search

SwinTransformer / Swin-Transformer-Object-Detection

This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows" on Object Detection and Instance Segmentation.
https://arxiv.org/abs/2103.14030
Apache License 2.0
1.79k stars 378 forks source link

Inference results AP AR mAP mAR are all -1 #67

Closed ngfuong closed 3 years ago

ngfuong commented 3 years ago

I trained Swin Transformers Cityscapes Instance Segmentation from pretrained Swin Base on ImageNet 22K. During training, evaluations are very normal. E.g:

2021-06-29 00:57:35,334 - mmdet - INFO - Evaluating segm...
Loading and preparing results...
DONE (t=0.44s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *segm*
DONE (t=7.74s).
Accumulating evaluation results...
DONE (t=0.63s).
Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.339
Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=1000 ] = 0.590
Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=1000 ] = -1.000
Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.132
Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.317
Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.530
Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.412
Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=300 ] = 0.412
Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=1000 ] = 0.412
Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = 0.182
Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = 0.386
Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = 0.620

However, after testing (test.py) using the epoch_*.pth checkpoint file, all the AP and AR are -1. If I export all of the predictions the images are all black. This is the testing output:

apex is not installed
apex is not installed
apex is not installed
apex is not installed
loading annotations into memory...
Done (t=0.00s)
creating index...
index created!
Use load_from_local loader
[                                                  ] 0/1525, elapsed: 0s, ETA:/usr/local/lib/python3.7/dist-packages/torch/utils/checkpoint.py:25: UserWarning: None of the inputs have requires_grad=True. Gradients will be None
  warnings.warn("None of the inputs have requires_grad=True. Gradients will be None")
[>>] 1525/1525, 1.4 task/s, elapsed: 1085s, ETA:     0sloading annotations into memory...
Done (t=0.00s)
creating index...
index created!
loading annotations into memory...
Done (t=0.00s)
creating index...
index created!

Evaluating bbox...
Loading and preparing results...
DONE (t=0.73s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=2.28s).
Accumulating evaluation results...
DONE (t=0.55s).
Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = -1.000
Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=1000 ] = -1.000
Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=1000 ] = -1.000
Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = -1.000
Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = -1.000
Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = -1.000
Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = -1.000
Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=300 ] = -1.000
Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=1000 ] = -1.000
Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = -1.000
Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = -1.000
Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = -1.000

Evaluating segm...
Loading and preparing results...
DONE (t=0.71s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *segm*
DONE (t=2.37s).
Accumulating evaluation results...
DONE (t=0.56s).
Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = -1.000
Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=1000 ] = -1.000
Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=1000 ] = -1.000
Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = -1.000
Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = -1.000
Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = -1.000
Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = -1.000
Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=300 ] = -1.000
Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=1000 ] = -1.000
Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=1000 ] = -1.000
Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=1000 ] = -1.000
Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=1000 ] = -1.000
{'bbox_mAP': -1.0, 'bbox_mAP_50': -1.0, 'bbox_mAP_75': -1.0, 'bbox_mAP_s': -1.0, 'bbox_mAP_m': -1.0, 'bbox_mAP_l': -1.0, 'bbox_mAP_copypaste': '-1.000 -1.000 -1.000 -1.000 -1.000 -1.000', 'segm_mAP': -1.0, 'segm_mAP_50': -1.0, 'segm_mAP_75': -1.0, 'segm_mAP_s': -1.0, 'segm_mAP_m': -1.0, 'segm_mAP_l': -1.0, 'segm_mAP_copypaste': '-1.000 -1.000 -1.000 -1.000 -1.000 -1.000'}

This is my configs:

TorchVision: 0.7.0+cu101
OpenCV: 4.1.2
MMCV: 1.3.0
MMCV Compiler: GCC 7.3
MMCV CUDA Compiler: 10.1
MMDetection: 2.11.0+
------------------------------------------------------------

2021-07-11 04:40:34,123 - mmdet - INFO - Distributed training: False
2021-07-11 04:40:38,154 - mmdet - INFO - Config:
num_classes = 8
model = dict(
    type='CascadeRCNN',
    pretrained=None,
    backbone=dict(
        type='SwinTransformer',
        embed_dim=96,
        depths=[2, 2, 6, 2],
        num_heads=[3, 6, 12, 24],
        window_size=7,
        mlp_ratio=4.0,
        qkv_bias=True,
        qk_scale=None,
        drop_rate=0.0,
        attn_drop_rate=0.0,
        drop_path_rate=0.0,
        ape=False,
        patch_norm=True,
        out_indices=(0, 1, 2, 3),
        use_checkpoint=True),
    neck=dict(
        type='FPN',
        in_channels=[96, 192, 384, 768],
        out_channels=256,
        num_outs=5),
    rpn_head=dict(
        type='RPNHead',
        in_channels=256,
        feat_channels=256,
        anchor_generator=dict(
            type='AnchorGenerator',
            scales=[8],
            ratios=[0.5, 1.0, 2.0],
            strides=[4, 8, 16, 32, 64]),
        bbox_coder=dict(
            type='DeltaXYWHBBoxCoder',
            target_means=[0.0, 0.0, 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=0.1111111111111111, loss_weight=1.0)),
    roi_head=dict(
        type='CascadeRoIHead',
        num_stages=3,
        stage_loss_weights=[1, 0.5, 0.25],
        bbox_roi_extractor=dict(
            type='SingleRoIExtractor',
            roi_layer=dict(type='RoIAlign', output_size=7, sampling_ratio=0),
            out_channels=256,
            featmap_strides=[4, 8, 16, 32]),
        bbox_head=[
            dict(
                type='ConvFCBBoxHead',
                num_shared_convs=4,
                num_shared_fcs=1,
                in_channels=256,
                conv_out_channels=256,
                fc_out_channels=1024,
                roi_feat_size=7,
                num_classes=8,
                bbox_coder=dict(
                    type='DeltaXYWHBBoxCoder',
                    target_means=[0.0, 0.0, 0.0, 0.0],
                    target_stds=[0.1, 0.1, 0.2, 0.2]),
                reg_class_agnostic=False,
                reg_decoded_bbox=True,
                norm_cfg=dict(type='BN', requires_grad=True),
                loss_cls=dict(
                    type='CrossEntropyLoss',
                    use_sigmoid=False,
                    loss_weight=1.0),
                loss_bbox=dict(type='GIoULoss', loss_weight=10.0)),
            dict(
                type='ConvFCBBoxHead',
                num_shared_convs=4,
                num_shared_fcs=1,
                in_channels=256,
                conv_out_channels=256,
                fc_out_channels=1024,
                roi_feat_size=7,
                num_classes=8,
                bbox_coder=dict(
                    type='DeltaXYWHBBoxCoder',
                    target_means=[0.0, 0.0, 0.0, 0.0],
                    target_stds=[0.05, 0.05, 0.1, 0.1]),
                reg_class_agnostic=False,
                reg_decoded_bbox=True,
                norm_cfg=dict(type='BN', requires_grad=True),
                loss_cls=dict(
                    type='CrossEntropyLoss',
                    use_sigmoid=False,
                    loss_weight=1.0),
                loss_bbox=dict(type='GIoULoss', loss_weight=10.0)),
            dict(
                type='ConvFCBBoxHead',
                num_shared_convs=4,
                num_shared_fcs=1,
                in_channels=256,
                conv_out_channels=256,
                fc_out_channels=1024,
                roi_feat_size=7,
                num_classes=8,
                bbox_coder=dict(
                    type='DeltaXYWHBBoxCoder',
                    target_means=[0.0, 0.0, 0.0, 0.0],
                    target_stds=[0.033, 0.033, 0.067, 0.067]),
                reg_class_agnostic=False,
                reg_decoded_bbox=True,
                norm_cfg=dict(type='BN', requires_grad=True),
                loss_cls=dict(
                    type='CrossEntropyLoss',
                    use_sigmoid=False,
                    loss_weight=1.0),
                loss_bbox=dict(type='GIoULoss', loss_weight=10.0))
        ],
        mask_roi_extractor=dict(
            type='SingleRoIExtractor',
            roi_layer=dict(type='RoIAlign', output_size=14, sampling_ratio=0),
            out_channels=256,
            featmap_strides=[4, 8, 16, 32]),
        mask_head=dict(
            type='FCNMaskHead',
            num_convs=4,
            in_channels=256,
            conv_out_channels=256,
            num_classes=8,
            loss_mask=dict(
                type='CrossEntropyLoss', use_mask=True, loss_weight=1.0))),
    train_cfg=dict(
        rpn=dict(
            assigner=dict(
                type='MaxIoUAssigner',
                pos_iou_thr=0.7,
                neg_iou_thr=0.3,
                min_pos_iou=0.3,
                match_low_quality=True,
                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_per_img=2000,
            nms=dict(type='nms', iou_threshold=0.7),
            min_bbox_size=0),
        rcnn=[
            dict(
                assigner=dict(
                    type='MaxIoUAssigner',
                    pos_iou_thr=0.5,
                    neg_iou_thr=0.5,
                    min_pos_iou=0.5,
                    match_low_quality=False,
                    ignore_iof_thr=-1),
                sampler=dict(
                    type='RandomSampler',
                    num=512,
                    pos_fraction=0.25,
                    neg_pos_ub=-1,
                    add_gt_as_proposals=True),
                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,
                    match_low_quality=False,
                    ignore_iof_thr=-1),
                sampler=dict(
                    type='RandomSampler',
                    num=512,
                    pos_fraction=0.25,
                    neg_pos_ub=-1,
                    add_gt_as_proposals=True),
                mask_size=28,
                pos_weight=-1,
                debug=False),
            dict(
                assigner=dict(
                    type='MaxIoUAssigner',
                    pos_iou_thr=0.7,
                    neg_iou_thr=0.7,
                    min_pos_iou=0.7,
                    match_low_quality=False,
                    ignore_iof_thr=-1),
                sampler=dict(
                    type='RandomSampler',
                    num=512,
                    pos_fraction=0.25,
                    neg_pos_ub=-1,
                    add_gt_as_proposals=True),
                mask_size=28,
                pos_weight=-1,
                debug=False)
        ]),
    test_cfg=dict(
        rpn=dict(
            nms_across_levels=False,
            nms_pre=1000,
            nms_post=1000,
            max_per_img=1000,
            nms=dict(type='nms', iou_threshold=0.7),
            min_bbox_size=0),
        rcnn=dict(
            score_thr=0.05,
            nms=dict(type='nms', iou_threshold=0.5),
            max_per_img=100,
            mask_thr_binary=0.5)))
dataset_type = 'CityscapesDataset'
data_root = 'data/cityscapes/'
img_norm_cfg = dict(
    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
train_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
    dict(type='RandomFlip', flip_ratio=0.5),
    dict(
        type='AutoAugment',
        policies=[[{
            'type':
            'Resize',
            'img_scale': [(480, 1333), (512, 1333), (544, 1333), (576, 1333),
                          (608, 1333), (640, 1333), (672, 1333), (704, 1333),
                          (736, 1333), (768, 1333), (800, 1333)],
            'multiscale_mode':
            'value',
            'keep_ratio':
            True
        }],
                  [{
                      'type': 'Resize',
                      'img_scale': [(400, 1333), (500, 1333), (600, 1333)],
                      'multiscale_mode': 'value',
                      'keep_ratio': True
                  }, {
                      'type': 'RandomCrop',
                      'crop_type': 'absolute_range',
                      'crop_size': (384, 600),
                      'allow_negative_crop': True
                  }, {
                      'type':
                      'Resize',
                      'img_scale': [(480, 1333), (512, 1333), (544, 1333),
                                    (576, 1333), (608, 1333), (640, 1333),
                                    (672, 1333), (704, 1333), (736, 1333),
                                    (768, 1333), (800, 1333)],
                      'multiscale_mode':
                      'value',
                      'override':
                      True,
                      'keep_ratio':
                      True
                  }]]),
    dict(
        type='Normalize',
        mean=[123.675, 116.28, 103.53],
        std=[58.395, 57.12, 57.375],
        to_rgb=True),
    dict(type='Pad', size_divisor=32),
    dict(type='DefaultFormatBundle'),
    dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks'])
]
test_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(
        type='MultiScaleFlipAug',
        img_scale=(2048, 1024),
        flip=False,
        transforms=[
            dict(type='Resize', keep_ratio=True),
            dict(type='RandomFlip'),
            dict(
                type='Normalize',
                mean=[123.675, 116.28, 103.53],
                std=[58.395, 57.12, 57.375],
                to_rgb=True),
            dict(type='Pad', size_divisor=32),
            dict(type='ImageToTensor', keys=['img']),
            dict(type='Collect', keys=['img'])
        ])
]
data = dict(
    samples_per_gpu=2,
    workers_per_gpu=2,
    train=dict(
        type='RepeatDataset',
        times=8,
        dataset=dict(
            type='CityscapesDataset',
            ann_file=
            'data/cityscapes/annotations/instancesonly_filtered_gtFine_train.json',
            img_prefix='data/cityscapes/leftImg8bit/train/',
            pipeline=[
                dict(type='LoadImageFromFile'),
                dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
                dict(
                    type='Resize',
                    img_scale=[(2048, 800), (2048, 1024)],
                    keep_ratio=True),
                dict(type='RandomFlip', flip_ratio=0.5),
                dict(
                    type='Normalize',
                    mean=[123.675, 116.28, 103.53],
                    std=[58.395, 57.12, 57.375],
                    to_rgb=True),
                dict(type='Pad', size_divisor=32),
                dict(type='DefaultFormatBundle'),
                dict(
                    type='Collect',
                    keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks'])
            ]),
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
            dict(type='RandomFlip', flip_ratio=0.5),
            dict(
                type='AutoAugment',
                policies=[[{
                    'type':
                    'Resize',
                    'img_scale': [(480, 1333), (512, 1333), (544, 1333),
                                  (576, 1333), (608, 1333), (640, 1333),
                                  (672, 1333), (704, 1333), (736, 1333),
                                  (768, 1333), (800, 1333)],
                    'multiscale_mode':
                    'value',
                    'keep_ratio':
                    True
                }],
                          [{
                              'type': 'Resize',
                              'img_scale': [(400, 1333), (500, 1333),
                                            (600, 1333)],
                              'multiscale_mode': 'value',
                              'keep_ratio': True
                          }, {
                              'type': 'RandomCrop',
                              'crop_type': 'absolute_range',
                              'crop_size': (384, 600),
                              'allow_negative_crop': True
                          }, {
                              'type':
                              'Resize',
                              'img_scale': [(480, 1333), (512, 1333),
                                            (544, 1333), (576, 1333),
                                            (608, 1333), (640, 1333),
                                            (672, 1333), (704, 1333),
                                            (736, 1333), (768, 1333),
                                            (800, 1333)],
                              'multiscale_mode':
                              'value',
                              'override':
                              True,
                              'keep_ratio':
                              True
                          }]]),
            dict(
                type='Normalize',
                mean=[123.675, 116.28, 103.53],
                std=[58.395, 57.12, 57.375],
                to_rgb=True),
            dict(type='Pad', size_divisor=32),
            dict(type='DefaultFormatBundle'),
            dict(
                type='Collect',
                keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks'])
        ]),
    val=dict(
        type='CityscapesDataset',
        ann_file=
        'data/cityscapes/annotations/instancesonly_filtered_gtFine_val.json',
        img_prefix='data/cityscapes/leftImg8bit/val/',
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(
                type='MultiScaleFlipAug',
                img_scale=(2048, 1024),
                flip=False,
                transforms=[
                    dict(type='Resize', keep_ratio=True),
                    dict(type='RandomFlip'),
                    dict(
                        type='Normalize',
                        mean=[123.675, 116.28, 103.53],
                        std=[58.395, 57.12, 57.375],
                        to_rgb=True),
                    dict(type='Pad', size_divisor=32),
                    dict(type='ImageToTensor', keys=['img']),
                    dict(type='Collect', keys=['img'])
                ])
        ]),
    test=dict(
        type='CityscapesDataset',
        ann_file=
        'data/cityscapes/annotations/instancesonly_filtered_gtFine_test.json',
        img_prefix='data/cityscapes/leftImg8bit/test/',
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(
                type='MultiScaleFlipAug',
                img_scale=(2048, 1024),
                flip=False,
                transforms=[
                    dict(type='Resize', keep_ratio=True),
                    dict(type='RandomFlip'),
                    dict(
                        type='Normalize',
                        mean=[123.675, 116.28, 103.53],
                        std=[58.395, 57.12, 57.375],
                        to_rgb=True),
                    dict(type='Pad', size_divisor=32),
                    dict(type='ImageToTensor', keys=['img']),
                    dict(type='Collect', keys=['img'])
                ])
        ]))
evaluation = dict(metric=['bbox', 'segm'])
optimizer = dict(
    type='AdamW',
    lr=0.0001,
    betas=(0.9, 0.999),
    weight_decay=0.05,
    paramwise_cfg=dict(
        custom_keys=dict(
            absolute_pos_embed=dict(decay_mult=0.0),
            relative_position_bias_table=dict(decay_mult=0.0),
            norm=dict(decay_mult=0.0))))
optimizer_config = dict(grad_clip=None)
lr_config = dict(
    policy='step',
    warmup='linear',
    warmup_iters=500,
    warmup_ratio=0.001,
    step=[8, 11])
runner = dict(type='EpochBasedRunner', max_epochs=12)
checkpoint_config = dict(interval=1)
log_config = dict(
    interval=50,
    hooks=[dict(type='TextLoggerHook'),
           dict(type='TensorboardLoggerHook')])
custom_hooks = [dict(type='NumClassCheckHook')]
dist_params = dict(backend='nccl')
log_level = 'INFO'
load_from = None
resume_from = '/content/drive/MyDrive/Object-Detection/Swin-Transformer-Object-Detection/checkpoints/swin_tiny_cascade_mask_cityscapes_1x/epoch_12.pth'
workflow = [('train', 1)]
work_dir = '/content/drive/MyDrive/Object-Detection/Swin-Transformer-Object-Detection/checkpoints/swin_tiny_cascade_mask_cityscapes_1x'
gpu_ids = range(0, 1)

Specifically, I changed SyncBN to BN, changed some AutoAugment policies and changed use_checkpoints from False to True to save GPU memory. There is a warning None of the inputs have requires_grad=True. Gradients will be None while inferencing. Does this have anything to do with my output? How can I fix this? Thank you!

colsona commented 2 years ago

Hello, I want to ask you how to solve this question. I meet the same question with you, and the train will end after 2 epochs.