Upload config/uwsam_teacher_paper.py with huggingface_hub

config/uwsam_teacher_paper.py

@@ -0,0 +1,342 @@
+custom_imports = dict(
+    allow_failed_imports=False, imports=[
+        'uwsam_teacher',
+    ])
+data_root = '/home/cuongvt/My-works/code/UWSAM/UIIS10K/UIIS10K/'
+default_hooks = dict(
+    checkpoint=dict(interval=1, max_keep_ckpts=3, type='CheckpointHook'),
+    logger=dict(interval=50, type='LoggerHook'),
+    param_scheduler=dict(type='ParamSchedulerHook'),
+    sampler_seed=dict(type='DistSamplerSeedHook'),
+    timer=dict(type='IterTimerHook'),
+    visualization=dict(type='DetVisualizationHook'))
+default_scope = 'mmdet'
+env_cfg = dict(
+    cuda_check_device=True,
+    cudnn_benchmark=False,
+    dist_cfg=dict(backend='nccl'),
+    mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0))
+launcher = 'none'
+load_from = None
+log_level = 'INFO'
+log_processor = dict(by_epoch=True, type='LogProcessor', window_size=50)
+metainfo = dict(
+    classes=(
+        'fish',
+        'reptiles',
+        'arthropoda',
+        'corals',
+        'mollusk',
+        'plants',
+        'ruins',
+        'garbage',
+        'human',
+        'robots',
+    ))
+model = dict(
+    backbone=dict(
+        freeze_base=True,
+        img_size=1024,
+        lora_alpha=16,
+        lora_dropout=0.0,
+        lora_r=8,
+        sam_checkpoint='/home/cuongvt/My-works/code/UWSAM/sam_vit_h_4b8939.pth',
+        sam_type='vit_h',
+        type='SAMImageEncoderLoRABackbone'),
+    data_preprocessor=dict(
+        bgr_to_rgb=True,
+        mean=[
+            123.675,
+            116.28,
+            103.53,
+        ],
+        pad_mask=True,
+        pad_size_divisor=32,
+        std=[
+            58.395,
+            57.12,
+            57.375,
+        ],
+        type='DetDataPreprocessor'),
+    mask_head=dict(
+        freeze_mask_decoder=False,
+        in_channels=256,
+        loss_iou=dict(loss_weight=0.5, type='MSELoss'),
+        loss_mask=dict(
+            loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=True),
+        num_classes=10,
+        sam_checkpoint='/home/cuongvt/My-works/code/UWSAM/sam_vit_h_4b8939.pth',
+        sam_dim=256,
+        sam_type='vit_h',
+        type='SAMHeadV2'),
+    neck=dict(
+        in_channels=256,
+        rpn_head=dict(
+            anchor_generator=dict(
+                ratios=[
+                    0.5,
+                    1.0,
+                    2.0,
+                ],
+                scales=[
+                    8,
+                ],
+                strides=[
+                    16,
+                ],
+                type='AnchorGenerator'),
+            bbox_coder=dict(
+                target_means=[
+                    0.0,
+                    0.0,
+                    0.0,
+                    0.0,
+                ],
+                target_stds=[
+                    1.0,
+                    1.0,
+                    1.0,
+                    1.0,
+                ],
+                type='DeltaXYWHBBoxCoder'),
+            feat_channels=256,
+            in_channels=256,
+            loss_bbox=dict(loss_weight=1.0, type='L1Loss'),
+            loss_cls=dict(
+                loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=True),
+            num_convs=1,
+            test_cfg=dict(
+                max_per_img=100,
+                min_bbox_size=0,
+                nms=dict(iou_threshold=0.7, type='nms'),
+                nms_pre=300),
+            train_cfg=dict(
+                allowed_border=-1,
+                assigner=dict(
+                    ignore_iof_thr=-1,
+                    match_low_quality=True,
+                    min_pos_iou=0.3,
+                    neg_iou_thr=0.3,
+                    pos_iou_thr=0.7,
+                    type='MaxIoUAssigner'),
+                debug=False,
+                pos_weight=-1,
+                sampler=dict(
+                    add_gt_as_proposals=False,
+                    neg_pos_ub=-1,
+                    num=256,
+                    pos_fraction=0.5,
+                    type='RandomSampler')),
+            type='RPNHead'),
+        type='EUPGHead'),
+    test_cfg=None,
+    train_cfg=None,
+    type='UWSAM')
+optim_wrapper = dict(
+    accumulative_counts=4,
+    clip_grad=dict(max_norm=1.0, norm_type=2),
+    loss_scale=dict(
+        backoff_factor=0.5,
+        growth_factor=2.0,
+        growth_interval=2000,
+        init_scale=512.0),
+    optimizer=dict(eps=1e-08, lr=0.0001, type='AdamW', weight_decay=0.05),
+    type='AmpOptimWrapper')
+param_scheduler = [
+    dict(
+        begin=0, by_epoch=False, end=500, start_factor=0.001, type='LinearLR'),
+    dict(
+        begin=0,
+        by_epoch=True,
+        end=24,
+        gamma=0.1,
+        milestones=[
+            16,
+            22,
+        ],
+        type='MultiStepLR'),
+]
+resume = False
+sam_checkpoint = '/home/cuongvt/My-works/code/UWSAM/sam_vit_h_4b8939.pth'
+test_cfg = dict(type='TestLoop')
+test_dataloader = dict(
+    batch_size=1,
+    dataset=dict(
+        ann_file='annotations/multiclass_test.json',
+        data_prefix=dict(img='img/'),
+        data_root='/home/cuongvt/My-works/code/UWSAM/UIIS10K/UIIS10K/',
+        metainfo=dict(
+            classes=(
+                'fish',
+                'reptiles',
+                'arthropoda',
+                'corals',
+                'mollusk',
+                'plants',
+                'ruins',
+                'garbage',
+                'human',
+                'robots',
+            )),
+        pipeline=[
+            dict(type='LoadImageFromFile'),
+            dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
+            dict(keep_ratio=False, scale=(
+                1024,
+                1024,
+            ), type='Resize'),
+            dict(type='PackDetInputs'),
+        ],
+        type='CocoDataset'),
+    num_workers=0,
+    persistent_workers=False,
+    sampler=dict(shuffle=False, type='DefaultSampler'))
+test_evaluator = dict(
+    ann_file=
+    '/home/cuongvt/My-works/code/UWSAM/UIIS10K/UIIS10K/annotations/multiclass_test.json',
+    format_only=False,
+    metric=[
+        'bbox',
+        'segm',
+    ],
+    type='CocoMetric')
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
+    dict(keep_ratio=False, scale=(
+        1024,
+        1024,
+    ), type='Resize'),
+    dict(type='PackDetInputs'),
+]
+train_cfg = dict(max_epochs=24, type='EpochBasedTrainLoop', val_interval=24)
+train_dataloader = dict(
+    batch_size=1,
+    dataset=dict(
+        ann_file='annotations/multiclass_train.json',
+        data_prefix=dict(img='img/'),
+        data_root='/home/cuongvt/My-works/code/UWSAM/UIIS10K/UIIS10K/',
+        filter_cfg=dict(filter_empty_gt=True, min_size=32),
+        metainfo=dict(
+            classes=(
+                'fish',
+                'reptiles',
+                'arthropoda',
+                'corals',
+                'mollusk',
+                'plants',
+                'ruins',
+                'garbage',
+                'human',
+                'robots',
+            )),
+        pipeline=[
+            dict(type='LoadImageFromFile'),
+            dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
+            dict(
+                keep_ratio=True,
+                ratio_range=(
+                    0.8,
+                    1.2,
+                ),
+                scale=(
+                    1024,
+                    1024,
+                ),
+                type='RandomResize'),
+            dict(
+                allow_negative_crop=False,
+                crop_size=(
+                    1024,
+                    1024,
+                ),
+                type='RandomCrop'),
+            dict(keep_ratio=False, scale=(
+                1024,
+                1024,
+            ), type='Resize'),
+            dict(prob=0.5, type='RandomFlip'),
+            dict(type='PackDetInputs'),
+        ],
+        type='CocoDataset'),
+    num_workers=0,
+    persistent_workers=False,
+    sampler=dict(shuffle=True, type='DefaultSampler'))
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
+    dict(
+        keep_ratio=True,
+        ratio_range=(
+            0.8,
+            1.2,
+        ),
+        scale=(
+            1024,
+            1024,
+        ),
+        type='RandomResize'),
+    dict(
+        allow_negative_crop=False, crop_size=(
+            1024,
+            1024,
+        ), type='RandomCrop'),
+    dict(keep_ratio=False, scale=(
+        1024,
+        1024,
+    ), type='Resize'),
+    dict(prob=0.5, type='RandomFlip'),
+    dict(type='PackDetInputs'),
+]
+val_cfg = dict(type='ValLoop')
+val_dataloader = dict(
+    batch_size=1,
+    dataset=dict(
+        ann_file='annotations/multiclass_test.json',
+        data_prefix=dict(img='img/'),
+        data_root='/home/cuongvt/My-works/code/UWSAM/UIIS10K/UIIS10K/',
+        metainfo=dict(
+            classes=(
+                'fish',
+                'reptiles',
+                'arthropoda',
+                'corals',
+                'mollusk',
+                'plants',
+                'ruins',
+                'garbage',
+                'human',
+                'robots',
+            )),
+        pipeline=[
+            dict(type='LoadImageFromFile'),
+            dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
+            dict(keep_ratio=False, scale=(
+                1024,
+                1024,
+            ), type='Resize'),
+            dict(type='PackDetInputs'),
+        ],
+        type='CocoDataset'),
+    num_workers=0,
+    persistent_workers=False,
+    sampler=dict(shuffle=False, type='DefaultSampler'))
+val_evaluator = dict(
+    ann_file=
+    '/home/cuongvt/My-works/code/UWSAM/UIIS10K/UIIS10K/annotations/multiclass_test.json',
+    format_only=False,
+    metric=[
+        'bbox',
+        'segm',
+    ],
+    type='CocoMetric')
+vis_backends = [
+    dict(type='LocalVisBackend'),
+]
+visualizer = dict(
+    name='visualizer',
+    type='DetLocalVisualizer',
+    vis_backends=[
+        dict(type='LocalVisBackend'),
+    ])
+work_dir = './work_dirs/uwsam_teacher_paper'