configs/segmentation/fcn/fcn_r50-d8_512x512_8xb4_60e_voc12aug.py [64:165]: - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - train_img_root = data_root + 'JPEGImages' train_label_root = data_root + 'SegmentationClass' train_list_file = data_root + 'ImageSets/Segmentation/train.txt' train_aug_img_root = data_root + 'JPEGImages' train_aug_label_root = data_root + 'SegmentationClassAug' train_aug_list_file = data_root + 'ImageSets/Segmentation/aug.txt' val_img_root = data_root + 'JPEGImages' val_label_root = data_root + 'SegmentationClass' val_list_file = data_root + 'ImageSets/Segmentation/val.txt' test_img_root = data_root + 'JPEGImages' test_label_root = data_root + 'SegmentationClass' test_list_file = data_root + 'ImageSets/Segmentation/test.txt' img_norm_cfg = dict( mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True) crop_size = (512, 512) train_pipeline = [ dict(type='MMResize', img_scale=(2048, 512), ratio_range=(0.5, 2.0)), dict(type='SegRandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='MMRandomFlip', flip_ratio=0.5), dict(type='MMPhotoMetricDistortion'), dict(type='MMNormalize', **img_norm_cfg), dict(type='MMPad', size=crop_size), dict(type='DefaultFormatBundle'), dict( type='Collect', keys=['img', 'gt_semantic_seg'], meta_keys=('filename', 'ori_filename', 'ori_shape', 'img_shape', 'pad_shape', 'scale_factor', 'flip', 'flip_direction', 'img_norm_cfg')), ] test_pipeline = [ dict( type='MMMultiScaleFlipAug', img_scale=(2048, 512), # img_ratios=[0.5, 0.75, 1.0, 1.25, 1.5, 1.75], flip=False, transforms=[ dict(type='MMResize', keep_ratio=True), dict(type='MMRandomFlip'), dict(type='MMNormalize', **img_norm_cfg), dict(type='ImageToTensor', keys=['img']), dict( type='Collect', keys=['img'], meta_keys=('filename', 'ori_filename', 'ori_shape', 'img_shape', 'pad_shape', 'scale_factor', 'flip', 'flip_direction', 'img_norm_cfg')), ]) ] data = dict( imgs_per_gpu=4, workers_per_gpu=4, train=dict( type=dataset_type, ignore_index=255, data_source=dict( type='SourceConcat', data_source_list=[ dict( type='SegSourceRaw', img_root=train_img_root, label_root=train_label_root, split=train_list_file, classes=CLASSES), dict( type='SegSourceRaw', img_root=train_aug_img_root, label_root=train_aug_label_root, split=train_aug_list_file, classes=CLASSES), ]), pipeline=train_pipeline), val=dict( imgs_per_gpu=1, ignore_index=255, type=dataset_type, data_source=dict( type='SegSourceRaw', img_root=val_img_root, label_root=val_label_root, split=val_list_file, classes=CLASSES, ), pipeline=test_pipeline), test=dict( type=dataset_type, data_source=dict( type='SegSourceRaw', img_root=test_img_root, label_root=test_label_root, split=test_list_file, classes=CLASSES, ), pipeline=test_pipeline)) # optimizer optimizer = dict(type='SGD', lr=0.01, momentum=0.9, weight_decay=0.0005) optimizer_config = dict() - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - configs/segmentation/upernet/upernet_r50_512x512_8xb4_60e_voc12aug.py [62:163]: - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - train_img_root = data_root + 'JPEGImages' train_label_root = data_root + 'SegmentationClass' train_list_file = data_root + 'ImageSets/Segmentation/train.txt' train_aug_img_root = data_root + 'JPEGImages' train_aug_label_root = data_root + 'SegmentationClassAug' train_aug_list_file = data_root + 'ImageSets/Segmentation/aug.txt' val_img_root = data_root + 'JPEGImages' val_label_root = data_root + 'SegmentationClass' val_list_file = data_root + 'ImageSets/Segmentation/val.txt' test_img_root = data_root + 'JPEGImages' test_label_root = data_root + 'SegmentationClass' test_list_file = data_root + 'ImageSets/Segmentation/test.txt' img_norm_cfg = dict( mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True) crop_size = (512, 512) train_pipeline = [ dict(type='MMResize', img_scale=(2048, 512), ratio_range=(0.5, 2.0)), dict(type='SegRandomCrop', crop_size=crop_size, cat_max_ratio=0.75), dict(type='MMRandomFlip', flip_ratio=0.5), dict(type='MMPhotoMetricDistortion'), dict(type='MMNormalize', **img_norm_cfg), dict(type='MMPad', size=crop_size), dict(type='DefaultFormatBundle'), dict( type='Collect', keys=['img', 'gt_semantic_seg'], meta_keys=('filename', 'ori_filename', 'ori_shape', 'img_shape', 'pad_shape', 'scale_factor', 'flip', 'flip_direction', 'img_norm_cfg')), ] test_pipeline = [ dict( type='MMMultiScaleFlipAug', img_scale=(2048, 512), # img_ratios=[0.5, 0.75, 1.0, 1.25, 1.5, 1.75], flip=False, transforms=[ dict(type='MMResize', keep_ratio=True), dict(type='MMRandomFlip'), dict(type='MMNormalize', **img_norm_cfg), dict(type='ImageToTensor', keys=['img']), dict( type='Collect', keys=['img'], meta_keys=('filename', 'ori_filename', 'ori_shape', 'img_shape', 'pad_shape', 'scale_factor', 'flip', 'flip_direction', 'img_norm_cfg')), ]) ] data = dict( imgs_per_gpu=4, workers_per_gpu=4, train=dict( type=dataset_type, ignore_index=255, data_source=dict( type='SourceConcat', data_source_list=[ dict( type='SegSourceRaw', img_root=train_img_root, label_root=train_label_root, split=train_list_file, classes=CLASSES), dict( type='SegSourceRaw', img_root=train_aug_img_root, label_root=train_aug_label_root, split=train_aug_list_file, classes=CLASSES), ]), pipeline=train_pipeline), val=dict( imgs_per_gpu=1, ignore_index=255, type=dataset_type, data_source=dict( type='SegSourceRaw', img_root=val_img_root, label_root=val_label_root, split=val_list_file, classes=CLASSES, ), pipeline=test_pipeline), test=dict( type=dataset_type, data_source=dict( type='SegSourceRaw', img_root=test_img_root, label_root=test_label_root, split=test_list_file, classes=CLASSES, ), pipeline=test_pipeline)) # optimizer optimizer = dict(type='SGD', lr=0.01, momentum=0.9, weight_decay=0.0005) optimizer_config = dict() - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -