18', 4 | backbone=dict( 5 | extra=dict( 6 | stage2=dict(num_channels=(18, 36)), 7 | stage3=dict(num_channels=(18, 36, 72)), 8 | stage4=dict(num_channels=(18, 36, 72, 144)))), 9 | neck=dict(type='HRFPN', in_channels=[18, 36, 72, 144], out_channels=256)) 10 | -------------------------------------------------------------------------------- /configs/hrnet/fcos_hrnetv2p_w32_gn-head_4x4_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fcos_hrnetv2p_w32_gn-head_4x4_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 22]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/hrnet/fcos_hrnetv2p_w40_gn-head_mstrain_640-800_4x4_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './fcos_hrnetv2p_w32_gn-head_mstrain_640-800_4x4_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://msra/hrnetv2_w40', 4 | backbone=dict( 5 | type='HRNet', 6 | extra=dict( 7 | stage2=dict(num_channels=(40, 80)), 8 | stage3=dict(num_channels=(40, 80, 160)), 9 | stage4=dict(num_channels=(40, 80, 160, 320)))), 10 | neck=dict(type='HRFPN', in_channels=[40, 80, 160, 320], out_channels=256)) 11 | -------------------------------------------------------------------------------- /configs/hrnet/htc_hrnetv2p_w18_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './htc_hrnetv2p_w32_20e_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://msra/hrnetv2_w18', 4 | backbone=dict( 5 | extra=dict( 6 | stage2=dict(num_channels=(18, 36)), 7 | stage3=dict(num_channels=(18, 36, 72)), 8 | stage4=dict(num_channels=(18, 36, 72, 144)))), 9 | neck=dict(type='HRFPN', in_channels=[18, 36, 72, 144], out_channels=256)) 10 | -------------------------------------------------------------------------------- /configs/hrnet/htc_hrnetv2p_w32_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../htc/htc_r50_fpn_20e_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://msra/hrnetv2_w32', 4 | backbone=dict( 5 | _delete_=True, 6 | type='HRNet', 7 | extra=dict( 8 | stage1=dict( 9 | num_modules=1, 10 | num_branches=1, 11 | block='BOTTLENECK', 12 | num_blocks=(4, ), 13 | num_channels=(64, )), 14 | stage2=dict( 15 | num_modules=1, 16 | num_branches=2, 17 | block='BASIC', 18 | num_blocks=(4, 4), 19 | num_channels=(32, 64)), 20 | stage3=dict( 21 | num_modules=4, 22 | num_branches=3, 23 | block='BASIC', 24 | num_blocks=(4, 4, 4), 25 | num_channels=(32, 64, 128)), 26 | stage4=dict( 27 | num_modules=3, 28 | num_branches=4, 29 | block='BASIC', 30 | num_blocks=(4, 4, 4, 4), 31 | num_channels=(32, 64, 128, 256)))), 32 | neck=dict( 33 | _delete_=True, 34 | type='HRFPN', 35 | in_channels=[32, 64, 128, 256], 36 | out_channels=256)) 37 | -------------------------------------------------------------------------------- /configs/hrnet/htc_hrnetv2p_w40_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './htc_hrnetv2p_w32_20e_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://msra/hrnetv2_w40', 4 | backbone=dict( 5 | type='HRNet', 6 | extra=dict( 7 | stage2=dict(num_channels=(40, 80)), 8 | stage3=dict(num_channels=(40, 80, 160)), 9 | stage4=dict(num_channels=(40, 80, 160, 320)))), 10 | neck=dict(type='HRFPN', in_channels=[40, 80, 160, 320], out_channels=256)) 11 | -------------------------------------------------------------------------------- /configs/hrnet/htc_hrnetv2p_w40_28e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './htc_hrnetv2p_w40_20e_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[24, 27]) 4 | total_epochs = 28 5 | -------------------------------------------------------------------------------- /configs/hrnet/htc_x101_64x4d_fpn_16x1_28e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../htc/htc_x101_64x4d_fpn_16x1_20e_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[24, 27]) 4 | total_epochs = 28 5 | -------------------------------------------------------------------------------- /configs/hrnet/mask_rcnn_hrnetv2p_w18_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_hrnetv2p_w32_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://msra/hrnetv2_w18', 4 | backbone=dict( 5 | extra=dict( 6 | stage2=dict(num_channels=(18, 36)), 7 | stage3=dict(num_channels=(18, 36, 72)), 8 | stage4=dict(num_channels=(18, 36, 72, 144)))), 9 | neck=dict(type='HRFPN', in_channels=[18, 36, 72, 144], out_channels=256)) 10 | -------------------------------------------------------------------------------- /configs/hrnet/mask_rcnn_hrnetv2p_w18_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_hrnetv2p_w18_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 22]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/hrnet/mask_rcnn_hrnetv2p_w32_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://msra/hrnetv2_w32', 4 | backbone=dict( 5 | _delete_=True, 6 | type='HRNet', 7 | extra=dict( 8 | stage1=dict( 9 | num_modules=1, 10 | num_branches=1, 11 | block='BOTTLENECK', 12 | num_blocks=(4, ), 13 | num_channels=(64, )), 14 | stage2=dict( 15 | num_modules=1, 16 | num_branches=2, 17 | block='BASIC', 18 | num_blocks=(4, 4), 19 | num_channels=(32, 64)), 20 | stage3=dict( 21 | num_modules=4, 22 | num_branches=3, 23 | block='BASIC', 24 | num_blocks=(4, 4, 4), 25 | num_channels=(32, 64, 128)), 26 | stage4=dict( 27 | num_modules=3, 28 | num_branches=4, 29 | block='BASIC', 30 | num_blocks=(4, 4, 4, 4), 31 | num_channels=(32, 64, 128, 256)))), 32 | neck=dict( 33 | _delete_=True, 34 | type='HRFPN', 35 | in_channels=[32, 64, 128, 256], 36 | out_channels=256)) 37 | -------------------------------------------------------------------------------- /configs/hrnet/mask_rcnn_hrnetv2p_w32_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_hrnetv2p_w32_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 22]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/hrnet/mask_rcnn_hrnetv2p_w40_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_hrnetv2p_w18_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://msra/hrnetv2_w40', 4 | backbone=dict( 5 | type='HRNet', 6 | extra=dict( 7 | stage2=dict(num_channels=(40, 80)), 8 | stage3=dict(num_channels=(40, 80, 160)), 9 | stage4=dict(num_channels=(40, 80, 160, 320)))), 10 | neck=dict(type='HRFPN', in_channels=[40, 80, 160, 320], out_channels=256)) 11 | -------------------------------------------------------------------------------- /configs/hrnet/mask_rcnn_hrnetv2p_w40_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_hrnetv2p_w40_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 22]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/htc/htc_r101_fpn_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './htc_r50_fpn_1x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | # learning policy 4 | lr_config = dict(step=[16, 19]) 5 | total_epochs = 20 6 | -------------------------------------------------------------------------------- /configs/htc/htc_r50_fpn_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './htc_r50_fpn_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 19]) 4 | total_epochs = 20 5 | -------------------------------------------------------------------------------- /configs/htc/htc_x101_32x4d_fpn_16x1_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './htc_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | norm_eval=True, 14 | style='pytorch')) 15 | data = dict(samples_per_gpu=1, workers_per_gpu=1) 16 | # learning policy 17 | lr_config = dict(step=[16, 19]) 18 | total_epochs = 20 19 | -------------------------------------------------------------------------------- /configs/htc/htc_x101_64x4d_fpn_16x1_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './htc_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | norm_eval=True, 14 | style='pytorch')) 15 | data = dict(samples_per_gpu=1, workers_per_gpu=1) 16 | # learning policy 17 | lr_config = dict(step=[16, 19]) 18 | total_epochs = 20 19 | -------------------------------------------------------------------------------- /configs/instaboost/cascade_mask_rcnn_r101_fpn_instaboost_4x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_mask_rcnn_r50_fpn_instaboost_4x_coco.py' 2 | 3 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 4 | -------------------------------------------------------------------------------- /configs/instaboost/cascade_mask_rcnn_r50_fpn_instaboost_4x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../cascade_rcnn/cascade_mask_rcnn_r50_fpn_1x_coco.py' 2 | img_norm_cfg = dict( 3 | mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True) 4 | train_pipeline = [ 5 | dict(type='LoadImageFromFile'), 6 | dict( 7 | type='InstaBoost', 8 | action_candidate=('normal', 'horizontal', 'skip'), 9 | action_prob=(1, 0, 0), 10 | scale=(0.8, 1.2), 11 | dx=15, 12 | dy=15, 13 | theta=(-1, 1), 14 | color_prob=0.5, 15 | hflag=False, 16 | aug_ratio=0.5), 17 | dict(type='LoadAnnotations', with_bbox=True, with_mask=True), 18 | dict(type='Resize', img_scale=(1333, 800), keep_ratio=True), 19 | dict(type='RandomFlip', flip_ratio=0.5), 20 | dict(type='Normalize', **img_norm_cfg), 21 | dict(type='Pad', size_divisor=32), 22 | dict(type='DefaultFormatBundle'), 23 | dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks']), 24 | ] 25 | data = dict(train=dict(pipeline=train_pipeline)) 26 | # learning policy 27 | lr_config = dict(step=[32, 44]) 28 | total_epochs = 48 29 | -------------------------------------------------------------------------------- /configs/instaboost/cascade_mask_rcnn_x101_64x4d_fpn_instaboost_4x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './cascade_mask_rcnn_r50_fpn_instaboost_4x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/instaboost/mask_rcnn_r101_fpn_instaboost_4x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_fpn_instaboost_4x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/instaboost/mask_rcnn_r50_fpn_instaboost_4x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py' 2 | img_norm_cfg = dict( 3 | mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True) 4 | train_pipeline = [ 5 | dict(type='LoadImageFromFile'), 6 | dict( 7 | type='InstaBoost', 8 | action_candidate=('normal', 'horizontal', 'skip'), 9 | action_prob=(1, 0, 0), 10 | scale=(0.8, 1.2), 11 | dx=15, 12 | dy=15, 13 | theta=(-1, 1), 14 | color_prob=0.5, 15 | hflag=False, 16 | aug_ratio=0.5), 17 | dict(type='LoadAnnotations', with_bbox=True, with_mask=True), 18 | dict(type='Resize', img_scale=(1333, 800), keep_ratio=True), 19 | dict(type='RandomFlip', flip_ratio=0.5), 20 | dict(type='Normalize', **img_norm_cfg), 21 | dict(type='Pad', size_divisor=32), 22 | dict(type='DefaultFormatBundle'), 23 | dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks']), 24 | ] 25 | data = dict(train=dict(pipeline=train_pipeline)) 26 | # learning policy 27 | lr_config = dict(step=[32, 44]) 28 | total_epochs = 48 29 | -------------------------------------------------------------------------------- /configs/instaboost/mask_rcnn_x101_64x4d_fpn_instaboost_4x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_fpn_instaboost_4x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/legacy_1.x/mask_rcnn_r50_fpn_1x_coco_v1.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/mask_rcnn_r50_fpn.py', 3 | '../_base_/datasets/coco_instance.py', 4 | '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py' 5 | ] 6 | 7 | model = dict( 8 | rpn_head=dict( 9 | anchor_generator=dict(type='LegacyAnchorGenerator', center_offset=0.5), 10 | bbox_coder=dict(type='LegacyDeltaXYWHBBoxCoder'), 11 | loss_bbox=dict(type='SmoothL1Loss', beta=1.0 / 9.0, loss_weight=1.0)), 12 | roi_head=dict( 13 | bbox_roi_extractor=dict( 14 | type='SingleRoIExtractor', 15 | roi_layer=dict( 16 | type='RoIAlign', out_size=7, sample_num=2, aligned=False)), 17 | mask_roi_extractor=dict( 18 | type='SingleRoIExtractor', 19 | roi_layer=dict( 20 | type='RoIAlign', out_size=14, sample_num=2, aligned=False)), 21 | bbox_head=dict( 22 | bbox_coder=dict(type='LegacyDeltaXYWHBBoxCoder'), 23 | loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)))) 24 | # model training and testing settings 25 | train_cfg = dict( 26 | rpn_proposal=dict(nms_post=2000, max_num=2000), 27 | rcnn=dict(assigner=dict(match_low_quality=True))) 28 | -------------------------------------------------------------------------------- /configs/legacy_1.x/retinanet_r50_fpn_1x_coco_v1.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/retinanet_r50_fpn.py', 3 | '../_base_/datasets/coco_detection.py', 4 | '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py' 5 | ] 6 | model = dict( 7 | bbox_head=dict( 8 | type='RetinaHead', 9 | anchor_generator=dict( 10 | type='LegacyAnchorGenerator', 11 | center_offset=0.5, 12 | octave_base_scale=4, 13 | scales_per_octave=3, 14 | ratios=[0.5, 1.0, 2.0], 15 | strides=[8, 16, 32, 64, 128]), 16 | bbox_coder=dict(type='LegacyDeltaXYWHBBoxCoder'), 17 | loss_bbox=dict(type='SmoothL1Loss', beta=0.11, loss_weight=1.0))) 18 | -------------------------------------------------------------------------------- /configs/libra_rcnn/libra_faster_rcnn_r101_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './libra_faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/libra_rcnn/libra_faster_rcnn_x101_64x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './libra_faster_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/libra_rcnn/libra_retinanet_r50_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../retinanet/retinanet_r50_fpn_1x_coco.py' 2 | # model settings 3 | model = dict( 4 | neck=[ 5 | dict( 6 | type='FPN', 7 | in_channels=[256, 512, 1024, 2048], 8 | out_channels=256, 9 | start_level=1, 10 | add_extra_convs='on_input', 11 | num_outs=5), 12 | dict( 13 | type='BFP', 14 | in_channels=256, 15 | num_levels=5, 16 | refine_level=1, 17 | refine_type='non_local') 18 | ], 19 | bbox_head=dict( 20 | loss_bbox=dict( 21 | _delete_=True, 22 | type='BalancedL1Loss', 23 | alpha=0.5, 24 | gamma=1.5, 25 | beta=0.11, 26 | loss_weight=1.0))) 27 | -------------------------------------------------------------------------------- /configs/lvis/mask_rcnn_r101_fpn_sample1e-3_mstrain_2x_lvis.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_fpn_sample1e-3_mstrain_2x_lvis.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/lvis/mask_rcnn_r50_fpn_sample1e-3_mstrain_2x_lvis.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/mask_rcnn_r50_fpn.py', 3 | '../_base_/datasets/lvis_instance.py', 4 | '../_base_/schedules/schedule_2x.py', '../_base_/default_runtime.py' 5 | ] 6 | model = dict( 7 | roi_head=dict( 8 | bbox_head=dict(num_classes=1230), mask_head=dict(num_classes=1230))) 9 | test_cfg = dict( 10 | rcnn=dict( 11 | score_thr=0.0001, 12 | # LVIS allows up to 300 13 | max_per_img=300)) 14 | img_norm_cfg = dict( 15 | mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True) 16 | train_pipeline = [ 17 | dict(type='LoadImageFromFile'), 18 | dict(type='LoadAnnotations', with_bbox=True, with_mask=True), 19 | dict( 20 | type='Resize', 21 | img_scale=[(1333, 640), (1333, 672), (1333, 704), (1333, 736), 22 | (1333, 768), (1333, 800)], 23 | multiscale_mode='value', 24 | keep_ratio=True), 25 | dict(type='RandomFlip', flip_ratio=0.5), 26 | dict(type='Normalize', **img_norm_cfg), 27 | dict(type='Pad', size_divisor=32), 28 | dict(type='DefaultFormatBundle'), 29 | dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks']), 30 | ] 31 | data = dict(train=dict(dataset=dict(pipeline=train_pipeline))) 32 | -------------------------------------------------------------------------------- /configs/lvis/mask_rcnn_x101_32x4d_fpn_sample1e-3_mstrain_2x_lvis.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_fpn_sample1e-3_mstrain_2x_lvis.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/lvis/mask_rcnn_x101_64x4d_fpn_sample1e-3_mstrain_2x_lvis.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_fpn_sample1e-3_mstrain_2x_lvis.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/mask_rcnn/mask_rcnn_r101_caffe_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_caffe_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://detectron2/resnet101_caffe', 4 | backbone=dict(depth=101)) 5 | -------------------------------------------------------------------------------- /configs/mask_rcnn/mask_rcnn_r101_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_fpn_1x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/mask_rcnn/mask_rcnn_r101_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_fpn_2x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/mask_rcnn/mask_rcnn_r50_caffe_fpn_mstrain-poly_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_caffe_fpn_mstrain-poly_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 23]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/mask_rcnn/mask_rcnn_r50_caffe_fpn_mstrain-poly_3x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r50_caffe_fpn_mstrain-poly_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[28, 34]) 4 | total_epochs = 36 5 | -------------------------------------------------------------------------------- /configs/mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/mask_rcnn_r50_fpn.py', 3 | '../_base_/datasets/coco_instance.py', 4 | '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py' 5 | ] 6 | -------------------------------------------------------------------------------- /configs/mask_rcnn/mask_rcnn_r50_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/mask_rcnn_r50_fpn.py', 3 | '../_base_/datasets/coco_instance.py', 4 | '../_base_/schedules/schedule_2x.py', '../_base_/default_runtime.py' 5 | ] 6 | -------------------------------------------------------------------------------- /configs/mask_rcnn/mask_rcnn_r50_fpn_poly_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/mask_rcnn_r50_fpn.py', 3 | '../_base_/datasets/coco_instance.py', 4 | '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py' 5 | ] 6 | 7 | img_norm_cfg = dict( 8 | mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True) 9 | train_pipeline = [ 10 | dict(type='LoadImageFromFile'), 11 | dict( 12 | type='LoadAnnotations', 13 | with_bbox=True, 14 | with_mask=True, 15 | poly2mask=False), 16 | dict(type='Resize', img_scale=(1333, 800), keep_ratio=True), 17 | dict(type='RandomFlip', flip_ratio=0.5), 18 | dict(type='Normalize', **img_norm_cfg), 19 | dict(type='Pad', size_divisor=32), 20 | dict(type='DefaultFormatBundle'), 21 | dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks']), 22 | ] 23 | data = dict(train=dict(pipeline=train_pipeline)) 24 | -------------------------------------------------------------------------------- /configs/mask_rcnn/mask_rcnn_x101_32x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r101_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/mask_rcnn/mask_rcnn_x101_32x4d_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_r101_fpn_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/mask_rcnn/mask_rcnn_x101_64x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_x101_32x4d_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/mask_rcnn/mask_rcnn_x101_64x4d_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_x101_32x4d_fpn_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/ms_rcnn/ms_rcnn_r101_caffe_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './ms_rcnn_r50_caffe_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://detectron2/resnet101_caffe', 4 | backbone=dict(depth=101)) 5 | -------------------------------------------------------------------------------- /configs/ms_rcnn/ms_rcnn_r101_caffe_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './ms_rcnn_r101_caffe_fpn_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 22]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/ms_rcnn/ms_rcnn_r50_caffe_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r50_caffe_fpn_1x_coco.py' 2 | model = dict( 3 | type='MaskScoringRCNN', 4 | roi_head=dict( 5 | type='MaskScoringRoIHead', 6 | mask_iou_head=dict( 7 | type='MaskIoUHead', 8 | num_convs=4, 9 | num_fcs=2, 10 | roi_feat_size=14, 11 | in_channels=256, 12 | conv_out_channels=256, 13 | fc_out_channels=1024, 14 | num_classes=80))) 15 | # model training and testing settings 16 | train_cfg = dict(rcnn=dict(mask_thr_binary=0.5)) 17 | -------------------------------------------------------------------------------- /configs/ms_rcnn/ms_rcnn_r50_caffe_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './ms_rcnn_r50_caffe_fpn_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 22]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/ms_rcnn/ms_rcnn_r50_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | type='MaskScoringRCNN', 4 | roi_head=dict( 5 | type='MaskScoringRoIHead', 6 | mask_iou_head=dict( 7 | type='MaskIoUHead', 8 | num_convs=4, 9 | num_fcs=2, 10 | roi_feat_size=14, 11 | in_channels=256, 12 | conv_out_channels=256, 13 | fc_out_channels=1024, 14 | num_classes=80))) 15 | # model training and testing settings 16 | train_cfg = dict(rcnn=dict(mask_thr_binary=0.5)) 17 | -------------------------------------------------------------------------------- /configs/ms_rcnn/ms_rcnn_x101_32x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './ms_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/ms_rcnn/ms_rcnn_x101_64x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './ms_rcnn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/ms_rcnn/ms_rcnn_x101_64x4d_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './ms_rcnn_x101_64x4d_fpn_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 22]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/pafpn/README.md: -------------------------------------------------------------------------------- 1 | # Path Aggregation Network for Instance Segmentation 2 | 3 | ## Introduction 4 | 5 | ``` 6 | @inproceedings{liu2018path, 7 | author = {Shu Liu and 8 | Lu Qi and 9 | Haifang Qin and 10 | Jianping Shi and 11 | Jiaya Jia}, 12 | title = {Path Aggregation Network for Instance Segmentation}, 13 | booktitle = {Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR)}, 14 | year = {2018} 15 | } 16 | ``` 17 | 18 | ## Results and Models 19 | 20 | ## Results and Models 21 | 22 | | Backbone | style | Lr schd | Mem (GB) | Inf time (fps) | box AP | mask AP | Download | 23 | |:-------------:|:----------:|:-------:|:--------:|:--------------:|:------:|:-------:|:--------:| 24 | | R-50-FPN | pytorch | 1x | 4.0 | 17.2 | 37.5 | | [model](https://open-mmlab.s3.ap-northeast-2.amazonaws.com/mmdetection/v2.0/pafpn/faster_rcnn_r50_pafpn_1x_coco/faster_rcnn_r50_pafpn_1x_coco_bbox_mAP-0.375_20200503_105836-b7b4b9bd.pth) | [log](https://open-mmlab.s3.ap-northeast-2.amazonaws.com/mmdetection/v2.0/pafpn/faster_rcnn_r50_pafpn_1x_coco/faster_rcnn_r50_pafpn_1x_coco_20200503_105836.log.json) | 25 | -------------------------------------------------------------------------------- /configs/pafpn/faster_rcnn_r50_pafpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | 3 | model = dict( 4 | neck=dict( 5 | type='PAFPN', 6 | in_channels=[256, 512, 1024, 2048], 7 | out_channels=256, 8 | num_outs=5)) 9 | -------------------------------------------------------------------------------- /configs/pascal_voc/README.md: -------------------------------------------------------------------------------- 1 | ## Results and Models 2 | 3 | | Architecture | Backbone | Style | Lr schd | Mem (GB) | Inf time (fps) | box AP | Download | 4 | |:------------:|:---------:|:-------:|:-------:|:--------:|:--------------:|:------:|:--------:| 5 | | Faster R-CNN | R-50 | pytorch | 1x | 2.6 | - | 79.5 |[model](https://open-mmlab.s3.ap-northeast-2.amazonaws.com/mmdetection/v2.0/pascal_voc/faster_rcnn_r50_fpn_1x_voc0712/faster_rcnn_r50_fpn_1x_voc0712_20200624-c9895d40.pth) | [log](https://open-mmlab.s3.ap-northeast-2.amazonaws.com/mmdetection/v2.0/pascal_voc/faster_rcnn_r50_fpn_1x_voc0712/20200623_015208.log.json) | 6 | | Retinanet | R-50 | pytorch | 1x | 2.1 | - | 77.3 |[model](https://open-mmlab.s3.ap-northeast-2.amazonaws.com/mmdetection/v2.0/pascal_voc/retinanet_r50_fpn_1x_voc0712/retinanet_r50_fpn_1x_voc0712_20200617-47cbdd0e.pth) | [log](https://open-mmlab.s3.ap-northeast-2.amazonaws.com/mmdetection/v2.0/pascal_voc/retinanet_r50_fpn_1x_voc0712/retinanet_r50_fpn_1x_voc0712_20200616_014642.log.json) | 7 | -------------------------------------------------------------------------------- /configs/pascal_voc/faster_rcnn_r50_fpn_1x_voc0712.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/faster_rcnn_r50_fpn.py', '../_base_/datasets/voc0712.py', 3 | '../_base_/default_runtime.py' 4 | ] 5 | model = dict(roi_head=dict(bbox_head=dict(num_classes=20))) 6 | # optimizer 7 | optimizer = dict(type='SGD', lr=0.01, momentum=0.9, weight_decay=0.0001) 8 | optimizer_config = dict(grad_clip=None) 9 | # learning policy 10 | # actual epoch = 3 * 3 = 9 11 | lr_config = dict(policy='step', step=[3]) 12 | # runtime settings 13 | total_epochs = 4 # actual epoch = 4 * 3 = 12 14 | -------------------------------------------------------------------------------- /configs/pascal_voc/retinanet_r50_fpn_1x_voc0712.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/retinanet_r50_fpn.py', '../_base_/datasets/voc0712.py', 3 | '../_base_/default_runtime.py' 4 | ] 5 | model = dict(bbox_head=dict(num_classes=20)) 6 | # optimizer 7 | optimizer = dict(type='SGD', lr=0.01, momentum=0.9, weight_decay=0.0001) 8 | optimizer_config = dict(grad_clip=None) 9 | # learning policy 10 | # actual epoch = 3 * 3 = 9 11 | lr_config = dict(policy='step', step=[3]) 12 | # runtime settings 13 | total_epochs = 4 # actual epoch = 4 * 3 = 12 14 | -------------------------------------------------------------------------------- /configs/pisa/pisa_faster_rcnn_r50_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py' 2 | 3 | model = dict( 4 | roi_head=dict( 5 | type='PISARoIHead', 6 | bbox_head=dict( 7 | loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)))) 8 | 9 | train_cfg = dict( 10 | rpn_proposal=dict( 11 | nms_across_levels=False, 12 | nms_pre=2000, 13 | nms_post=2000, 14 | max_num=2000, 15 | nms_thr=0.7, 16 | min_bbox_size=0), 17 | rcnn=dict( 18 | sampler=dict( 19 | type='ScoreHLRSampler', 20 | num=512, 21 | pos_fraction=0.25, 22 | neg_pos_ub=-1, 23 | add_gt_as_proposals=True, 24 | k=0.5, 25 | bias=0.), 26 | isr=dict(k=2, bias=0), 27 | carl=dict(k=1, bias=0.2))) 28 | 29 | test_cfg = dict( 30 | rpn=dict( 31 | nms_across_levels=False, 32 | nms_pre=2000, 33 | nms_post=2000, 34 | max_num=2000, 35 | nms_thr=0.7, 36 | min_bbox_size=0)) 37 | -------------------------------------------------------------------------------- /configs/pisa/pisa_faster_rcnn_x101_32x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_x101_32x4d_fpn_1x_coco.py' 2 | 3 | model = dict( 4 | roi_head=dict( 5 | type='PISARoIHead', 6 | bbox_head=dict( 7 | loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)))) 8 | 9 | train_cfg = dict( 10 | rpn_proposal=dict( 11 | nms_across_levels=False, 12 | nms_pre=2000, 13 | nms_post=2000, 14 | max_num=2000, 15 | nms_thr=0.7, 16 | min_bbox_size=0), 17 | rcnn=dict( 18 | sampler=dict( 19 | type='ScoreHLRSampler', 20 | num=512, 21 | pos_fraction=0.25, 22 | neg_pos_ub=-1, 23 | add_gt_as_proposals=True, 24 | k=0.5, 25 | bias=0.), 26 | isr=dict(k=2, bias=0), 27 | carl=dict(k=1, bias=0.2))) 28 | 29 | test_cfg = dict( 30 | rpn=dict( 31 | nms_across_levels=False, 32 | nms_pre=2000, 33 | nms_post=2000, 34 | max_num=2000, 35 | nms_thr=0.7, 36 | min_bbox_size=0)) 37 | -------------------------------------------------------------------------------- /configs/pisa/pisa_mask_rcnn_r50_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py' 2 | 3 | model = dict( 4 | roi_head=dict( 5 | type='PISARoIHead', 6 | bbox_head=dict( 7 | loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)))) 8 | 9 | train_cfg = dict( 10 | rpn_proposal=dict( 11 | nms_across_levels=False, 12 | nms_pre=2000, 13 | nms_post=2000, 14 | max_num=2000, 15 | nms_thr=0.7, 16 | min_bbox_size=0), 17 | rcnn=dict( 18 | sampler=dict( 19 | type='ScoreHLRSampler', 20 | num=512, 21 | pos_fraction=0.25, 22 | neg_pos_ub=-1, 23 | add_gt_as_proposals=True, 24 | k=0.5, 25 | bias=0.), 26 | isr=dict(k=2, bias=0), 27 | carl=dict(k=1, bias=0.2))) 28 | 29 | test_cfg = dict( 30 | rpn=dict( 31 | nms_across_levels=False, 32 | nms_pre=2000, 33 | nms_post=2000, 34 | max_num=2000, 35 | nms_thr=0.7, 36 | min_bbox_size=0)) 37 | -------------------------------------------------------------------------------- /configs/pisa/pisa_mask_rcnn_x101_32x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_x101_32x4d_fpn_1x_coco.py' 2 | 3 | model = dict( 4 | roi_head=dict( 5 | type='PISARoIHead', 6 | bbox_head=dict( 7 | loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)))) 8 | 9 | train_cfg = dict( 10 | rpn_proposal=dict( 11 | nms_across_levels=False, 12 | nms_pre=2000, 13 | nms_post=2000, 14 | max_num=2000, 15 | nms_thr=0.7, 16 | min_bbox_size=0), 17 | rcnn=dict( 18 | sampler=dict( 19 | type='ScoreHLRSampler', 20 | num=512, 21 | pos_fraction=0.25, 22 | neg_pos_ub=-1, 23 | add_gt_as_proposals=True, 24 | k=0.5, 25 | bias=0.), 26 | isr=dict(k=2, bias=0), 27 | carl=dict(k=1, bias=0.2))) 28 | 29 | test_cfg = dict( 30 | rpn=dict( 31 | nms_across_levels=False, 32 | nms_pre=2000, 33 | nms_post=2000, 34 | max_num=2000, 35 | nms_thr=0.7, 36 | min_bbox_size=0)) 37 | -------------------------------------------------------------------------------- /configs/pisa/pisa_retinanet_r50_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../retinanet/retinanet_r50_fpn_1x_coco.py' 2 | 3 | model = dict( 4 | bbox_head=dict( 5 | type='PISARetinaHead', 6 | loss_bbox=dict(type='SmoothL1Loss', beta=0.11, loss_weight=1.0))) 7 | 8 | train_cfg = dict(isr=dict(k=2., bias=0.), carl=dict(k=1., bias=0.2)) 9 | -------------------------------------------------------------------------------- /configs/pisa/pisa_retinanet_x101_32x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../retinanet/retinanet_x101_32x4d_fpn_1x_coco.py' 2 | 3 | model = dict( 4 | bbox_head=dict( 5 | type='PISARetinaHead', 6 | loss_bbox=dict(type='SmoothL1Loss', beta=0.11, loss_weight=1.0))) 7 | 8 | train_cfg = dict(isr=dict(k=2., bias=0.), carl=dict(k=1., bias=0.2)) 9 | -------------------------------------------------------------------------------- /configs/pisa/pisa_ssd300_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../ssd/ssd300_coco.py' 2 | 3 | model = dict(bbox_head=dict(type='PISASSDHead')) 4 | 5 | train_cfg = dict(isr=dict(k=2., bias=0.), carl=dict(k=1., bias=0.2)) 6 | 7 | optimizer_config = dict( 8 | _delete_=True, grad_clip=dict(max_norm=35, norm_type=2)) 9 | -------------------------------------------------------------------------------- /configs/pisa/pisa_ssd512_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../ssd/ssd512_coco.py' 2 | 3 | model = dict(bbox_head=dict(type='PISASSDHead')) 4 | 5 | train_cfg = dict(isr=dict(k=2., bias=0.), carl=dict(k=1., bias=0.2)) 6 | 7 | optimizer_config = dict( 8 | _delete_=True, grad_clip=dict(max_norm=35, norm_type=2)) 9 | -------------------------------------------------------------------------------- /configs/point_rend/point_rend_r50_caffe_fpn_mstrain_3x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './point_rend_r50_caffe_fpn_mstrain_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[28, 34]) 4 | total_epochs = 36 5 | -------------------------------------------------------------------------------- /configs/regnet/faster_rcnn_regnetx-3.2GF_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './faster_rcnn_regnetx-3.2GF_fpn_1x_coco.py' 2 | lr_config = dict(step=[16, 22]) 3 | total_epochs = 24 4 | -------------------------------------------------------------------------------- /configs/regnet/mask_rcnn_regnetx-12GF_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_regnetx-3.2GF_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://regnetx_12gf', 4 | backbone=dict( 5 | type='RegNet', 6 | arch='regnetx_12gf', 7 | out_indices=(0, 1, 2, 3), 8 | frozen_stages=1, 9 | norm_cfg=dict(type='BN', requires_grad=True), 10 | norm_eval=True, 11 | style='pytorch'), 12 | neck=dict( 13 | type='FPN', 14 | in_channels=[224, 448, 896, 2240], 15 | out_channels=256, 16 | num_outs=5)) 17 | -------------------------------------------------------------------------------- /configs/regnet/mask_rcnn_regnetx-3.2GF_fpn_mdconv_c3-c5_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = 'mask_rcnn_regnetx-3.2GF_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://regnetx_3.2gf', 4 | backbone=dict( 5 | dcn=dict(type='DCNv2', deformable_groups=1, fallback_on_stride=False), 6 | stage_with_dcn=(False, True, True, True))) 7 | -------------------------------------------------------------------------------- /configs/regnet/mask_rcnn_regnetx-4GF_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_regnetx-3.2GF_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://regnetx_4.0gf', 4 | backbone=dict( 5 | type='RegNet', 6 | arch='regnetx_4.0gf', 7 | out_indices=(0, 1, 2, 3), 8 | frozen_stages=1, 9 | norm_cfg=dict(type='BN', requires_grad=True), 10 | norm_eval=True, 11 | style='pytorch'), 12 | neck=dict( 13 | type='FPN', 14 | in_channels=[80, 240, 560, 1360], 15 | out_channels=256, 16 | num_outs=5)) 17 | -------------------------------------------------------------------------------- /configs/regnet/mask_rcnn_regnetx-6.4GF_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_regnetx-3.2GF_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://regnetx_6.4gf', 4 | backbone=dict( 5 | type='RegNet', 6 | arch='regnetx_6.4gf', 7 | out_indices=(0, 1, 2, 3), 8 | frozen_stages=1, 9 | norm_cfg=dict(type='BN', requires_grad=True), 10 | norm_eval=True, 11 | style='pytorch'), 12 | neck=dict( 13 | type='FPN', 14 | in_channels=[168, 392, 784, 1624], 15 | out_channels=256, 16 | num_outs=5)) 17 | -------------------------------------------------------------------------------- /configs/regnet/mask_rcnn_regnetx-8GF_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './mask_rcnn_regnetx-3.2GF_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://regnetx_8.0gf', 4 | backbone=dict( 5 | type='RegNet', 6 | arch='regnetx_8.0gf', 7 | out_indices=(0, 1, 2, 3), 8 | frozen_stages=1, 9 | norm_cfg=dict(type='BN', requires_grad=True), 10 | norm_eval=True, 11 | style='pytorch'), 12 | neck=dict( 13 | type='FPN', 14 | in_channels=[80, 240, 720, 1920], 15 | out_channels=256, 16 | num_outs=5)) 17 | -------------------------------------------------------------------------------- /configs/regnet/retinanet_regnetx-1.6GF_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_regnetx-3.2GF_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://regnetx_1.6gf', 4 | backbone=dict( 5 | type='RegNet', 6 | arch='regnetx_1.6gf', 7 | out_indices=(0, 1, 2, 3), 8 | frozen_stages=1, 9 | norm_cfg=dict(type='BN', requires_grad=True), 10 | norm_eval=True, 11 | style='pytorch'), 12 | neck=dict( 13 | type='FPN', 14 | in_channels=[72, 168, 408, 912], 15 | out_channels=256, 16 | num_outs=5)) 17 | -------------------------------------------------------------------------------- /configs/regnet/retinanet_regnetx-800MF_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_regnetx-3.2GF_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://regnetx_800mf', 4 | backbone=dict( 5 | type='RegNet', 6 | arch='regnetx_800mf', 7 | out_indices=(0, 1, 2, 3), 8 | frozen_stages=1, 9 | norm_cfg=dict(type='BN', requires_grad=True), 10 | norm_eval=True, 11 | style='pytorch'), 12 | neck=dict( 13 | type='FPN', 14 | in_channels=[64, 128, 288, 672], 15 | out_channels=256, 16 | num_outs=5)) 17 | -------------------------------------------------------------------------------- /configs/reppoints/bbox_r50_grid_center_fpn_gn-neck+head_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './reppoints_moment_r50_fpn_gn-neck+head_1x_coco.py' 2 | model = dict(bbox_head=dict(transform_method='minmax', use_grid_points=True)) 3 | -------------------------------------------------------------------------------- /configs/reppoints/bbox_r50_grid_fpn_gn-neck+head_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './reppoints_moment_r50_fpn_gn-neck+head_1x_coco.py' 2 | model = dict(bbox_head=dict(transform_method='minmax', use_grid_points=True)) 3 | # training and testing settings 4 | train_cfg = dict( 5 | init=dict( 6 | assigner=dict( 7 | _delete_=True, 8 | type='MaxIoUAssigner', 9 | pos_iou_thr=0.5, 10 | neg_iou_thr=0.4, 11 | min_pos_iou=0, 12 | ignore_iof_thr=-1))) 13 | -------------------------------------------------------------------------------- /configs/reppoints/reppoints.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/longrongyang/LNCIS/6b0ad08b79e0b372ae90cba7a31db00d23f43b3d/configs/reppoints/reppoints.png -------------------------------------------------------------------------------- /configs/reppoints/reppoints_minmax_r50_fpn_gn-neck+head_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './reppoints_moment_r50_fpn_gn-neck+head_1x_coco.py' 2 | model = dict(bbox_head=dict(transform_method='minmax')) 3 | -------------------------------------------------------------------------------- /configs/reppoints/reppoints_moment_r101_fpn_dconv_c3-c5_gn-neck+head_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './reppoints_moment_r50_fpn_gn-neck+head_2x_coco.py' 2 | model = dict( 3 | pretrained='torchvision://resnet101', 4 | backbone=dict( 5 | depth=101, 6 | dcn=dict(type='DCN', deformable_groups=1, fallback_on_stride=False), 7 | stage_with_dcn=(False, True, True, True))) 8 | -------------------------------------------------------------------------------- /configs/reppoints/reppoints_moment_r101_fpn_gn-neck+head_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './reppoints_moment_r50_fpn_gn-neck+head_2x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/reppoints/reppoints_moment_r50_fpn_gn-neck+head_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './reppoints_moment_r50_fpn_1x_coco.py' 2 | norm_cfg = dict(type='GN', num_groups=32, requires_grad=True) 3 | model = dict(neck=dict(norm_cfg=norm_cfg), bbox_head=dict(norm_cfg=norm_cfg)) 4 | optimizer = dict(lr=0.01) 5 | -------------------------------------------------------------------------------- /configs/reppoints/reppoints_moment_r50_fpn_gn-neck+head_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './reppoints_moment_r50_fpn_gn-neck+head_1x_coco.py' 2 | lr_config = dict(step=[16, 22]) 3 | total_epochs = 24 4 | -------------------------------------------------------------------------------- /configs/reppoints/reppoints_moment_x101_fpn_dconv_c3-c5_gn-neck+head_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './reppoints_moment_r50_fpn_gn-neck+head_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch', 14 | dcn=dict(type='DCN', deformable_groups=1, fallback_on_stride=False), 15 | stage_with_dcn=(False, True, True, True))) 16 | -------------------------------------------------------------------------------- /configs/reppoints/reppoints_partial_minmax_r50_fpn_gn-neck+head_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './reppoints_moment_r50_fpn_gn-neck+head_1x_coco.py' 2 | model = dict(bbox_head=dict(transform_method='partial_minmax')) 3 | -------------------------------------------------------------------------------- /configs/res2net/cascade_mask_rcnn_r2_101_fpn_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../cascade_rcnn/cascade_mask_rcnn_r50_fpn_20e_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://res2net101_v1d_26w_4s', 4 | backbone=dict(type='Res2Net', depth=101, scales=4, base_width=26)) 5 | -------------------------------------------------------------------------------- /configs/res2net/cascade_rcnn_r2_101_fpn_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../cascade_rcnn/cascade_rcnn_r50_fpn_20e_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://res2net101_v1d_26w_4s', 4 | backbone=dict(type='Res2Net', depth=101, scales=4, base_width=26)) 5 | -------------------------------------------------------------------------------- /configs/res2net/faster_rcnn_r2_101_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../faster_rcnn/faster_rcnn_r50_fpn_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://res2net101_v1d_26w_4s', 4 | backbone=dict(type='Res2Net', depth=101, scales=4, base_width=26)) 5 | -------------------------------------------------------------------------------- /configs/res2net/htc_r2_101_fpn_20e_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../htc/htc_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://res2net101_v1d_26w_4s', 4 | backbone=dict(type='Res2Net', depth=101, scales=4, base_width=26)) 5 | # learning policy 6 | lr_config = dict(step=[16, 19]) 7 | total_epochs = 20 8 | -------------------------------------------------------------------------------- /configs/res2net/mask_rcnn_r2_101_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = '../mask_rcnn/mask_rcnn_r50_fpn_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://res2net101_v1d_26w_4s', 4 | backbone=dict(type='Res2Net', depth=101, scales=4, base_width=26)) 5 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_r101_caffe_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_r50_caffe_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://detectron2/resnet101_caffe', 4 | backbone=dict(depth=101)) 5 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_r101_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_r50_fpn_1x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_r101_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_r50_fpn_2x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_r50_caffe_fpn_mstrain_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_r50_caffe_fpn_mstrain_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 23]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_r50_caffe_fpn_mstrain_3x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_r50_caffe_fpn_mstrain_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[28, 34]) 4 | total_epochs = 36 5 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_r50_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/retinanet_r50_fpn.py', 3 | '../_base_/datasets/coco_detection.py', 4 | '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py' 5 | ] 6 | # optimizer 7 | optimizer = dict(type='SGD', lr=0.01, momentum=0.9, weight_decay=0.0001) 8 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_r50_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_r50_fpn_1x_coco.py' 2 | # learning policy 3 | lr_config = dict(step=[16, 22]) 4 | total_epochs = 24 5 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_x101_32x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_x101_32x4d_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_r50_fpn_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_x101_64x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/retinanet/retinanet_x101_64x4d_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './retinanet_r50_fpn_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/rpn/rpn_r101_caffe_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './rpn_r50_caffe_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://detectron2/resnet101_caffe', 4 | backbone=dict(depth=101)) 5 | -------------------------------------------------------------------------------- /configs/rpn/rpn_r101_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './rpn_r50_fpn_1x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/rpn/rpn_r101_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './rpn_r50_fpn_2x_coco.py' 2 | model = dict(pretrained='torchvision://resnet101', backbone=dict(depth=101)) 3 | -------------------------------------------------------------------------------- /configs/rpn/rpn_r50_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/rpn_r50_fpn.py', '../_base_/datasets/coco_detection.py', 3 | '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py' 4 | ] 5 | img_norm_cfg = dict( 6 | mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True) 7 | train_pipeline = [ 8 | dict(type='LoadImageFromFile'), 9 | dict(type='LoadAnnotations', with_bbox=True, with_label=False), 10 | dict(type='Resize', img_scale=(1333, 800), keep_ratio=True), 11 | dict(type='RandomFlip', flip_ratio=0.5), 12 | dict(type='Normalize', **img_norm_cfg), 13 | dict(type='Pad', size_divisor=32), 14 | dict(type='DefaultFormatBundle'), 15 | dict(type='Collect', keys=['img', 'gt_bboxes']), 16 | ] 17 | data = dict(train=dict(pipeline=train_pipeline)) 18 | evaluation = dict(interval=1, metric='proposal_fast') 19 | -------------------------------------------------------------------------------- /configs/rpn/rpn_r50_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './rpn_r50_fpn_1x_coco.py' 2 | 3 | # learning policy 4 | lr_config = dict(step=[16, 22]) 5 | total_epochs = 24 6 | -------------------------------------------------------------------------------- /configs/rpn/rpn_x101_32x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './rpn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/rpn/rpn_x101_32x4d_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './rpn_r50_fpn_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_32x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=32, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/rpn/rpn_x101_64x4d_fpn_1x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './rpn_r50_fpn_1x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/rpn/rpn_x101_64x4d_fpn_2x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = './rpn_r50_fpn_2x_coco.py' 2 | model = dict( 3 | pretrained='open-mmlab://resnext101_64x4d', 4 | backbone=dict( 5 | type='ResNeXt', 6 | depth=101, 7 | groups=64, 8 | base_width=4, 9 | num_stages=4, 10 | out_indices=(0, 1, 2, 3), 11 | frozen_stages=1, 12 | norm_cfg=dict(type='BN', requires_grad=True), 13 | style='pytorch')) 14 | -------------------------------------------------------------------------------- /configs/scratch/faster_rcnn_r50_fpn_gn-all_scratch_6x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/faster_rcnn_r50_fpn.py', 3 | '../_base_/datasets/coco_detection.py', 4 | '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py' 5 | ] 6 | norm_cfg = dict(type='GN', num_groups=32, requires_grad=True) 7 | model = dict( 8 | pretrained=None, 9 | backbone=dict( 10 | frozen_stages=-1, zero_init_residual=False, norm_cfg=norm_cfg), 11 | neck=dict(norm_cfg=norm_cfg), 12 | roi_head=dict( 13 | bbox_head=dict( 14 | type='Shared4Conv1FCBBoxHead', 15 | conv_out_channels=256, 16 | norm_cfg=norm_cfg))) 17 | # optimizer 18 | optimizer = dict(paramwise_cfg=dict(norm_decay_mult=0)) 19 | optimizer_config = dict(_delete_=True, grad_clip=None) 20 | # learning policy 21 | lr_config = dict(warmup_ratio=0.1, step=[65, 71]) 22 | total_epochs = 73 23 | -------------------------------------------------------------------------------- /configs/scratch/mask_rcnn_r50_fpn_gn-all_scratch_6x_coco.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/mask_rcnn_r50_fpn.py', 3 | '../_base_/datasets/coco_instance.py', 4 | '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py' 5 | ] 6 | norm_cfg = dict(type='GN', num_groups=32, requires_grad=True) 7 | model = dict( 8 | pretrained=None, 9 | backbone=dict( 10 | frozen_stages=-1, zero_init_residual=False, norm_cfg=norm_cfg), 11 | neck=dict(norm_cfg=norm_cfg), 12 | roi_head=dict( 13 | bbox_head=dict( 14 | type='Shared4Conv1FCBBoxHead', 15 | conv_out_channels=256, 16 | norm_cfg=norm_cfg), 17 | mask_head=dict(norm_cfg=norm_cfg))) 18 | # optimizer 19 | optimizer = dict(paramwise_cfg=dict(norm_decay_mult=0)) 20 | optimizer_config = dict(_delete_=True, grad_clip=None) 21 | # learning policy 22 | lr_config = dict(warmup_ratio=0.1, step=[65, 71]) 23 | total_epochs = 73 24 | -------------------------------------------------------------------------------- /configs/wider_face/README.md: -------------------------------------------------------------------------------- 1 | ## WIDER Face Dataset 2 | 3 | To use the WIDER Face dataset you need to download it 4 | and extract to the `data/WIDERFace` folder. Annotation in the VOC format 5 | can be found in this [repo](https://github.com/sovrasov/wider-face-pascal-voc-annotations.git). 6 | You should move the annotation files from `WIDER_train_annotations` and `WIDER_val_annotations` folders 7 | to the `Annotation` folders inside the corresponding directories `WIDER_train` and `WIDER_val`. 8 | Also annotation lists `val.txt` and `train.txt` should be copied to `data/WIDERFace` from `WIDER_train_annotations` and `WIDER_val_annotations`. 9 | The directory should be like this: 10 | 11 | ``` 12 | mmdetection 13 | ├── mmdet 14 | ├── tools 15 | ├── configs 16 | ├── data 17 | │ ├── WIDERFace 18 | │ │ ├── WIDER_train 19 | │ | │ ├──0--Parade 20 | │ | │ ├── ... 21 | │ | │ ├── Annotations 22 | │ │ ├── WIDER_val 23 | │ | │ ├──0--Parade 24 | │ | │ ├── ... 25 | │ | │ ├── Annotations 26 | │ │ ├── val.txt 27 | │ │ ├── train.txt 28 | 29 | ``` 30 | 31 | After that you can train the SSD300 on WIDER by launching training with the `ssd300_wider_face.py` config or 32 | create your own config based on the presented one. 33 | -------------------------------------------------------------------------------- /configs/wider_face/ssd300_wider_face.py: -------------------------------------------------------------------------------- 1 | _base_ = [ 2 | '../_base_/models/ssd300.py', '../_base_/datasets/wider_face.py', 3 | '../_base_/default_runtime.py' 4 | ] 5 | model = dict(bbox_head=dict(num_classes=1)) 6 | # optimizer 7 | optimizer = dict(type='SGD', lr=0.012, momentum=0.9, weight_decay=5e-4) 8 | optimizer_config = dict() 9 | # learning policy 10 | lr_config = dict( 11 | policy='step', 12 | warmup='linear', 13 | warmup_iters=1000, 14 | warmup_ratio=0.001, 15 | step=[16, 20]) 16 | # runtime settings 17 | total_epochs = 24 18 | log_config = dict(interval=1) 19 | -------------------------------------------------------------------------------- /demo/coco_test_12510.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/longrongyang/LNCIS/6b0ad08b79e0b372ae90cba7a31db00d23f43b3d/demo/coco_test_12510.jpg -------------------------------------------------------------------------------- /demo/corruptions_sev_3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/longrongyang/LNCIS/6b0ad08b79e0b372ae90cba7a31db00d23f43b3d/demo/corruptions_sev_3.png -------------------------------------------------------------------------------- /demo/data_pipeline.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/longrongyang/LNCIS/6b0ad08b79e0b372ae90cba7a31db00d23f43b3d/demo/data_pipeline.png -------------------------------------------------------------------------------- /demo/demo.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/longrongyang/LNCIS/6b0ad08b79e0b372ae90cba7a31db00d23f43b3d/demo/demo.jpg -------------------------------------------------------------------------------- /demo/image_demo.py: -------------------------------------------------------------------------------- 1 | from argparse import ArgumentParser 2 | 3 | from mmdet.apis import inference_detector, init_detector, show_result_pyplot 4 | 5 | 6 | def main(): 7 | parser = ArgumentParser() 8 | parser.add_argument('img', help='Image file') 9 | parser.add_argument('config', help='Config file') 10 | parser.add_argument('checkpoint', help='Checkpoint file') 11 | parser.add_argument( 12 | '--device', default='cuda:0', help='Device used for inference') 13 | parser.add_argument( 14 | '--score-thr', type=float, default=0.3, help='bbox score threshold') 15 | args = parser.parse_args() 16 | 17 | # build the model from a config file and a checkpoint file 18 | model = init_detector(args.config, args.checkpoint, device=args.device) 19 | # test a single image 20 | result = inference_detector(model, args.img) 21 | # show the results 22 | show_result_pyplot(model, args.img, result, score_thr=args.score_thr) 23 | 24 | 25 | if __name__ == '__main__': 26 | main() 27 | -------------------------------------------------------------------------------- /demo/loss_curve.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/longrongyang/LNCIS/6b0ad08b79e0b372ae90cba7a31db00d23f43b3d/demo/loss_curve.png -------------------------------------------------------------------------------- /docker/Dockerfile: -------------------------------------------------------------------------------- 1 | ARG PYTORCH="1.5" 2 | ARG CUDA="10.1" 3 | ARG CUDNN="7" 4 | 5 | FROM pytorch/pytorch:${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel 6 | 7 | ENV TORCH_CUDA_ARCH_LIST="6.0 6.1 7.0+PTX" 8 | ENV TORCH_NVCC_FLAGS="-Xfatbin -compress-all" 9 | ENV CMAKE_PREFIX_PATH="$(dirname $(which conda))/../" 10 | 11 | RUN apt-get update && apt-get install -y git ninja-build libglib2.0-0 libsm6 libxrender-dev libxext6 \ 12 | && apt-get clean \ 13 | && rm -rf /var/lib/apt/lists/* 14 | 15 | # Install mmdetection 16 | RUN conda clean --all 17 | RUN git clone https://github.com/open-mmlab/mmdetection.git /mmdetection 18 | WORKDIR /mmdetection 19 | ENV FORCE_CUDA="1" 20 | RUN pip install cython --no-cache-dir 21 | RUN pip install "git+https://github.com/open-mmlab/cocoapi.git#subdirectory=pycocotools" 22 | RUN pip install --no-cache-dir -e . 23 | -------------------------------------------------------------------------------- /docs/Makefile: -------------------------------------------------------------------------------- 1 | # Minimal makefile for Sphinx documentation 2 | # 3 | 4 | # You can set these variables from the command line, and also 5 | # from the environment for the first two. 6 | SPHINXOPTS ?= 7 | SPHINXBUILD ?= sphinx-build 8 | SOURCEDIR = . 9 | BUILDDIR = _build 10 | 11 | # Put it first so that "make" without argument is like "make help". 12 | help: 13 | @$(SPHINXBUILD) -M help "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O) 14 | 15 | .PHONY: help Makefile 16 | 17 | # Catch-all target: route all unknown targets to Sphinx using the new 18 | # "make mode" option. $(O) is meant as a shortcut for $(SPHINXOPTS). 19 | %: Makefile 20 | @$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O) 21 | -------------------------------------------------------------------------------- /docs/index.rst: -------------------------------------------------------------------------------- 1 | Welcome to MMDetection's documentation! 2 | ======================================= 3 | 4 | .. toctree:: 5 | :maxdepth: 2 6 | 7 | install.md 8 | getting_started.md 9 | config.md 10 | model_zoo.md 11 | tutorials/finetune.md 12 | tutorials/new_dataset.md 13 | tutorials/data_pipeline.md 14 | tutorials/new_modules.md 15 | compatibility.md 16 | changelog.md 17 | projects.md 18 | api.rst 19 | 20 | 21 | Indices and tables 22 | ================== 23 | 24 | * :ref:`genindex` 25 | * :ref:`search` 26 | -------------------------------------------------------------------------------- /docs/make.bat: -------------------------------------------------------------------------------- 1 | @ECHO OFF 2 | 3 | pushd %~dp0 4 | 5 | REM Command file for Sphinx documentation 6 | 7 | if "%SPHINXBUILD%" == "" ( 8 | set SPHINXBUILD=sphinx-build 9 | ) 10 | set SOURCEDIR=. 11 | set BUILDDIR=_build 12 | 13 | if "%1" == "" goto help 14 | 15 | %SPHINXBUILD% >NUL 2>NUL 16 | if errorlevel 9009 ( 17 | echo. 18 | echo.The 'sphinx-build' command was not found. Make sure you have Sphinx 19 | echo.installed, then set the SPHINXBUILD environment variable to point 20 | echo.to the full path of the 'sphinx-build' executable. Alternatively you 21 | echo.may add the Sphinx directory to PATH. 22 | echo. 23 | echo.If you don't have Sphinx installed, grab it from 24 | echo.http://sphinx-doc.org/ 25 | exit /b 1 26 | ) 27 | 28 | %SPHINXBUILD% -M %1 %SOURCEDIR% %BUILDDIR% %SPHINXOPTS% %O% 29 | goto end 30 | 31 | :help 32 | %SPHINXBUILD% -M help %SOURCEDIR% %BUILDDIR% %SPHINXOPTS% %O% 33 | 34 | :end 35 | popd 36 | -------------------------------------------------------------------------------- /mmdet/VERSION: -------------------------------------------------------------------------------- 1 | 2.2.0 2 | -------------------------------------------------------------------------------- /mmdet/__init__.py: -------------------------------------------------------------------------------- 1 | from .version import __version__, short_version 2 | 3 | __all__ = ['__version__', 'short_version'] 4 | -------------------------------------------------------------------------------- /mmdet/apis/__init__.py: -------------------------------------------------------------------------------- 1 | from .inference import (async_inference_detector, inference_detector, 2 | init_detector, show_result_pyplot) 3 | from .test import multi_gpu_test, single_gpu_test 4 | from .train import get_root_logger, set_random_seed, train_detector 5 | 6 | __all__ = [ 7 | 'get_root_logger', 'set_random_seed', 'train_detector', 'init_detector', 8 | 'async_inference_detector', 'inference_detector', 'show_result_pyplot', 9 | 'multi_gpu_test', 'single_gpu_test' 10 | ] 11 | -------------------------------------------------------------------------------- /mmdet/core/__init__.py: -------------------------------------------------------------------------------- 1 | from .anchor import * # noqa: F401, F403 2 | from .bbox import * # noqa: F401, F403 3 | from .evaluation import * # noqa: F401, F403 4 | from .fp16 import * # noqa: F401, F403 5 | from .mask import * # noqa: F401, F403 6 | from .post_processing import * # noqa: F401, F403 7 | from .utils import * # noqa: F401, F403 8 | -------------------------------------------------------------------------------- /mmdet/core/anchor/__init__.py: -------------------------------------------------------------------------------- 1 | from .anchor_generator import AnchorGenerator, LegacyAnchorGenerator 2 | from .builder import ANCHOR_GENERATORS, build_anchor_generator 3 | from .point_generator import PointGenerator 4 | from .utils import anchor_inside_flags, calc_region, images_to_levels 5 | 6 | __all__ = [ 7 | 'AnchorGenerator', 'LegacyAnchorGenerator', 'anchor_inside_flags', 8 | 'PointGenerator', 'images_to_levels', 'calc_region', 9 | 'build_anchor_generator', 'ANCHOR_GENERATORS' 10 | ] 11 | -------------------------------------------------------------------------------- /mmdet/core/anchor/builder.py: -------------------------------------------------------------------------------- 1 | from mmcv.utils import Registry, build_from_cfg 2 | 3 | ANCHOR_GENERATORS = Registry('Anchor generator') 4 | 5 | 6 | def build_anchor_generator(cfg, default_args=None): 7 | return build_from_cfg(cfg, ANCHOR_GENERATORS, default_args) 8 | -------------------------------------------------------------------------------- /mmdet/core/bbox/assigners/__init__.py: -------------------------------------------------------------------------------- 1 | from .approx_max_iou_assigner import ApproxMaxIoUAssigner 2 | from .assign_result import AssignResult 3 | from .atss_assigner import ATSSAssigner 4 | from .base_assigner import BaseAssigner 5 | from .center_region_assigner import CenterRegionAssigner 6 | from .max_iou_assigner import MaxIoUAssigner 7 | from .point_assigner import PointAssigner 8 | 9 | __all__ = [ 10 | 'BaseAssigner', 'MaxIoUAssigner', 'ApproxMaxIoUAssigner', 'AssignResult', 11 | 'PointAssigner', 'ATSSAssigner', 'CenterRegionAssigner' 12 | ] 13 | -------------------------------------------------------------------------------- /mmdet/core/bbox/assigners/base_assigner.py: -------------------------------------------------------------------------------- 1 | from abc import ABCMeta, abstractmethod 2 | 3 | 4 | class BaseAssigner(metaclass=ABCMeta): 5 | """Base assigner that assigns boxes to ground truth boxes.""" 6 | 7 | @abstractmethod 8 | def assign(self, bboxes, gt_bboxes, gt_bboxes_ignore=None, gt_labels=None): 9 | """Assign boxes to either a ground truth boxe or a negative boxes.""" 10 | pass 11 | -------------------------------------------------------------------------------- /mmdet/core/bbox/builder.py: -------------------------------------------------------------------------------- 1 | from mmcv.utils import Registry, build_from_cfg 2 | 3 | BBOX_ASSIGNERS = Registry('bbox_assigner') 4 | BBOX_SAMPLERS = Registry('bbox_sampler') 5 | BBOX_CODERS = Registry('bbox_coder') 6 | 7 | 8 | def build_assigner(cfg, **default_args): 9 | """Builder of box assigner.""" 10 | return build_from_cfg(cfg, BBOX_ASSIGNERS, default_args) 11 | 12 | 13 | def build_sampler(cfg, **default_args): 14 | """Builder of box sampler.""" 15 | return build_from_cfg(cfg, BBOX_SAMPLERS, default_args) 16 | 17 | 18 | def build_bbox_coder(cfg, **default_args): 19 | """Builder of box coder.""" 20 | return build_from_cfg(cfg, BBOX_CODERS, default_args) 21 | -------------------------------------------------------------------------------- /mmdet/core/bbox/coder/__init__.py: -------------------------------------------------------------------------------- 1 | from .base_bbox_coder import BaseBBoxCoder 2 | from .delta_xywh_bbox_coder import DeltaXYWHBBoxCoder 3 | from .legacy_delta_xywh_bbox_coder import LegacyDeltaXYWHBBoxCoder 4 | from .pseudo_bbox_coder import PseudoBBoxCoder 5 | from .tblr_bbox_coder import TBLRBBoxCoder 6 | 7 | __all__ = [ 8 | 'BaseBBoxCoder', 'PseudoBBoxCoder', 'DeltaXYWHBBoxCoder', 9 | 'LegacyDeltaXYWHBBoxCoder', 'TBLRBBoxCoder' 10 | ] 11 | -------------------------------------------------------------------------------- /mmdet/core/bbox/coder/base_bbox_coder.py: -------------------------------------------------------------------------------- 1 | from abc import ABCMeta, abstractmethod 2 | 3 | 4 | class BaseBBoxCoder(metaclass=ABCMeta): 5 | """Base bounding box coder.""" 6 | 7 | def __init__(self, **kwargs): 8 | pass 9 | 10 | @abstractmethod 11 | def encode(self, bboxes, gt_bboxes): 12 | """Encode deltas between bboxes and ground truth boxes.""" 13 | pass 14 | 15 | @abstractmethod 16 | def decode(self, bboxes, bboxes_pred): 17 | """Decode the predicted bboxes according to prediction and base 18 | boxes.""" 19 | pass 20 | -------------------------------------------------------------------------------- /mmdet/core/bbox/coder/pseudo_bbox_coder.py: -------------------------------------------------------------------------------- 1 | from ..builder import BBOX_CODERS 2 | from .base_bbox_coder import BaseBBoxCoder 3 | 4 | 5 | @BBOX_CODERS.register_module() 6 | class PseudoBBoxCoder(BaseBBoxCoder): 7 | """Pseudo bounding box coder.""" 8 | 9 | def __init__(self, **kwargs): 10 | super(BaseBBoxCoder, self).__init__(**kwargs) 11 | 12 | def encode(self, bboxes, gt_bboxes): 13 | """torch.Tensor: return the given ``bboxes``""" 14 | return gt_bboxes 15 | 16 | def decode(self, bboxes, pred_bboxes): 17 | """torch.Tensor: return the given ``pred_bboxes``""" 18 | return pred_bboxes 19 | -------------------------------------------------------------------------------- /mmdet/core/bbox/iou_calculators/__init__.py: -------------------------------------------------------------------------------- 1 | from .builder import build_iou_calculator 2 | from .iou2d_calculator import BboxOverlaps2D, bbox_overlaps 3 | 4 | __all__ = ['build_iou_calculator', 'BboxOverlaps2D', 'bbox_overlaps'] 5 | -------------------------------------------------------------------------------- /mmdet/core/bbox/iou_calculators/builder.py: -------------------------------------------------------------------------------- 1 | from mmcv.utils import Registry, build_from_cfg 2 | 3 | IOU_CALCULATORS = Registry('IoU calculator') 4 | 5 | 6 | def build_iou_calculator(cfg, default_args=None): 7 | """Builder of IoU calculator.""" 8 | return build_from_cfg(cfg, IOU_CALCULATORS, default_args) 9 | -------------------------------------------------------------------------------- /mmdet/core/bbox/samplers/__init__.py: -------------------------------------------------------------------------------- 1 | from .base_sampler import BaseSampler 2 | from .combined_sampler import CombinedSampler 3 | from .instance_balanced_pos_sampler import InstanceBalancedPosSampler 4 | from .iou_balanced_neg_sampler import IoUBalancedNegSampler 5 | from .ohem_sampler import OHEMSampler 6 | from .pseudo_sampler import PseudoSampler 7 | from .random_sampler import RandomSampler 8 | from .sampling_result import SamplingResult 9 | from .score_hlr_sampler import ScoreHLRSampler 10 | 11 | __all__ = [ 12 | 'BaseSampler', 'PseudoSampler', 'RandomSampler', 13 | 'InstanceBalancedPosSampler', 'IoUBalancedNegSampler', 'CombinedSampler', 14 | 'OHEMSampler', 'SamplingResult', 'ScoreHLRSampler' 15 | ] 16 | -------------------------------------------------------------------------------- /mmdet/core/bbox/samplers/combined_sampler.py: -------------------------------------------------------------------------------- 1 | from ..builder import BBOX_SAMPLERS, build_sampler 2 | from .base_sampler import BaseSampler 3 | 4 | 5 | @BBOX_SAMPLERS.register_module() 6 | class CombinedSampler(BaseSampler): 7 | """A sampler that combines positive sampler and negative sampler.""" 8 | 9 | def __init__(self, pos_sampler, neg_sampler, **kwargs): 10 | super(CombinedSampler, self).__init__(**kwargs) 11 | self.pos_sampler = build_sampler(pos_sampler, **kwargs) 12 | self.neg_sampler = build_sampler(neg_sampler, **kwargs) 13 | 14 | def _sample_pos(self, **kwargs): 15 | """Sample positive samples.""" 16 | raise NotImplementedError 17 | 18 | def _sample_neg(self, **kwargs): 19 | """Sample negative samples.""" 20 | raise NotImplementedError 21 | -------------------------------------------------------------------------------- /mmdet/core/evaluation/__init__.py: -------------------------------------------------------------------------------- 1 | from .class_names import (cityscapes_classes, coco_classes, dataset_aliases, 2 | get_classes, imagenet_det_classes, 3 | imagenet_vid_classes, voc_classes) 4 | from .eval_hooks import DistEvalHook, EvalHook 5 | from .mean_ap import average_precision, eval_map, print_map_summary 6 | from .recall import (eval_recalls, plot_iou_recall, plot_num_recall, 7 | print_recall_summary) 8 | 9 | __all__ = [ 10 | 'voc_classes', 'imagenet_det_classes', 'imagenet_vid_classes', 11 | 'coco_classes', 'cityscapes_classes', 'dataset_aliases', 'get_classes', 12 | 'DistEvalHook', 'EvalHook', 'average_precision', 'eval_map', 13 | 'print_map_summary', 'eval_recalls', 'print_recall_summary', 14 | 'plot_num_recall', 'plot_iou_recall' 15 | ] 16 | -------------------------------------------------------------------------------- /mmdet/core/fp16/__init__.py: -------------------------------------------------------------------------------- 1 | from .decorators import auto_fp16, force_fp32 2 | from .hooks import Fp16OptimizerHook, wrap_fp16_model 3 | 4 | __all__ = ['auto_fp16', 'force_fp32', 'Fp16OptimizerHook', 'wrap_fp16_model'] 5 | -------------------------------------------------------------------------------- /mmdet/core/fp16/utils.py: -------------------------------------------------------------------------------- 1 | from collections import abc 2 | 3 | import numpy as np 4 | import torch 5 | 6 | 7 | def cast_tensor_type(inputs, src_type, dst_type): 8 | """Recursively convert Tensor in inputs from src_type to dst_type. 9 | 10 | Args: 11 | inputs: Inputs that to be casted. 12 | src_type (torch.dtype): Source type.. 13 | dst_type (torch.dtype): Destination type. 14 | 15 | Returns: 16 | The same type with inputs, but all contained Tensors have been cast. 17 | """ 18 | if isinstance(inputs, torch.Tensor): 19 | return inputs.to(dst_type) 20 | elif isinstance(inputs, str): 21 | return inputs 22 | elif isinstance(inputs, np.ndarray): 23 | return inputs 24 | elif isinstance(inputs, abc.Mapping): 25 | return type(inputs)({ 26 | k: cast_tensor_type(v, src_type, dst_type) 27 | for k, v in inputs.items() 28 | }) 29 | elif isinstance(inputs, abc.Iterable): 30 | return type(inputs)( 31 | cast_tensor_type(item, src_type, dst_type) for item in inputs) 32 | else: 33 | return inputs 34 | -------------------------------------------------------------------------------- /mmdet/core/mask/__init__.py: -------------------------------------------------------------------------------- 1 | from .mask_target import mask_target 2 | from .structures import BaseInstanceMasks, BitmapMasks, PolygonMasks 3 | from .utils import encode_mask_results, split_combined_polys 4 | 5 | __all__ = [ 6 | 'split_combined_polys', 'mask_target', 'BaseInstanceMasks', 'BitmapMasks', 7 | 'PolygonMasks', 'encode_mask_results' 8 | ] 9 | -------------------------------------------------------------------------------- /mmdet/core/post_processing/__init__.py: -------------------------------------------------------------------------------- 1 | from .bbox_nms import multiclass_nms 2 | from .merge_augs import (merge_aug_bboxes, merge_aug_masks, 3 | merge_aug_proposals, merge_aug_scores) 4 | 5 | __all__ = [ 6 | 'multiclass_nms', 'merge_aug_proposals', 'merge_aug_bboxes', 7 | 'merge_aug_scores', 'merge_aug_masks' 8 | ] 9 | -------------------------------------------------------------------------------- /mmdet/core/utils/__init__.py: -------------------------------------------------------------------------------- 1 | from .dist_utils import DistOptimizerHook, allreduce_grads 2 | from .misc import multi_apply, tensor2imgs, unmap 3 | 4 | __all__ = [ 5 | 'allreduce_grads', 'DistOptimizerHook', 'tensor2imgs', 'multi_apply', 6 | 'unmap' 7 | ] 8 | -------------------------------------------------------------------------------- /mmdet/datasets/__init__.py: -------------------------------------------------------------------------------- 1 | from .builder import DATASETS, PIPELINES, build_dataloader, build_dataset 2 | from .cityscapes import CityscapesDataset 3 | from .coco import CocoDataset 4 | from .custom import CustomDataset 5 | from .dataset_wrappers import (ClassBalancedDataset, ConcatDataset, 6 | RepeatDataset) 7 | from .deepfashion import DeepFashionDataset 8 | from .lvis import LVISDataset 9 | from .samplers import DistributedGroupSampler, DistributedSampler, GroupSampler 10 | from .voc import VOCDataset 11 | from .wider_face import WIDERFaceDataset 12 | from .xml_style import XMLDataset 13 | 14 | __all__ = [ 15 | 'CustomDataset', 'XMLDataset', 'CocoDataset', 'DeepFashionDataset', 16 | 'VOCDataset', 'CityscapesDataset', 'LVISDataset', 'GroupSampler', 17 | 'DistributedGroupSampler', 'DistributedSampler', 'build_dataloader', 18 | 'ConcatDataset', 'RepeatDataset', 'ClassBalancedDataset', 19 | 'WIDERFaceDataset', 'DATASETS', 'PIPELINES', 'build_dataset' 20 | ] 21 | -------------------------------------------------------------------------------- /mmdet/datasets/deepfashion.py: -------------------------------------------------------------------------------- 1 | from .builder import DATASETS 2 | from .coco import CocoDataset 3 | 4 | 5 | @DATASETS.register_module() 6 | class DeepFashionDataset(CocoDataset): 7 | 8 | CLASSES = ('top', 'skirt', 'leggings', 'dress', 'outer', 'pants', 'bag', 9 | 'neckwear', 'headwear', 'eyeglass', 'belt', 'footwear', 'hair', 10 | 'skin', 'face') 11 | -------------------------------------------------------------------------------- /mmdet/datasets/pipelines/__init__.py: -------------------------------------------------------------------------------- 1 | from .auto_augment import AutoAugment 2 | from .compose import Compose 3 | from .formating import (Collect, ImageToTensor, ToDataContainer, ToTensor, 4 | Transpose, to_tensor) 5 | from .instaboost import InstaBoost 6 | from .loading import (LoadAnnotations, LoadImageFromFile, 7 | LoadMultiChannelImageFromFiles, LoadProposals) 8 | from .test_time_aug import MultiScaleFlipAug 9 | from .transforms import (Albu, Expand, MinIoURandomCrop, Normalize, Pad, 10 | PhotoMetricDistortion, RandomCenterCropPad, 11 | RandomCrop, RandomFlip, Resize, SegRescale) 12 | 13 | __all__ = [ 14 | 'Compose', 'to_tensor', 'ToTensor', 'ImageToTensor', 'ToDataContainer', 15 | 'Transpose', 'Collect', 'LoadAnnotations', 'LoadImageFromFile', 16 | 'LoadMultiChannelImageFromFiles', 'LoadProposals', 'MultiScaleFlipAug', 17 | 'Resize', 'RandomFlip', 'Pad', 'RandomCrop', 'Normalize', 'SegRescale', 18 | 'MinIoURandomCrop', 'Expand', 'PhotoMetricDistortion', 'Albu', 19 | 'InstaBoost', 'RandomCenterCropPad', 'AutoAugment' 20 | ] 21 | -------------------------------------------------------------------------------- /mmdet/datasets/samplers/__init__.py: -------------------------------------------------------------------------------- 1 | from .distributed_sampler import DistributedSampler 2 | from .group_sampler import DistributedGroupSampler, GroupSampler 3 | 4 | __all__ = ['DistributedSampler', 'DistributedGroupSampler', 'GroupSampler'] 5 | -------------------------------------------------------------------------------- /mmdet/datasets/samplers/distributed_sampler.py: -------------------------------------------------------------------------------- 1 | import torch 2 | from torch.utils.data import DistributedSampler as _DistributedSampler 3 | 4 | 5 | class DistributedSampler(_DistributedSampler): 6 | 7 | def __init__(self, dataset, num_replicas=None, rank=None, shuffle=True): 8 | super().__init__(dataset, num_replicas=num_replicas, rank=rank) 9 | self.shuffle = shuffle 10 | 11 | def __iter__(self): 12 | # deterministically shuffle based on epoch 13 | if self.shuffle: 14 | g = torch.Generator() 15 | g.manual_seed(self.epoch) 16 | indices = torch.randperm(len(self.dataset), generator=g).tolist() 17 | else: 18 | indices = torch.arange(len(self.dataset)).tolist() 19 | 20 | # add extra samples to make it evenly divisible 21 | indices += indices[:(self.total_size - len(indices))] 22 | assert len(indices) == self.total_size 23 | 24 | # subsample 25 | indices = indices[self.rank:self.total_size:self.num_replicas] 26 | assert len(indices) == self.num_samples 27 | 28 | return iter(indices) 29 | -------------------------------------------------------------------------------- /mmdet/models/__init__.py: -------------------------------------------------------------------------------- 1 | from .backbones import * # noqa: F401,F403 2 | from .builder import (BACKBONES, DETECTORS, HEADS, LOSSES, NECKS, 3 | ROI_EXTRACTORS, SHARED_HEADS, build_backbone, 4 | build_detector, build_head, build_loss, build_neck, 5 | build_roi_extractor, build_shared_head) 6 | from .dense_heads import * # noqa: F401,F403 7 | from .detectors import * # noqa: F401,F403 8 | from .losses import * # noqa: F401,F403 9 | from .necks import * # noqa: F401,F403 10 | from .roi_heads import * # noqa: F401,F403 11 | 12 | __all__ = [ 13 | 'BACKBONES', 'NECKS', 'ROI_EXTRACTORS', 'SHARED_HEADS', 'HEADS', 'LOSSES', 14 | 'DETECTORS', 'build_backbone', 'build_neck', 'build_roi_extractor', 15 | 'build_shared_head', 'build_head', 'build_loss', 'build_detector' 16 | ] 17 | -------------------------------------------------------------------------------- /mmdet/models/backbones/__init__.py: -------------------------------------------------------------------------------- 1 | from .detectors_resnet import DetectoRS_ResNet 2 | from .detectors_resnext import DetectoRS_ResNeXt 3 | from .hourglass import HourglassNet 4 | from .hrnet import HRNet 5 | from .regnet import RegNet 6 | from .res2net import Res2Net 7 | from .resnet import ResNet, ResNetV1d 8 | from .resnext import ResNeXt 9 | from .ssd_vgg import SSDVGG 10 | 11 | __all__ = [ 12 | 'RegNet', 'ResNet', 'ResNetV1d', 'ResNeXt', 'SSDVGG', 'HRNet', 'Res2Net', 13 | 'HourglassNet', 'DetectoRS_ResNet', 'DetectoRS_ResNeXt' 14 | ] 15 | -------------------------------------------------------------------------------- /mmdet/models/dense_heads/__init__.py: -------------------------------------------------------------------------------- 1 | from .anchor_free_head import AnchorFreeHead 2 | from .anchor_head import AnchorHead 3 | from .atss_head import ATSSHead 4 | from .fcos_head import FCOSHead 5 | from .fovea_head import FoveaHead 6 | from .free_anchor_retina_head import FreeAnchorRetinaHead 7 | from .fsaf_head import FSAFHead 8 | from .ga_retina_head import GARetinaHead 9 | from .ga_rpn_head import GARPNHead 10 | from .gfl_head import GFLHead 11 | from .guided_anchor_head import FeatureAdaption, GuidedAnchorHead 12 | from .nasfcos_head import NASFCOSHead 13 | from .pisa_retinanet_head import PISARetinaHead 14 | from .pisa_ssd_head import PISASSDHead 15 | from .reppoints_head import RepPointsHead 16 | from .retina_head import RetinaHead 17 | from .retina_sepbn_head import RetinaSepBNHead 18 | from .rpn_head import RPNHead 19 | from .ssd_head import SSDHead 20 | 21 | __all__ = [ 22 | 'AnchorFreeHead', 'AnchorHead', 'GuidedAnchorHead', 'FeatureAdaption', 23 | 'RPNHead', 'GARPNHead', 'RetinaHead', 'RetinaSepBNHead', 'GARetinaHead', 24 | 'SSDHead', 'FCOSHead', 'RepPointsHead', 'FoveaHead', 25 | 'FreeAnchorRetinaHead', 'ATSSHead', 'FSAFHead', 'NASFCOSHead', 26 | 'PISARetinaHead', 'PISASSDHead', 'GFLHead' 27 | ] 28 | -------------------------------------------------------------------------------- /mmdet/models/detectors/__init__.py: -------------------------------------------------------------------------------- 1 | from .atss import ATSS 2 | from .base import BaseDetector 3 | from .cascade_rcnn import CascadeRCNN 4 | from .fast_rcnn import FastRCNN 5 | from .faster_rcnn import FasterRCNN 6 | from .fcos import FCOS 7 | from .fovea import FOVEA 8 | from .fsaf import FSAF 9 | from .gfl import GFL 10 | from .grid_rcnn import GridRCNN 11 | from .htc import HybridTaskCascade 12 | from .mask_rcnn import MaskRCNN 13 | from .mask_scoring_rcnn import MaskScoringRCNN 14 | from .nasfcos import NASFCOS 15 | from .point_rend import PointRend 16 | from .reppoints_detector import RepPointsDetector 17 | from .retinanet import RetinaNet 18 | from .rpn import RPN 19 | from .single_stage import SingleStageDetector 20 | from .two_stage import TwoStageDetector 21 | 22 | __all__ = [ 23 | 'ATSS', 'BaseDetector', 'SingleStageDetector', 'TwoStageDetector', 'RPN', 24 | 'FastRCNN', 'FasterRCNN', 'MaskRCNN', 'CascadeRCNN', 'HybridTaskCascade', 25 | 'RetinaNet', 'FCOS', 'GridRCNN', 'MaskScoringRCNN', 'RepPointsDetector', 26 | 'FOVEA', 'FSAF', 'NASFCOS', 'PointRend', 'GFL' 27 | ] 28 | -------------------------------------------------------------------------------- /mmdet/models/detectors/atss.py: -------------------------------------------------------------------------------- 1 | from ..builder import DETECTORS 2 | from .single_stage import SingleStageDetector 3 | 4 | 5 | @DETECTORS.register_module() 6 | class ATSS(SingleStageDetector): 7 | 8 | def __init__(self, 9 | backbone, 10 | neck, 11 | bbox_head, 12 | train_cfg=None, 13 | test_cfg=None, 14 | pretrained=None): 15 | super(ATSS, self).__init__(backbone, neck, bbox_head, train_cfg, 16 | test_cfg, pretrained) 17 | -------------------------------------------------------------------------------- /mmdet/models/detectors/faster_rcnn.py: -------------------------------------------------------------------------------- 1 | from ..builder import DETECTORS 2 | from .two_stage import TwoStageDetector 3 | 4 | 5 | @DETECTORS.register_module() 6 | class FasterRCNN(TwoStageDetector): 7 | """Implementation of `Faster R-CNN `_""" 8 | 9 | def __init__(self, 10 | backbone, 11 | rpn_head, 12 | roi_head, 13 | train_cfg, 14 | test_cfg, 15 | neck=None, 16 | pretrained=None): 17 | super(FasterRCNN, self).__init__( 18 | backbone=backbone, 19 | neck=neck, 20 | rpn_head=rpn_head, 21 | roi_head=roi_head, 22 | train_cfg=train_cfg, 23 | test_cfg=test_cfg, 24 | pretrained=pretrained) 25 | -------------------------------------------------------------------------------- /mmdet/models/detectors/fcos.py: -------------------------------------------------------------------------------- 1 | from ..builder import DETECTORS 2 | from .single_stage import SingleStageDetector 3 | 4 | 5 | @DETECTORS.register_module() 6 | class FCOS(SingleStageDetector): 7 | """Implementation of `FCOS `_""" 8 | 9 | def __init__(self, 10 | backbone, 11 | neck, 12 | bbox_head, 13 | train_cfg=None, 14 | test_cfg=None, 15 | pretrained=None): 16 | super(FCOS, self).__init__(backbone, neck, bbox_head, train_cfg, 17 | test_cfg, pretrained) 18 | -------------------------------------------------------------------------------- /mmdet/models/detectors/fovea.py: -------------------------------------------------------------------------------- 1 | from ..builder import DETECTORS 2 | from .single_stage import SingleStageDetector 3 | 4 | 5 | @DETECTORS.register_module() 6 | class FOVEA(SingleStageDetector): 7 | """Implementation of `FoveaBox `_""" 8 | 9 | def __init__(self, 10 | backbone, 11 | neck, 12 | bbox_head, 13 | train_cfg=None, 14 | test_cfg=None, 15 | pretrained=None): 16 | super(FOVEA, self).__init__(backbone, neck, bbox_head, train_cfg, 17 | test_cfg, pretrained) 18 | -------------------------------------------------------------------------------- /mmdet/models/detectors/fsaf.py: -------------------------------------------------------------------------------- 1 | from ..builder import DETECTORS 2 | from .single_stage import SingleStageDetector 3 | 4 | 5 | @DETECTORS.register_module() 6 | class FSAF(SingleStageDetector): 7 | """Implementation of `FSAF `_""" 8 | 9 | def __init__(self, 10 | backbone, 11 | neck, 12 | bbox_head, 13 | train_cfg=None, 14 | test_cfg=None, 15 | pretrained=None): 16 | super(FSAF, self).__init__(backbone, neck, bbox_head, train_cfg, 17 | test_cfg, pretrained) 18 | -------------------------------------------------------------------------------- /mmdet/models/detectors/gfl.py: -------------------------------------------------------------------------------- 1 | from ..builder import DETECTORS 2 | from .single_stage import SingleStageDetector 3 | 4 | 5 | @DETECTORS.register_module() 6 | class GFL(SingleStageDetector): 7 | 8 | def __init__(self, 9 | backbone, 10 | neck, 11 | bbox_head, 12 | train_cfg=None, 13 | test_cfg=None, 14 | pretrained=None): 15 | super(GFL, self).__init__(backbone, neck, bbox_head, train_cfg, 16 | test_cfg, pretrained) 17 | -------------------------------------------------------------------------------- /mmdet/models/detectors/grid_rcnn.py: -------------------------------------------------------------------------------- 1 | from ..builder import DETECTORS 2 | from .two_stage import TwoStageDetector 3 | 4 | 5 | @DETECTORS.register_module() 6 | class GridRCNN(TwoStageDetector): 7 | """Grid R-CNN. 8 | 9 | This detector is the implementation of: 10 | - Grid R-CNN (https://arxiv.org/abs/1811.12030) 11 | - Grid R-CNN Plus: Faster and Better (https://arxiv.org/abs/1906.05688) 12 | """ 13 | 14 | def __init__(self, 15 | backbone, 16 | rpn_head, 17 | roi_head, 18 | train_cfg, 19 | test_cfg, 20 | neck=None, 21 | pretrained=None): 22 | super(GridRCNN, self).__init__( 23 | backbone=backbone, 24 | neck=neck, 25 | rpn_head=rpn_head, 26 | roi_head=roi_head, 27 | train_cfg=train_cfg, 28 | test_cfg=test_cfg, 29 | pretrained=pretrained) 30 | -------------------------------------------------------------------------------- /mmdet/models/detectors/htc.py: -------------------------------------------------------------------------------- 1 | from ..builder import DETECTORS 2 | from .cascade_rcnn import CascadeRCNN 3 | 4 | 5 | @DETECTORS.register_module() 6 | class HybridTaskCascade(CascadeRCNN): 7 | """Implementation of `HTC `_""" 8 | 9 | def __init__(self, **kwargs): 10 | super(HybridTaskCascade, self).__init__(**kwargs) 11 | 12 | @property 13 | def with_semantic(self): 14 | """bool: whether the detector has a semantic head""" 15 | return self.roi_head.with_semantic 16 | -------------------------------------------------------------------------------- /mmdet/models/detectors/mask_rcnn.py: -------------------------------------------------------------------------------- 1 | from ..builder import DETECTORS 2 | from .two_stage import TwoStageDetector 3 | 4 | 5 | @DETECTORS.register_module() 6 | class MaskRCNN(TwoStageDetector): 7 | """Implementation of `Mask R-CNN `_""" 8 | 9 | def __init__(self, 10 | backbone, 11 | rpn_head, 12 | roi_head, 13 | train_cfg, 14 | test_cfg, 15 | neck=None, 16 | pretrained=None): 17 | super(MaskRCNN, self).__init__( 18 | backbone=backbone, 19 | neck=neck, 20 | rpn_head=rpn_head, 21 | roi_head=roi_head, 22 | train_cfg=train_cfg, 23 | test_cfg=test_cfg, 24 | pretrained=pretrained) 25 | -------------------------------------------------------------------------------- /mmdet/models/detectors/mask_scoring_rcnn.py: -------------------------------------------------------------------------------- 1 | from ..builder import DETECTORS 2 | from .two_stage import TwoStageDetector 3 | 4 | 5 | @DETECTORS.register_module() 6 | class MaskScoringRCNN(TwoStageDetector): 7 | """Mask Scoring RCNN. 8 | 9 | https://arxiv.org/abs/1903.00241 10 | """ 11 | 12 | def __init__(self, 13 | backbone, 14 | rpn_head, 15 | roi_head, 16 | train_cfg, 17 | test_cfg, 18 | neck=None, 19 | pretrained=None): 20 | super(MaskScoringRCNN, self).__init__( 21 | backbone=backbone, 22 | neck=neck, 23 | rpn_head=rpn_head, 24 | roi_head=roi_head, 25 | train_cfg=train_cfg, 26 | test_cfg=test_cfg, 27 | pretrained=pretrained) 28 | -------------------------------------------------------------------------------- /mmdet/models/detectors/nasfcos.py: -------------------------------------------------------------------------------- 1 | from ..builder import DETECTORS 2 | from .single_stage import SingleStageDetector 3 | 4 | 5 | @DETECTORS.register_module() 6 | class NASFCOS(SingleStageDetector): 7 | """NAS-FCOS: Fast Neural Architecture Search for Object Detection. 8 | 9 | https://arxiv.org/abs/1906.0442 10 | """ 11 | 12 | def __init__(self, 13 | backbone, 14 | neck, 15 | bbox_head, 16 | train_cfg=None, 17 | test_cfg=None, 18 | pretrained=None): 19 | super(NASFCOS, self).__init__(backbone, neck, bbox_head, train_cfg, 20 | test_cfg, pretrained) 21 | -------------------------------------------------------------------------------- /mmdet/models/detectors/point_rend.py: -------------------------------------------------------------------------------- 1 | from ..builder import DETECTORS 2 | from .two_stage import TwoStageDetector 3 | 4 | 5 | @DETECTORS.register_module() 6 | class PointRend(TwoStageDetector): 7 | """PointRend: Image Segmentation as Rendering 8 | 9 | This detector is the implementation of 10 | `PointRend `_. 11 | 12 | """ 13 | 14 | def __init__(self, 15 | backbone, 16 | rpn_head, 17 | roi_head, 18 | train_cfg, 19 | test_cfg, 20 | neck=None, 21 | pretrained=None): 22 | super(PointRend, self).__init__( 23 | backbone=backbone, 24 | neck=neck, 25 | rpn_head=rpn_head, 26 | roi_head=roi_head, 27 | train_cfg=train_cfg, 28 | test_cfg=test_cfg, 29 | pretrained=pretrained) 30 | -------------------------------------------------------------------------------- /mmdet/models/detectors/retinanet.py: -------------------------------------------------------------------------------- 1 | from ..builder import DETECTORS 2 | from .single_stage import SingleStageDetector 3 | 4 | 5 | @DETECTORS.register_module() 6 | class RetinaNet(SingleStageDetector): 7 | """Implementation of `RetinaNet `_""" 8 | 9 | def __init__(self, 10 | backbone, 11 | neck, 12 | bbox_head, 13 | train_cfg=None, 14 | test_cfg=None, 15 | pretrained=None): 16 | super(RetinaNet, self).__init__(backbone, neck, bbox_head, train_cfg, 17 | test_cfg, pretrained) 18 | -------------------------------------------------------------------------------- /mmdet/models/necks/__init__.py: -------------------------------------------------------------------------------- 1 | from .bfp import BFP 2 | from .fpn import FPN 3 | from .fpn_carafe import FPN_CARAFE 4 | from .hrfpn import HRFPN 5 | from .nas_fpn import NASFPN 6 | from .nasfcos_fpn import NASFCOS_FPN 7 | from .pafpn import PAFPN 8 | from .rfp import RFP 9 | 10 | __all__ = [ 11 | 'FPN', 'BFP', 'HRFPN', 'NASFPN', 'FPN_CARAFE', 'PAFPN', 'NASFCOS_FPN', 12 | 'RFP' 13 | ] 14 | -------------------------------------------------------------------------------- /mmdet/models/roi_heads/__init__.py: -------------------------------------------------------------------------------- 1 | from .base_roi_head import BaseRoIHead 2 | from .bbox_heads import (BBoxHead, ConvFCBBoxHead, DoubleConvFCBBoxHead, 3 | Shared2FCBBoxHead, Shared4Conv1FCBBoxHead) 4 | from .cascade_roi_head import CascadeRoIHead 5 | from .double_roi_head import DoubleHeadRoIHead 6 | from .dynamic_roi_head import DynamicRoIHead 7 | from .grid_roi_head import GridRoIHead 8 | from .htc_roi_head import HybridTaskCascadeRoIHead 9 | from .mask_heads import (CoarseMaskHead, FCNMaskHead, FusedSemanticHead, 10 | GridHead, HTCMaskHead, MaskIoUHead, MaskPointHead) 11 | from .mask_scoring_roi_head import MaskScoringRoIHead 12 | from .pisa_roi_head import PISARoIHead 13 | from .point_rend_roi_head import PointRendRoIHead 14 | from .roi_extractors import SingleRoIExtractor 15 | from .shared_heads import ResLayer 16 | 17 | __all__ = [ 18 | 'BaseRoIHead', 'CascadeRoIHead', 'DoubleHeadRoIHead', 'MaskScoringRoIHead', 19 | 'HybridTaskCascadeRoIHead', 'GridRoIHead', 'ResLayer', 'BBoxHead', 20 | 'ConvFCBBoxHead', 'Shared2FCBBoxHead', 'Shared4Conv1FCBBoxHead', 21 | 'DoubleConvFCBBoxHead', 'FCNMaskHead', 'HTCMaskHead', 'FusedSemanticHead', 22 | 'GridHead', 'MaskIoUHead', 'SingleRoIExtractor', 'PISARoIHead', 23 | 'PointRendRoIHead', 'MaskPointHead', 'CoarseMaskHead', 'DynamicRoIHead' 24 | ] 25 | -------------------------------------------------------------------------------- /mmdet/models/roi_heads/bbox_heads/__init__.py: -------------------------------------------------------------------------------- 1 | from .bbox_head import BBoxHead 2 | from .convfc_bbox_head import (ConvFCBBoxHead, Shared2FCBBoxHead, 3 | Shared4Conv1FCBBoxHead) 4 | from .double_bbox_head import DoubleConvFCBBoxHead 5 | 6 | __all__ = [ 7 | 'BBoxHead', 'ConvFCBBoxHead', 'Shared2FCBBoxHead', 8 | 'Shared4Conv1FCBBoxHead', 'DoubleConvFCBBoxHead' 9 | ] 10 | -------------------------------------------------------------------------------- /mmdet/models/roi_heads/double_roi_head.py: -------------------------------------------------------------------------------- 1 | from ..builder import HEADS 2 | from .standard_roi_head import StandardRoIHead 3 | 4 | 5 | @HEADS.register_module() 6 | class DoubleHeadRoIHead(StandardRoIHead): 7 | """RoI head for Double Head RCNN. 8 | 9 | https://arxiv.org/abs/1904.06493 10 | """ 11 | 12 | def __init__(self, reg_roi_scale_factor, **kwargs): 13 | super(DoubleHeadRoIHead, self).__init__(**kwargs) 14 | self.reg_roi_scale_factor = reg_roi_scale_factor 15 | 16 | def _bbox_forward(self, x, rois): 17 | """Box head forward function used in both training and testing time.""" 18 | bbox_cls_feats = self.bbox_roi_extractor( 19 | x[:self.bbox_roi_extractor.num_inputs], rois) 20 | bbox_reg_feats = self.bbox_roi_extractor( 21 | x[:self.bbox_roi_extractor.num_inputs], 22 | rois, 23 | roi_scale_factor=self.reg_roi_scale_factor) 24 | if self.with_shared_head: 25 | bbox_cls_feats = self.shared_head(bbox_cls_feats) 26 | bbox_reg_feats = self.shared_head(bbox_reg_feats) 27 | cls_score, bbox_pred = self.bbox_head(bbox_cls_feats, bbox_reg_feats) 28 | 29 | bbox_results = dict( 30 | cls_score=cls_score, 31 | bbox_pred=bbox_pred, 32 | bbox_feats=bbox_cls_feats) 33 | return bbox_results 34 | -------------------------------------------------------------------------------- /mmdet/models/roi_heads/mask_heads/__init__.py: -------------------------------------------------------------------------------- 1 | from .coarse_mask_head import CoarseMaskHead 2 | from .fcn_mask_head import FCNMaskHead 3 | from .fused_semantic_head import FusedSemanticHead 4 | from .grid_head import GridHead 5 | from .htc_mask_head import HTCMaskHead 6 | from .mask_point_head import MaskPointHead 7 | from .maskiou_head import MaskIoUHead 8 | 9 | __all__ = [ 10 | 'FCNMaskHead', 'HTCMaskHead', 'FusedSemanticHead', 'GridHead', 11 | 'MaskIoUHead', 'CoarseMaskHead', 'MaskPointHead' 12 | ] 13 | -------------------------------------------------------------------------------- /mmdet/models/roi_heads/roi_extractors/__init__.py: -------------------------------------------------------------------------------- 1 | from .generic_roi_extractor import GenericRoIExtractor 2 | from .single_level_roi_extractor import SingleRoIExtractor 3 | 4 | __all__ = [ 5 | 'SingleRoIExtractor', 6 | 'GenericRoIExtractor', 7 | ] 8 | -------------------------------------------------------------------------------- /mmdet/models/roi_heads/shared_heads/__init__.py: -------------------------------------------------------------------------------- 1 | from .res_layer import ResLayer 2 | 3 | __all__ = ['ResLayer'] 4 | -------------------------------------------------------------------------------- /mmdet/models/utils/__init__.py: -------------------------------------------------------------------------------- 1 | from .res_layer import ResLayer 2 | 3 | __all__ = ['ResLayer'] 4 | -------------------------------------------------------------------------------- /mmdet/ops/carafe/__init__.py: -------------------------------------------------------------------------------- 1 | from .carafe import CARAFE, CARAFENaive, CARAFEPack, carafe, carafe_naive 2 | 3 | __all__ = ['carafe', 'carafe_naive', 'CARAFE', 'CARAFENaive', 'CARAFEPack'] 4 | -------------------------------------------------------------------------------- /mmdet/ops/carafe/setup.py: -------------------------------------------------------------------------------- 1 | from setuptools import setup 2 | 3 | from torch.utils.cpp_extension import BuildExtension, CUDAExtension 4 | 5 | NVCC_ARGS = [ 6 | '-D__CUDA_NO_HALF_OPERATORS__', 7 | '-D__CUDA_NO_HALF_CONVERSIONS__', 8 | '-D__CUDA_NO_HALF2_OPERATORS__', 9 | ] 10 | 11 | setup( 12 | name='carafe', 13 | ext_modules=[ 14 | CUDAExtension( 15 | 'carafe_ext', [ 16 | 'src/cuda/carafe_cuda.cpp', 'src/cuda/carafe_cuda_kernel.cu', 17 | 'src/carafe_ext.cpp' 18 | ], 19 | define_macros=[('WITH_CUDA', None)], 20 | extra_compile_args={ 21 | 'cxx': [], 22 | 'nvcc': NVCC_ARGS 23 | }), 24 | CUDAExtension( 25 | 'carafe_naive_ext', [ 26 | 'src/cuda/carafe_naive_cuda.cpp', 27 | 'src/cuda/carafe_naive_cuda_kernel.cu', 28 | 'src/carafe_naive_ext.cpp' 29 | ], 30 | define_macros=[('WITH_CUDA', None)], 31 | extra_compile_args={ 32 | 'cxx': [], 33 | 'nvcc': NVCC_ARGS 34 | }) 35 | ], 36 | cmdclass={'build_ext': BuildExtension}) 37 | -------------------------------------------------------------------------------- /mmdet/ops/corner_pool/__init__.py: -------------------------------------------------------------------------------- 1 | from .corner_pool import CornerPool 2 | 3 | __all__ = ['CornerPool'] 4 | -------------------------------------------------------------------------------- /mmdet/ops/dcn/__init__.py: -------------------------------------------------------------------------------- 1 | from .deform_conv import (DeformConv, DeformConvPack, ModulatedDeformConv, 2 | ModulatedDeformConvPack, deform_conv, 3 | modulated_deform_conv) 4 | from .deform_pool import (DeformRoIPooling, DeformRoIPoolingPack, 5 | ModulatedDeformRoIPoolingPack, deform_roi_pooling) 6 | 7 | __all__ = [ 8 | 'DeformConv', 'DeformConvPack', 'ModulatedDeformConv', 9 | 'ModulatedDeformConvPack', 'DeformRoIPooling', 'DeformRoIPoolingPack', 10 | 'ModulatedDeformRoIPoolingPack', 'deform_conv', 'modulated_deform_conv', 11 | 'deform_roi_pooling' 12 | ] 13 | -------------------------------------------------------------------------------- /mmdet/ops/masked_conv/__init__.py: -------------------------------------------------------------------------------- 1 | from .masked_conv import MaskedConv2d, masked_conv2d 2 | 3 | __all__ = ['masked_conv2d', 'MaskedConv2d'] 4 | -------------------------------------------------------------------------------- /mmdet/ops/nms/__init__.py: -------------------------------------------------------------------------------- 1 | from .nms_wrapper import batched_nms, nms, nms_match, soft_nms 2 | 3 | __all__ = ['nms', 'soft_nms', 'batched_nms', 'nms_match'] 4 | -------------------------------------------------------------------------------- /mmdet/ops/nms/src/cuda/nms_cuda.cpp: -------------------------------------------------------------------------------- 1 | // Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. 2 | #include 3 | 4 | #define CHECK_CUDA(x) TORCH_CHECK(x.device().is_cuda(), #x, " must be a CUDAtensor ") 5 | 6 | at::Tensor nms_cuda_forward(const at::Tensor boxes, float nms_overlap_thresh); 7 | 8 | at::Tensor nms_cuda(const at::Tensor& dets, const float threshold) { 9 | CHECK_CUDA(dets); 10 | if (dets.numel() == 0) 11 | return at::empty({0}, dets.options().dtype(at::kLong).device(at::kCPU)); 12 | return nms_cuda_forward(dets, threshold); 13 | } 14 | -------------------------------------------------------------------------------- /mmdet/ops/roi_align/__init__.py: -------------------------------------------------------------------------------- 1 | from .roi_align import RoIAlign, roi_align 2 | 3 | __all__ = ['roi_align', 'RoIAlign'] 4 | -------------------------------------------------------------------------------- /mmdet/ops/roi_align/gradcheck.py: -------------------------------------------------------------------------------- 1 | import os.path as osp 2 | import sys 3 | 4 | import numpy as np 5 | import torch 6 | from torch.autograd import gradcheck 7 | 8 | sys.path.append(osp.abspath(osp.join(__file__, '../../'))) 9 | from roi_align import RoIAlign # noqa: E402, isort:skip 10 | 11 | feat_size = 15 12 | spatial_scale = 1.0 / 8 13 | img_size = feat_size / spatial_scale 14 | num_imgs = 2 15 | num_rois = 20 16 | 17 | batch_ind = np.random.randint(num_imgs, size=(num_rois, 1)) 18 | rois = np.random.rand(num_rois, 4) * img_size * 0.5 19 | rois[:, 2:] += img_size * 0.5 20 | rois = np.hstack((batch_ind, rois)) 21 | 22 | feat = torch.randn( 23 | num_imgs, 16, feat_size, feat_size, requires_grad=True, device='cuda:0') 24 | rois = torch.from_numpy(rois).float().cuda() 25 | inputs = (feat, rois) 26 | print('Gradcheck for roi align...') 27 | test = gradcheck(RoIAlign(3, spatial_scale), inputs, atol=1e-3, eps=1e-3) 28 | print(test) 29 | test = gradcheck(RoIAlign(3, spatial_scale, 2), inputs, atol=1e-3, eps=1e-3) 30 | print(test) 31 | -------------------------------------------------------------------------------- /mmdet/ops/roi_pool/__init__.py: -------------------------------------------------------------------------------- 1 | from .roi_pool import RoIPool, roi_pool 2 | 3 | __all__ = ['roi_pool', 'RoIPool'] 4 | -------------------------------------------------------------------------------- /mmdet/ops/roi_pool/gradcheck.py: -------------------------------------------------------------------------------- 1 | import os.path as osp 2 | import sys 3 | 4 | import torch 5 | from torch.autograd import gradcheck 6 | 7 | sys.path.append(osp.abspath(osp.join(__file__, '../../'))) 8 | from roi_pool import RoIPool # noqa: E402, isort:skip 9 | 10 | feat = torch.randn(4, 16, 15, 15, requires_grad=True).cuda() 11 | rois = torch.Tensor([[0, 0, 0, 50, 50], [0, 10, 30, 43, 55], 12 | [1, 67, 40, 110, 120]]).cuda() 13 | inputs = (feat, rois) 14 | print('Gradcheck for roi pooling...') 15 | test = gradcheck(RoIPool(4, 1.0 / 8), inputs, eps=1e-5, atol=1e-3) 16 | print(test) 17 | -------------------------------------------------------------------------------- /mmdet/ops/sigmoid_focal_loss/__init__.py: -------------------------------------------------------------------------------- 1 | from .sigmoid_focal_loss import SigmoidFocalLoss, sigmoid_focal_loss 2 | 3 | __all__ = ['SigmoidFocalLoss', 'sigmoid_focal_loss'] 4 | -------------------------------------------------------------------------------- /mmdet/ops/utils/__init__.py: -------------------------------------------------------------------------------- 1 | # from . import compiling_info 2 | from .compiling_info import get_compiler_version, get_compiling_cuda_version 3 | 4 | # get_compiler_version = compiling_info.get_compiler_version 5 | # get_compiling_cuda_version = compiling_info.get_compiling_cuda_version 6 | 7 | __all__ = ['get_compiler_version', 'get_compiling_cuda_version'] 8 | -------------------------------------------------------------------------------- /mmdet/utils/__init__.py: -------------------------------------------------------------------------------- 1 | from .collect_env import collect_env 2 | from .logger import get_root_logger 3 | 4 | __all__ = ['get_root_logger', 'collect_env'] 5 | -------------------------------------------------------------------------------- /mmdet/utils/logger.py: -------------------------------------------------------------------------------- 1 | import logging 2 | 3 | from mmcv.utils import get_logger 4 | 5 | 6 | def get_root_logger(log_file=None, log_level=logging.INFO): 7 | """Get root logger. 8 | 9 | Args: 10 | log_file (str, optional): File path of log. Defaults to None. 11 | log_level (int, optional): The level of logger. 12 | Defaults to logging.INFO. 13 | 14 | Returns: 15 | :obj:`logging.Logger`: The obtained logger 16 | """ 17 | logger = get_logger(name='mmdet', log_file=log_file, log_level=log_level) 18 | 19 | return logger 20 | -------------------------------------------------------------------------------- /noisy_labels_SN_COCO.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | import json 3 | 4 | 5 | def unjson(file): 6 | with open(file, 'r') as fo: 7 | dict = json.load(fo) 8 | return dict 9 | 10 | # r is noise rate 11 | r = 0.2 12 | 13 | count = 0 14 | 15 | p_a = '' 16 | p_g = '' 17 | a = unjson(p_a) 18 | for i in range(len(a['annotations'])): 19 | if np.random.random() < r: 20 | id = np.array([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 21 | 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 22 | 24, 25, 27, 28, 31, 32, 33, 34, 35, 36, 37, 23 | 38, 39, 40, 41, 42, 43, 44, 46, 47, 48, 49, 24 | 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 25 | 61, 62, 63, 64, 65, 67, 70, 72, 73, 74, 75, 26 | 76, 77, 78, 79, 80, 81, 82, 84, 85, 86, 87, 27 | 88, 89, 90]) 28 | index = np.random.randint(0, 79) 29 | a['annotations'][i]['category_id'] = int(id[index]) 30 | count += 1 31 | with open(p_g, 'w') as file: 32 | json.dump(a, file) 33 | 34 | print(count) 35 | -------------------------------------------------------------------------------- /noisy_labels_SN_VOC.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | import json 3 | 4 | 5 | def unjson(file): 6 | with open(file, 'r') as fo: 7 | dict = json.load(fo) 8 | return dict 9 | 10 | 11 | # r is noise rate 12 | r = 0.2 13 | 14 | count = 0 15 | 16 | p_a = '' 17 | p_g = '' 18 | a = unjson(p_a) 19 | for i in range(len(a['annotations'])): 20 | if np.random.random() < r: 21 | a['annotations'][i]['category_id'] = np.random.randint(1, 20) 22 | count += 1 23 | with open(p_g, 'w') as file: 24 | json.dump(a, file) 25 | 26 | print(count) 27 | -------------------------------------------------------------------------------- /noisy_labels_SN_cityscapes.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | import json 3 | 4 | 5 | def unjson(file): 6 | with open(file, 'r') as fo: 7 | dict = json.load(fo) 8 | return dict 9 | 10 | 11 | # r is noise rate 12 | r = 0.2 13 | 14 | count = 0 15 | 16 | p_a = '' 17 | p_g = '' 18 | a = unjson(p_a) 19 | for i in range(len(a['annotations'])): 20 | if np.random.random() < r: 21 | id = np.array([24, 25, 26, 27, 28, 31, 32, 33]) 22 | index = np.random.randint(0, 7) 23 | a['annotations'][i]['category_id'] = int(id[index]) 24 | count += 1 25 | with open(p_g, 'w') as file: 26 | json.dump(a, file) 27 | 28 | print(count) 29 | -------------------------------------------------------------------------------- /pytest.ini: -------------------------------------------------------------------------------- 1 | [pytest] 2 | addopts = --xdoctest --xdoctest-style=auto 3 | norecursedirs = .git ignore build __pycache__ data docker docs .eggs 4 | 5 | filterwarnings= default 6 | ignore:.*No cfgstr given in Cacher constructor or call.*:Warning 7 | ignore:.*Define the __nice__ method for.*:Warning 8 | -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- 1 | -r requirements/build.txt 2 | -r requirements/optional.txt 3 | -r requirements/runtime.txt 4 | -r requirements/tests.txt 5 | -------------------------------------------------------------------------------- /requirements/build.txt: -------------------------------------------------------------------------------- 1 | # These must be installed before building mmdetection 2 | cython 3 | numpy 4 | torch>=1.3 5 | -------------------------------------------------------------------------------- /requirements/docs.txt: -------------------------------------------------------------------------------- 1 | recommonmark 2 | sphinx 3 | sphinx_markdown_tables 4 | sphinx_rtd_theme 5 | -------------------------------------------------------------------------------- /requirements/optional.txt: -------------------------------------------------------------------------------- 1 | albumentations>=0.3.2 2 | cityscapesscripts 3 | imagecorruptions 4 | -------------------------------------------------------------------------------- /requirements/readthedocs.txt: -------------------------------------------------------------------------------- 1 | mmcv 2 | torch 3 | torchvision 4 | -------------------------------------------------------------------------------- /requirements/runtime.txt: -------------------------------------------------------------------------------- 1 | matplotlib 2 | mmcv>=0.6.2 3 | numpy 4 | # need older pillow until torchvision is fixed 5 | Pillow<=6.2.2 6 | six 7 | terminaltables 8 | torch>=1.3 9 | torchvision 10 | -------------------------------------------------------------------------------- /requirements/tests.txt: -------------------------------------------------------------------------------- 1 | asynctest 2 | codecov 3 | flake8 4 | interrogate 5 | isort==4.3.21 6 | # Note: used for kwarray.group_items, this may be ported to mmcv in the future. 7 | kwarray 8 | pytest 9 | ubelt 10 | xdoctest>=0.10.0 11 | yapf 12 | -------------------------------------------------------------------------------- /tests/data/color.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/longrongyang/LNCIS/6b0ad08b79e0b372ae90cba7a31db00d23f43b3d/tests/data/color.jpg -------------------------------------------------------------------------------- /tests/data/gray.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/longrongyang/LNCIS/6b0ad08b79e0b372ae90cba7a31db00d23f43b3d/tests/data/gray.jpg -------------------------------------------------------------------------------- /tests/test_data/test_formatting.py: -------------------------------------------------------------------------------- 1 | import os.path as osp 2 | 3 | from mmcv.utils import build_from_cfg 4 | 5 | from mmdet.datasets.builder import PIPELINES 6 | 7 | 8 | def test_default_format_bundle(): 9 | results = dict( 10 | img_prefix=osp.join(osp.dirname(__file__), '../data'), 11 | img_info=dict(filename='color.jpg')) 12 | load = dict(type='LoadImageFromFile') 13 | load = build_from_cfg(load, PIPELINES) 14 | bundle = dict(type='DefaultFormatBundle') 15 | bundle = build_from_cfg(bundle, PIPELINES) 16 | results = load(results) 17 | assert 'pad_shape' not in results 18 | assert 'scale_factor' not in results 19 | assert 'img_norm_cfg' not in results 20 | results = bundle(results) 21 | assert 'pad_shape' in results 22 | assert 'scale_factor' in results 23 | assert 'img_norm_cfg' in results 24 | -------------------------------------------------------------------------------- /tools/dist_test.sh: -------------------------------------------------------------------------------- 1 | #!/usr/bin/env bash 2 | 3 | CONFIG=$1 4 | CHECKPOINT=$2 5 | GPUS=$3 6 | PORT=${PORT:-29500} 7 | 8 | PYTHONPATH="$(dirname $0)/..":$PYTHONPATH \ 9 | python -m torch.distributed.launch --nproc_per_node=$GPUS --master_port=$PORT \ 10 | $(dirname "$0")/test.py $CONFIG $CHECKPOINT --launcher pytorch ${@:4} 11 | -------------------------------------------------------------------------------- /tools/dist_train.sh: -------------------------------------------------------------------------------- 1 | #!/usr/bin/env bash 2 | 3 | CONFIG=$1 4 | GPUS=$2 5 | PORT=${PORT:-29500} 6 | 7 | PYTHONPATH="$(dirname $0)/..":$PYTHONPATH \ 8 | python -m torch.distributed.launch --nproc_per_node=$GPUS --master_port=$PORT \ 9 | $(dirname "$0")/train.py $CONFIG --launcher pytorch ${@:3} 10 | -------------------------------------------------------------------------------- /tools/print_config.py: -------------------------------------------------------------------------------- 1 | import argparse 2 | 3 | from mmcv import Config, DictAction 4 | 5 | 6 | def parse_args(): 7 | parser = argparse.ArgumentParser(description='Print the whole config') 8 | parser.add_argument('config', help='config file path') 9 | parser.add_argument( 10 | '--options', nargs='+', action=DictAction, help='arguments in dict') 11 | args = parser.parse_args() 12 | 13 | return args 14 | 15 | 16 | def main(): 17 | args = parse_args() 18 | 19 | cfg = Config.fromfile(args.config) 20 | if args.options is not None: 21 | cfg.merge_from_dict(args.options) 22 | print(f'Config:\n{cfg.pretty_text}') 23 | 24 | 25 | if __name__ == '__main__': 26 | main() 27 | -------------------------------------------------------------------------------- /tools/publish_model.py: -------------------------------------------------------------------------------- 1 | import argparse 2 | import subprocess 3 | 4 | import torch 5 | 6 | 7 | def parse_args(): 8 | parser = argparse.ArgumentParser( 9 | description='Process a checkpoint to be published') 10 | parser.add_argument('in_file', help='input checkpoint filename') 11 | parser.add_argument('out_file', help='output checkpoint filename') 12 | args = parser.parse_args() 13 | return args 14 | 15 | 16 | def process_checkpoint(in_file, out_file): 17 | checkpoint = torch.load(in_file, map_location='cpu') 18 | # remove optimizer for smaller file size 19 | if 'optimizer' in checkpoint: 20 | del checkpoint['optimizer'] 21 | # if it is necessary to remove some sensitive data in checkpoint['meta'], 22 | # add the code here. 23 | torch.save(checkpoint, out_file) 24 | sha = subprocess.check_output(['sha256sum', out_file]).decode() 25 | if out_file.endswith('.pth'): 26 | out_file = out_file[:-4] 27 | final_file = out_file + f'-{sha[:8]}.pth' 28 | subprocess.Popen(['mv', out_file, final_file]) 29 | 30 | 31 | def main(): 32 | args = parse_args() 33 | process_checkpoint(args.in_file, args.out_file) 34 | 35 | 36 | if __name__ == '__main__': 37 | main() 38 | -------------------------------------------------------------------------------- /tools/slurm_test.sh: -------------------------------------------------------------------------------- 1 | #!/usr/bin/env bash 2 | 3 | set -x 4 | 5 | PARTITION=$1 6 | JOB_NAME=$2 7 | CONFIG=$3 8 | CHECKPOINT=$4 9 | GPUS=${GPUS:-8} 10 | GPUS_PER_NODE=${GPUS_PER_NODE:-8} 11 | CPUS_PER_TASK=${CPUS_PER_TASK:-5} 12 | PY_ARGS=${@:5} 13 | SRUN_ARGS=${SRUN_ARGS:-""} 14 | 15 | PYTHONPATH="$(dirname $0)/..":$PYTHONPATH \ 16 | srun -p ${PARTITION} \ 17 | --job-name=${JOB_NAME} \ 18 | --gres=gpu:${GPUS_PER_NODE} \ 19 | --ntasks=${GPUS} \ 20 | --ntasks-per-node=${GPUS_PER_NODE} \ 21 | --cpus-per-task=${CPUS_PER_TASK} \ 22 | --kill-on-bad-exit=1 \ 23 | ${SRUN_ARGS} \ 24 | python -u tools/test.py ${CONFIG} ${CHECKPOINT} --launcher="slurm" ${PY_ARGS} 25 | -------------------------------------------------------------------------------- /tools/slurm_train.sh: -------------------------------------------------------------------------------- 1 | #!/usr/bin/env bash 2 | 3 | set -x 4 | 5 | PARTITION=$1 6 | JOB_NAME=$2 7 | CONFIG=$3 8 | WORK_DIR=$4 9 | GPUS=${GPUS:-8} 10 | GPUS_PER_NODE=${GPUS_PER_NODE:-8} 11 | CPUS_PER_TASK=${CPUS_PER_TASK:-5} 12 | SRUN_ARGS=${SRUN_ARGS:-""} 13 | PY_ARGS=${@:5} 14 | 15 | PYTHONPATH="$(dirname $0)/..":$PYTHONPATH \ 16 | srun -p ${PARTITION} \ 17 | --job-name=${JOB_NAME} \ 18 | --gres=gpu:${GPUS_PER_NODE} \ 19 | --ntasks=${GPUS} \ 20 | --ntasks-per-node=${GPUS_PER_NODE} \ 21 | --cpus-per-task=${CPUS_PER_TASK} \ 22 | --kill-on-bad-exit=1 \ 23 | ${SRUN_ARGS} \ 24 | python -u tools/train.py ${CONFIG} --work-dir=${WORK_DIR} --launcher="slurm" ${PY_ARGS} 25 | --------------------------------------------------------------------------------