search

open-mmlab / mmdetection

OpenMMLab Detection Toolbox and Benchmark
https://mmdetection.readthedocs.io
Apache License 2.0
29.61k stars 9.47k forks source link

mmdetection v1.0 cascade rcnn模型 转成v2后finetune问题 #5679

Open charles3349 opened 3 years ago

charles3349 commented 3 years ago

mmdetection v1.0 cascade rcnn模型 转成v2后finetune，与直接在mmdet v1.0 finetune对比，发现v2版本 loss刚开始为8.1，第一个epoch后map 为74.8。

2021-07-24 16:36:45,713 - mmdet - INFO - Epoch [1][100/14009] lr: 4.653e-05, eta: 13 days, 22:43:14, time: 2.868, data_time: 0.802, memory: 25562, loss_rpn_cls: 0.0145, loss_rpn_bbox: 0.0251, s0.loss_cls: 4.3144, s0.acc: 45.0299, s0.loss_bbox: 0.1248, s1.loss_cls: 2.4995, s1.acc: 35.3806, s1.loss_bbox: 0.2038, s2.loss_cls: 1.3757, s2.acc: 26.1961, s2.loss_bbox: 0.1660, loss: 8.7236, grad_norm: 40.1798 INFO:mmdet:Epoch [1][100/14009] lr: 4.653e-05, eta: 13 days, 22:43:14, time: 2.868, data_time: 0.802, memory: 25562, loss_rpn_cls: 0.0145, loss_rpn_bbox: 0.0251, s0.loss_cls: 4.3144, s0.acc: 45.0299, s0.loss_bbox: 0.1248, s1.loss_cls: 2.4995, s1.acc: 35.3806, s1.loss_bbox: 0.2038, s2.loss_cls: 1.3757, s2.acc: 26.1961, s2.loss_bbox: 0.1660, loss: 8.7236, grad_norm: 40.1798 v1.0 直接finetune，刚开始loss为0.6，第一个epoch map为78，已经收敛 2021-05-31 12:09:32,203 - INFO - Epoch [1][100/15757] lr: 0.00005, eta: 2 days, 18:25:47, time: 1.899, data_time: 0.274, memory: 16455, loss_rpn_cls: 0.0119, loss_rpn_bbox: 0.0203, s0.loss_cls: 0.1438, s0.acc: 94.4299, s0.loss_bbox: 0.0904, s1.loss_cls: 0.0782, s1.acc: 93.8354, s1.loss_bbox: 0.1670, s2.loss_cls: 0.0401, s2.acc: 93.6331, s2.loss_bbox: 0.1367, loss: 0.6883 两次训练数据都一样，v1的config如下

model= dict(
    type='CascadeRCNN',
    num_stages=3,
    #pretrained='/media/senter/0a0d2450-80d0-4ab4-87c4-45613e982e6b/Models/pretrain_models/open-mmlab/resnext101_64x4d-ee2c6f71.pth',
    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'),
    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, 16, 32, 64],
        anchor_ratios=[0.5, 1.0, 2.0],
        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=11,
            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=11,
            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=11,
            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))
    ])
# 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='OHEMSampler',
                num=512,
                pos_fraction=0.25,
                neg_pos_ub=-1,
                add_gt_as_proposals=True),
            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='OHEMSampler',
                num=512,
                pos_fraction=0.25,
                neg_pos_ub=-1,
                add_gt_as_proposals=True),
            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='OHEMSampler',
                num=512,
                pos_fraction=0.25,
                neg_pos_ub=-1,
                add_gt_as_proposals=True),
            pos_weight=-1,
            debug=False)
    ],
    stage_loss_weights=[1, 0.5, 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='soft_nms', iou_thr=0.5, min_score=0.05), max_per_img=100),
    keep_all_stages=False)
# dataset settings
dataset_type = 'VOCDataset'

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),
    dict(type='Resize', img_scale=[(1000, 800),(1600,1200)], keep_ratio=True),
    dict(type='RandomFlip', flip_ratio=0.5),
    dict(type='Normalize', **img_norm_cfg),
    dict(type='Pad', size_divisor=32),
    dict(type='DefaultFormatBundle'),
    dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels']),
]
test_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(
        type='MultiScaleFlipAug',
        img_scale=(1600, 1200),
        flip=False,
        transforms=[
            dict(type='Resize', keep_ratio=True),
            dict(type='RandomFlip'),
            dict(type='Normalize', **img_norm_cfg),
            dict(type='Pad', size_divisor=32),
            dict(type='ImageToTensor', keys=['img']),
            dict(type='Collect', keys=['img']),
        ])
]
data = dict(
    imgs_per_gpu=3,
    workers_per_gpu=16,
    train=dict(
        type=dataset_type,
        ann_file='/media/senter/disk2/ShuDianTongDao/00_TrainData/imagesets/20210531_train.txt',
        img_prefix='/media/senter',
        pipeline=train_pipeline),
    val=dict(
        type=dataset_type,
        ann_file='/media/senter/0a0d2450-80d0-4ab4-87c4-45613e982e6b/ShuDianYinHuanDataset/00_DatasetComplete/01TrainTestDataset/EvaluateDataset20200912-useless/FillNiaoHai16781/images.txt',
        img_prefix='/media/senter',
        pipeline=test_pipeline),
    test=dict(
        type=dataset_type,
        ann_file='/media/senter/0a0d2450-80d0-4ab4-87c4-45613e982e6b/ShuDianYinHuanDataset/00_DatasetComplete/01TrainTestDataset/EvaluateDataset20200912-useless/FillNiaoHai16781/images.txt',
        img_prefix='/media/senter',
        pipeline=test_pipeline))
# optimizer
optimizer = dict(type='SGD', lr=0.0001, 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=[3, 5])
checkpoint_config = dict(interval=1)
# yapf:disable
log_config = dict(
    interval=100,
    hooks=[
        dict(type='TextLoggerHook'),
        dict(type='TensorboardLoggerHook')
    ])
# yapf:enable
# runtime settings
total_epochs = 8
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = '/media/senter/0a0d2450-80d0-4ab4-87c4-45613e982e6b/Models/trained_models/ShuDianTongDao/20210531_cascade_rcnn_x101_64x4d_fpn_old_10cls'
load_from = None#'/media/senter/0a0d2450-80d0-4ab4-87c4-45613e982e6b/Models/trained_models/ShuDianTongDao/20210227_cascade_rcnn_x101_64x4d_fpn_old_10cls/epoch_13.pth'
resume_from = '/media/senter/0a0d2450-80d0-4ab4-87c4-45613e982e6b/Models/trained_models/ShuDianTongDao/20210531_cascade_rcnn_x101_64x4d_fpn_old_10cls/epoch_1.pth'
workflow = [('train', `1)]`

v2 config为

`model = dict(
    type='CascadeRCNN',
    pretrained=
    '/media/senter/0a0d2450-80d0-4ab4-87c4-45613e982e6b/Models/pretrain_models/open-mmlab/resnext101_64x4d-ee2c6f71.pth',
    backbone=dict(
        type='ResNeXt',
        depth=101,
        num_stages=4,
        out_indices=(0, 1, 2, 3),
        frozen_stages=1,
        style='pytorch',
        groups=64,
        base_width=4),
    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_generator=dict(
            type='AnchorGenerator',
            scales=[8, 16, 32, 64],
            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='Shared2FCBBoxHead',
                in_channels=256,
                fc_out_channels=1024,
                roi_feat_size=7,
                num_classes=10,
                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=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='Shared2FCBBoxHead',
                in_channels=256,
                fc_out_channels=1024,
                roi_feat_size=7,
                num_classes=10,
                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=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='Shared2FCBBoxHead',
                in_channels=256,
                fc_out_channels=1024,
                roi_feat_size=7,
                num_classes=10,
                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=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))
        ]))
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=0.4,
                neg_iou_thr=0.4,
                min_pos_iou=0.4,
                match_low_quality=False,
                ignore_iof_thr=-1),
            sampler=dict(
                type='OHEMSampler',
                num=512,
                pos_fraction=0.25,
                neg_pos_ub=-1,
                add_gt_as_proposals=True),
            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,
                match_low_quality=False,
                ignore_iof_thr=-1),
            sampler=dict(
                type='OHEMSampler',
                num=512,
                pos_fraction=0.25,
                neg_pos_ub=-1,
                add_gt_as_proposals=True),
            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='OHEMSampler',
                num=512,
                pos_fraction=0.25,
                neg_pos_ub=-1,
                add_gt_as_proposals=True),
            pos_weight=-1,
            debug=False)
    ],
    stage_loss_weights=[1, 0.5, 0.25])
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='soft_nms', iou_threshold=0.5, min_score=0.05),
        max_per_img=100),
    keep_all_stages=False)
dataset_type = 'VOCDataset'
data_root = '/media/senter'
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),
    dict(
        type='Resize',
        img_scale=[(1000, 800), (1600, 1200)],
        #multiscale_mode='range',
        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'])
]
test_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(
        type='MultiScaleFlipAug',
        img_scale=(1600, 1200),
        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=3,
    workers_per_gpu=16,
    train=dict(
        type='VOCDataset',
        ann_file=
        '/media/senter/disk2/ShuDianTongDao/00_TrainData/imagesets/20210608_train_.txt',
        img_prefix='/media/senter',
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(type='LoadAnnotations', with_bbox=True),
            dict(
                type='Resize',
                img_scale=[(1000, 800), (1600, 1200)],
                #multiscale_mode='range',
                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'])
        ]),
    val=dict(
        type='VOCDataset',
        ann_file=
        '/media/senter/0a0d2450-80d0-4ab4-87c4-45613e982e6b/ShuDianYinHuanDataset/00_DatasetComplete/01TrainTestDataset/EvaluateDataset20200912-useless/FillNiaoHai16781/images.txt',
        img_prefix='/media/senter',
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(
                type='MultiScaleFlipAug',
                img_scale=(1600, 1200),
                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='VOCDataset',
        ann_file='home/bj/mmdetection_202105/images2.txt',
        img_prefix='/media/senter',
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(
                type='MultiScaleFlipAug',
                img_scale=(1600, 1200),
                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(interval=1, metric='mAP')
optimizer = dict(type='SGD', lr=0.0001, momentum=0.9, weight_decay=0.0001)
optimizer_config = dict(grad_clip=dict(max_norm=35, norm_type=2))
lr_config = dict(
    policy='step',
    warmup='linear',
    warmup_iters=500,
    warmup_ratio=0.3333333333333333,
    step=[10, 15])
runner = dict(type='EpochBasedRunner', max_epochs=30)
checkpoint_config = dict(interval=1)
log_config = dict(interval=100, hooks=[dict(type='TextLoggerHook')])
custom_hooks = [dict(type='NumClassCheckHook')]
dist_params = dict(backend='nccl')
log_level = 'INFO'
load_from = "2020_mm_v44.pth"
resume_from = None#"2020_mm_v44.pth"
workflow = [('train', 1)]
work_dir = './work_dirs/jxgs_from_v44'
gpu_ids = range(0, 4)
HappyPeanuts commented 3 years ago

你好，请教一下，我现在想在我的模型中使用TTFNet的TTFHead，我应该择那么操作，我现在的虚拟环境中已经安装好了mmdet，我是否需要在自己的项目中将mmdet文件夹放在我自己的项目里边，具体放在什么位置，应该如何调用，希望你能给一点点建议，非常感谢