add test pipeline

configs/rs_cascade_mask_rcnn_r50_fpn_ohem.py

@@ -196,7 +196,7 @@
     keep_all_stages=False)
 # dataset settings
 dataset_type = 'CocoDataset'
-data_root = './data/rscup/'
+data_root = '/home/xfr/git_mm/mmdetection/data/rscup/'
 aug_root = "./data/rscup/aug/"
 img_norm_cfg = dict(
     mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)

configs/rscup/htc_next_3s.py

@@ -276,8 +276,8 @@
         with_label=True),
     test=dict(
         type=dataset_type,
-        ann_file='./data/rscup/annotation/annos_rscup_test.json',
-        img_prefix='./data/rscup/test',
+        ann_file='./data/rscup/annotation/annos_rscup_val.json',
+        img_prefix='./data/rscup/val',
         img_scale=(512, 512),
         img_norm_cfg=img_norm_cfg,
         size_divisor=32,

configs/rscup/htc_next_3s_ft.py

@@ -0,0 +1,314 @@
+# model settings
+fp16 = dict(loss_scale=512.)
+# norm_cfg = dict(type='GN', num_groups=32, requires_grad=True)
+norm_cfg = dict(type='SyncBN', requires_grad=True)
+model = dict(
+    type='HybridTaskCascade',
+    num_stages=3,
+    pretrained='open-mmlab://resnext101_64x4d',
+    interleaved=True,
+    mask_info_flow=True,
+    backbone=dict(
+        type='ResNeXt',
+        depth=101,
+        groups=64,
+        base_width=4,
+        num_stages=4,
+        out_indices=(0, 1, 2, 3),
+        frozen_stages=1,
+        style='pytorch',
+        dcn=dict(
+            modulated=False,
+            groups=64,
+            deformable_groups=1,
+            fallback_on_stride=False),
+            stage_with_dcn=(False, True, True, True),
+        ),
+    neck=dict(
+        type='FPN',
+        in_channels=[256, 512, 1024, 2048],
+        out_channels=256,
+        num_outs=5),
+    rpn_head=dict(
+        type='RPNHead',
+        in_channels=256,
+        feat_channels=256,
+        anchor_scales=[8],
+        anchor_ratios=[0.5, 1.0, 1.5, 2.5, 5.0, 7.0],
+        anchor_strides=[4, 8, 16, 32, 64],
+        target_means=[.0, .0, .0, .0],
+        target_stds=[1.0, 1.0, 1.0, 1.0],
+        loss_cls=dict(
+            type='CrossEntropyLoss', use_sigmoid=True, loss_weight=1.0),
+        loss_bbox=dict(type='SmoothL1Loss', beta=1.0 / 9.0, loss_weight=1.0)),
+    bbox_roi_extractor=dict(
+        type='SingleRoIExtractor',
+        roi_layer=dict(type='RoIAlign', out_size=7, sample_num=2),
+        out_channels=256,
+        featmap_strides=[4, 8, 16, 32]),
+    bbox_head=[
+        dict(
+            type='SharedFCBBoxHead',
+            num_fcs=2,
+            in_channels=256,
+            fc_out_channels=1024,
+            roi_feat_size=7,
+            num_classes=19,
+            target_means=[0., 0., 0., 0.],
+            target_stds=[0.1, 0.1, 0.2, 0.2],
+            reg_class_agnostic=True,
+            loss_cls=dict(
+                type='CrossEntropyLoss',
+                use_sigmoid=False,
+                loss_weight=1.0),
+            loss_bbox=dict(
+                type='SmoothL1Loss',
+                beta=1.0,
+                loss_weight=1.0)),
+
+        dict(
+            type='SharedFCBBoxHead',
+            num_fcs=2,
+            in_channels=256,
+            fc_out_channels=1024,
+            roi_feat_size=7,
+            num_classes=19,
+            target_means=[0., 0., 0., 0.],
+            target_stds=[0.05, 0.05, 0.1, 0.1],
+            reg_class_agnostic=True,
+            loss_cls=dict(
+                type='CrossEntropyLoss',
+                use_sigmoid=False,
+                loss_weight=1.0),
+            loss_bbox=dict(
+                type='SmoothL1Loss',
+                beta=1.0,
+                loss_weight=1.0)),
+        dict(
+            type='SharedFCBBoxHead',
+            num_fcs=2,
+            in_channels=256,
+            fc_out_channels=1024,
+            roi_feat_size=7,
+            num_classes=19,
+            target_means=[0., 0., 0., 0.],
+            target_stds=[0.033, 0.033, 0.067, 0.067],
+            reg_class_agnostic=True,
+            loss_cls=dict(
+                type='CrossEntropyLoss',
+                use_sigmoid=False,
+                loss_weight=1.0),
+            loss_bbox=dict(
+                type='SmoothL1Loss',
+                beta=1.0,
+                loss_weight=1.0))
+    ],
+    mask_roi_extractor=dict(
+        type='SingleRoIExtractor',
+        roi_layer=dict(type='RoIAlign', out_size=14, sample_num=2),
+        out_channels=256,
+        featmap_strides=[4, 8, 16, 32]),
+    mask_head=dict(
+        type='HTCMaskHead',
+        num_convs=4,
+        in_channels=256,
+        conv_out_channels=256,
+        num_classes=19,
+        loss_mask=dict(
+            type='CrossEntropyLoss', use_mask=True, loss_weight=1.0)))
+# model training and testing settings
+train_cfg = dict(
+    rpn=dict(
+        assigner=dict(
+            type='MaxIoUAssigner',
+            pos_iou_thr=0.7,
+            neg_iou_thr=0.3,
+            min_pos_iou=0.3,
+            ignore_iof_thr=-1),
+        sampler=dict(
+            type='RandomSampler',
+            num=256,
+            pos_fraction=0.5,
+            neg_pos_ub=-1,
+            add_gt_as_proposals=False),
+        allowed_border=0,
+        pos_weight=-1,
+        debug=False),
+    rpn_proposal=dict(
+        nms_across_levels=False,
+        nms_pre=2000,
+        nms_post=2000,
+        max_num=2000,
+        nms_thr=0.7,
+        min_bbox_size=0),
+    rcnn=[
+        dict(
+            assigner=dict(
+                type='MaxIoUAssigner',
+                pos_iou_thr=0.4,
+                neg_iou_thr=0.4,
+                min_pos_iou=0.4,
+                ignore_iof_thr=-1),
+            sampler=dict(
+                type='CombinedSampler',
+                num=512,
+                pos_fraction=0.25,
+                add_gt_as_proposals=True,
+                pos_sampler=dict(type='InstanceBalancedPosSampler'),
+                neg_sampler=dict(
+                    type='IoUBalancedNegSampler',
+                    floor_thr=-1,
+                    floor_fraction=0,
+                    num_bins=3)),
+            mask_size=28,
+            pos_weight=-1,
+            debug=False),
+        dict(
+            assigner=dict(
+                type='MaxIoUAssigner',
+                pos_iou_thr=0.5,
+                neg_iou_thr=0.5,
+                min_pos_iou=0.5,
+                ignore_iof_thr=-1),
+            sampler=dict(
+                type='CombinedSampler',
+                num=512,
+                pos_fraction=0.25,
+                add_gt_as_proposals=True,
+                pos_sampler=dict(type='InstanceBalancedPosSampler'),
+                neg_sampler=dict(
+                    type='IoUBalancedNegSampler',
+                    floor_thr=-1,
+                    floor_fraction=0,
+                    num_bins=3)),
+            mask_size=28,
+            pos_weight=-1,
+            debug=False),
+        dict(
+            assigner=dict(
+                type='MaxIoUAssigner',
+                pos_iou_thr=0.6,
+                neg_iou_thr=0.6,
+                min_pos_iou=0.6,
+                ignore_iof_thr=-1),
+            sampler=dict(
+                type='CombinedSampler',
+                num=512,
+                pos_fraction=0.25,
+                add_gt_as_proposals=True,
+                pos_sampler=dict(type='InstanceBalancedPosSampler'),
+                neg_sampler=dict(
+                    type='IoUBalancedNegSampler',
+                    floor_thr=-1,
+                    floor_fraction=0,
+                    num_bins=3)),
+            mask_size=28,
+            pos_weight=-1,
+            debug=False)
+    ],
+    stage_loss_weights=[0.5, 1, 0.25])
+test_cfg = dict(
+    rpn=dict(
+        nms_across_levels=False,
+        nms_pre=1000,
+        nms_post=1000,
+        max_num=1000,
+        nms_thr=0.7,
+        min_bbox_size=0),
+    rcnn=dict(
+        score_thr=0.05,
+        nms=dict(type='nms', iou_thr=0.5),
+        max_per_img=1000,
+        mask_thr_binary=0.5),
+    keep_all_stages=False)
+# dataset settings
+dataset_type = 'CocoDataset'
+data_root = './data/rscup/'
+aug_root = "./data/rscup/aug/"
+other_aug_root = "./data/rscup/otheraug/"
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
+data = dict(
+    imgs_per_gpu=4,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        ann_file=(data_root + 'annotation/annos_rscup_train.json',
+                  data_root + 'annotation/annos_rscup_s2.json',
+                  aug_root + 'annos_rscup_airport.json',
+                  other_aug_root + "annos_rscup_baseball-diamond.json",
+                  other_aug_root + "annos_rscup_basketball-court.json",
+                  other_aug_root + "annos_rscup_container-crane.json",
+                  other_aug_root + "annos_rscup_helicopter.json",
+                  other_aug_root + "annos_rscup_helipad.json",
+                  other_aug_root + "annos_rscup_helipad_ship.json",
+                  other_aug_root + "annos_rscup_roundabout.json",
+                  other_aug_root + "annos_rscup_soccer-ball-field_ground-track-field.json",
+                  ),
+        img_prefix=(data_root + 'train/',
+                    data_root + 's2/',
+                    aug_root + "airport/",
+                    other_aug_root + "baseball-diamond",
+                    other_aug_root + "basketball-court",
+                    other_aug_root + "container-crane",
+                    other_aug_root + "helicopter",
+                    other_aug_root + "helipad",
+                    other_aug_root + "helipad_ship",
+                    other_aug_root + "roundabout",
+                    other_aug_root + "soccer-ball-field_ground-track-field"),
+        img_scale=(512, 512),
+        img_norm_cfg=img_norm_cfg,
+        size_divisor=32,
+        flip_ratio=0.5,
+        with_mask=True,
+        with_crowd=True,
+        with_label=True),
+    val=dict(
+        type=dataset_type,
+        ann_file=data_root + 'annotation/annos_rscup_val.json',
+        img_prefix=data_root + 'val/',
+        img_scale=(512, 512),
+        img_norm_cfg=img_norm_cfg,
+        size_divisor=32,
+        flip_ratio=0,
+        with_mask=True,
+        with_crowd=True,
+        with_label=True),
+    test=dict(
+        type=dataset_type,
+        ann_file='./data/rscup/annotation/annos_rscup_val.json',
+        img_prefix='./data/rscup/val',
+        img_scale=(512, 512),
+        img_norm_cfg=img_norm_cfg,
+        size_divisor=32,
+        flip_ratio=0,
+        with_mask=True,
+        with_label=False,
+        test_mode=True))
+# optimizer
+optimizer = dict(type='SGD', lr=6e-4, momentum=0.9, weight_decay=0.0001)
+optimizer_config = dict(grad_clip=dict(max_norm=35, norm_type=2))
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=1.0 / 3,
+    step=[1])
+checkpoint_config = dict(interval=1)
+# yapf:disable
+log_config = dict(
+    interval=50,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        dict(type='TensorboardLoggerHook')
+    ])
+# yapf:enable
+# runtime settings
+total_epochs = 2
+dist_params = dict(backend='nccl')
+log_level = 'INFO'
+work_dir = './work_dirs/htc_next_3s_ft'
+load_from = None
+resume_from = None
+workflow = [('train', 1)]

configs/rscup/htc_next_ft.py

@@ -276,10 +276,11 @@
         with_label=True),
     test=dict(
         type=dataset_type,
-        ann_file='./data/rscup/annotation/annos_rscup_test.json',
-        img_prefix='./data/rscup/test',
+        ann_file='./data/rscup/annotation/annos_rscup_val.json',
+        img_prefix='./data/rscup/val',
         img_scale=(512, 512),
         img_norm_cfg=img_norm_cfg,
+        resize_keep_ratio=True,
         size_divisor=32,
         flip_ratio=0,
         with_mask=True,

configs/rscup/htc_sy.py

@@ -1,5 +1,5 @@
 # model settings
-fp16 = dict(loss_scale=512.)
+# fp16 = dict(loss_scale=512.)
 # norm_cfg = dict(type='GN', num_groups=32, requires_grad=True)
 norm_cfg = dict(type='SyncBN', requires_grad=True)
 model = dict(
@@ -274,6 +274,7 @@
         img_scale=(512, 512),
         img_norm_cfg=img_norm_cfg,
         size_divisor=32,
+        resize_keep_ratio=True,
         flip_ratio=0,
         with_mask=True,
         with_label=False,

configs/rscup/htc_sy_next.py

@@ -274,8 +274,8 @@
         with_label=True),
     test=dict(
         type=dataset_type,
-        ann_file='./data/rscup/annotation/annos_rscup_test.json',
-        img_prefix='./data/rscup/test',
+        ann_file='./data/rscup/annotation/annos_rscup_val.json',
+        img_prefix='./data/rscup/val',
         img_scale=(512, 512),
         img_norm_cfg=img_norm_cfg,
         size_divisor=32,