| ReduceOnPlateauLR_patience = 50 |
| albu_train_transforms = [ |
| dict( |
| always_apply=True, |
| border_mode=0, |
| min_height=512, |
| min_width=512, |
| type='PadIfNeeded'), |
| dict(always_apply=True, type='Flip'), |
| dict( |
| always_apply=True, interpolation=4, limit=( |
| -180, |
| 180, |
| ), type='Rotate'), |
| dict( |
| p=0.5, |
| transforms=[ |
| dict( |
| alpha=20, |
| approximate=True, |
| border_mode=0, |
| interpolation=4, |
| mask_value=( |
| 0, |
| 0, |
| 0, |
| ), |
| p=0.5, |
| same_dxdy=True, |
| sigma=15, |
| type='ElasticTransform'), |
| dict( |
| alpha=40, |
| approximate=True, |
| border_mode=0, |
| interpolation=4, |
| mask_value=( |
| 0, |
| 0, |
| 0, |
| ), |
| p=0.5, |
| same_dxdy=False, |
| sigma=15, |
| type='ElasticTransform'), |
| ], |
| type='OneOf'), |
| dict( |
| always_apply=True, |
| brightness=0.2, |
| contrast=0.1, |
| hue=0.2, |
| saturation=0.2, |
| type='ColorJitter'), |
| dict(p=0.5, type='AdvancedBlur'), |
| dict(always_apply=True, height=512, type='CenterCrop', width=512), |
| ] |
| auto_scale_lr = dict(base_batch_size=16, enable=False) |
| batch_size = 8 |
| class_weight = [ |
| 1.0, |
| 1.0, |
| 1.0, |
| 1.0, |
| 1.0, |
| 1.0, |
| 0.1, |
| ] |
| classes = ( |
| 'background', |
| 'stomatal complex', |
| 'stoma', |
| 'outer ledge', |
| 'pore', |
| 'pavement cell', |
| ) |
| crop_size = ( |
| 512, |
| 512, |
| ) |
| data_root = 'StomataPy400K_filtered_train/' |
| dataset_type = 'StomataDataset' |
| default_hooks = dict( |
| checkpoint=dict( |
| by_epoch=True, |
| interval=999999, |
| save_best='mIoU', |
| save_last=True, |
| type='CheckpointHook'), |
| early_stopping=dict( |
| monitor='mIoU', patience=150, rule='greater', |
| type='EarlyStoppingHook'), |
| logger=dict(interval=600, log_metric_by_epoch=True, type='LoggerHook'), |
| param_scheduler=dict(type='ParamSchedulerHook'), |
| sampler_seed=dict(type='DistSamplerSeedHook'), |
| timer=dict(type='IterTimerHook'), |
| visualization=dict(draw=True, interval=50, type='SegVisualizationHook')) |
| default_scope = 'mmseg' |
| dinov2_checkpoint = 'train/checkpoints/dinov2_converted.pth' |
| early_stopping_patience = 150 |
| env_cfg = dict( |
| cudnn_benchmark=True, |
| dist_cfg=dict(backend='nccl'), |
| mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) |
| find_unused_parameters = True |
| fp16 = dict(loss_scale='dynamic') |
| launcher = 'pytorch' |
| load_from = None |
| log_level = 'INFO' |
| log_processor = dict(by_epoch=True) |
| lr = 1e-05 |
| max_epochs = 300 |
| model = dict( |
| backbone=dict( |
| block_chunks=0, |
| depth=24, |
| embed_dim=1024, |
| ffn_bias=True, |
| ffn_layer='mlp', |
| img_size=512, |
| init_cfg=dict( |
| checkpoint='train/checkpoints/dinov2_converted.pth', |
| type='Pretrained'), |
| init_values=1e-05, |
| mlp_ratio=4, |
| num_heads=16, |
| patch_size=16, |
| proj_bias=True, |
| qkv_bias=True, |
| reins_config=dict( |
| embed_dims=1024, |
| link_token_to_query=True, |
| lora_dim=16, |
| num_layers=24, |
| patch_size=16, |
| token_length=100, |
| type='LoRAReins'), |
| type='ReinsDinoVisionTransformer'), |
| data_preprocessor=dict( |
| bgr_to_rgb=True, |
| mean=[ |
| 123.675, |
| 116.28, |
| 103.53, |
| ], |
| pad_val=0, |
| seg_pad_val=255, |
| size=( |
| 512, |
| 512, |
| ), |
| std=[ |
| 58.395, |
| 57.12, |
| 57.375, |
| ], |
| type='SegDataPreProcessor'), |
| decode_head=dict( |
| align_corners=False, |
| enforce_decoder_input_project=False, |
| feat_channels=256, |
| in_channels=[ |
| 1024, |
| 1024, |
| 1024, |
| 1024, |
| ], |
| loss_cls=dict( |
| class_weight=[ |
| 1.0, |
| 1.0, |
| 1.0, |
| 1.0, |
| 1.0, |
| 1.0, |
| 0.1, |
| ], |
| loss_weight=2.0, |
| reduction='mean', |
| type='mmdet.CrossEntropyLoss', |
| use_sigmoid=False), |
| loss_dice=dict( |
| activate=True, |
| eps=1.0, |
| loss_weight=5.0, |
| naive_dice=True, |
| reduction='mean', |
| type='mmdet.DiceLoss', |
| use_sigmoid=True), |
| loss_mask=dict( |
| loss_weight=5.0, |
| reduction='mean', |
| type='mmdet.CrossEntropyLoss', |
| use_sigmoid=True), |
| num_classes=6, |
| num_queries=100, |
| num_transformer_feat_level=3, |
| out_channels=256, |
| pixel_decoder=dict( |
| act_cfg=dict(type='ReLU'), |
| encoder=dict( |
| init_cfg=None, |
| layer_cfg=dict( |
| ffn_cfg=dict( |
| act_cfg=dict(inplace=True, type='ReLU'), |
| embed_dims=256, |
| feedforward_channels=1024, |
| ffn_drop=0.0, |
| num_fcs=2), |
| self_attn_cfg=dict( |
| batch_first=True, |
| dropout=0.0, |
| embed_dims=256, |
| im2col_step=64, |
| init_cfg=None, |
| norm_cfg=None, |
| num_heads=8, |
| num_levels=3, |
| num_points=4)), |
| num_layers=6), |
| init_cfg=None, |
| norm_cfg=dict(num_groups=32, type='GN'), |
| num_outs=3, |
| positional_encoding=dict(normalize=True, num_feats=128), |
| type='mmdet.MSDeformAttnPixelDecoder'), |
| positional_encoding=dict(normalize=True, num_feats=128), |
| replace_query_feat=True, |
| strides=[ |
| 4, |
| 8, |
| 16, |
| 32, |
| ], |
| train_cfg=dict( |
| assigner=dict( |
| match_costs=[ |
| dict(type='mmdet.ClassificationCost', weight=2.0), |
| dict( |
| type='mmdet.CrossEntropyLossCost', |
| use_sigmoid=True, |
| weight=5.0), |
| dict( |
| eps=1.0, |
| pred_act=True, |
| type='mmdet.DiceCost', |
| weight=5.0), |
| ], |
| type='mmdet.HungarianAssigner'), |
| importance_sample_ratio=0.75, |
| num_points=12544, |
| oversample_ratio=3.0, |
| sampler=dict(type='mmdet.MaskPseudoSampler')), |
| transformer_decoder=dict( |
| init_cfg=None, |
| layer_cfg=dict( |
| cross_attn_cfg=dict( |
| attn_drop=0.0, |
| batch_first=True, |
| dropout_layer=None, |
| embed_dims=256, |
| num_heads=8, |
| proj_drop=0.0), |
| ffn_cfg=dict( |
| act_cfg=dict(inplace=True, type='ReLU'), |
| add_identity=True, |
| dropout_layer=None, |
| embed_dims=256, |
| feedforward_channels=2048, |
| ffn_drop=0.0, |
| num_fcs=2), |
| self_attn_cfg=dict( |
| attn_drop=0.0, |
| batch_first=True, |
| dropout_layer=None, |
| embed_dims=256, |
| num_heads=8, |
| proj_drop=0.0)), |
| num_layers=9, |
| return_intermediate=True), |
| type='ReinMask2FormerHead'), |
| test_cfg=dict(crop_size=( |
| 512, |
| 512, |
| ), mode='slide', stride=( |
| 1364, |
| 1364, |
| )), |
| train_cfg=dict(), |
| type='EncoderDecoder') |
| model_crop_size = ( |
| 512, |
| 512, |
| ) |
| n_gpus = 4 |
| num_classes = 6 |
| num_workers = 16 |
| optim_wrapper = dict( |
| constructor='PEFTOptimWrapperConstructor', |
| optimizer=dict( |
| betas=( |
| 0.9, |
| 0.999, |
| ), |
| eps=1e-08, |
| lr=1e-05, |
| type='AdamW', |
| weight_decay=0.05), |
| paramwise_cfg=dict( |
| custom_keys=dict({ |
| 'learnable_tokens': dict(decay_mult=0.0, lr_mult=1.0), |
| 'level_embed': dict(decay_mult=0.0, lr_mult=1.0), |
| 'norm': dict(decay_mult=0.0), |
| 'query_embed': dict(decay_mult=0.0, lr_mult=1.0), |
| 'reins.scale': dict(decay_mult=0.0, lr_mult=1.0) |
| }), |
| norm_decay_mult=0.0)) |
| optimizer_config = dict( |
| cumulative_iters=8, type='GradientCumulativeOptimizerHook') |
| original_batch_size = 4 |
| original_lr = 0.0001 |
| original_n_gpus = 8 |
| output_dir = 'StomataPy400K_aperture_512' |
| param_scheduler = [ |
| dict( |
| begin=0, |
| by_epoch=True, |
| convert_to_iter_based=True, |
| end=30, |
| end_factor=1.0, |
| start_factor=0.001, |
| type='LinearLR', |
| verbose=False), |
| dict( |
| T_max=270, |
| begin=30, |
| by_epoch=True, |
| convert_to_iter_based=True, |
| end=300, |
| eta_min=1.0000000000000002e-10, |
| eta_min_ratio=None, |
| type='CosineAnnealingLR', |
| verbose=False), |
| dict( |
| by_epoch=True, |
| factor=0.75, |
| monitor='mIoU', |
| patience=50, |
| rule='greater', |
| type='ReduceOnPlateauLR', |
| verbose=False), |
| ] |
| randomness = dict(deterministic=False, seed=42) |
| resume = False |
| test_cfg = dict(type='ValLoop') |
| test_dataloader = dict( |
| batch_size=1, |
| dataset=dict( |
| ann_file='splits//val.txt', |
| data_prefix=dict(img_path='images', seg_map_path='labels'), |
| data_root='StomataPy400K_filtered_train/', |
| pipeline=[ |
| dict(type='LoadImageFromFile'), |
| dict(prob=0.5, type='RandomFlip'), |
| dict(keep_ratio=False, scale=( |
| 512, |
| 512, |
| ), type='Resize'), |
| dict(reduce_zero_label=False, type='LoadAnnotations'), |
| dict(type='PackSegInputs'), |
| ], |
| type='StomataDataset'), |
| num_workers=16) |
| test_evaluator = dict( |
| iou_metrics=[ |
| 'mIoU', |
| ], type='IoUMetric') |
| test_pipeline = [ |
| dict(type='LoadImageFromFile'), |
| dict(prob=0.5, type='RandomFlip'), |
| dict(keep_ratio=False, scale=( |
| 512, |
| 512, |
| ), type='Resize'), |
| dict(reduce_zero_label=False, type='LoadAnnotations'), |
| dict(type='PackSegInputs'), |
| ] |
| train_cfg = dict(max_epochs=300, type='EpochBasedTrainLoop', val_interval=1) |
| train_dataloader = dict( |
| batch_size=8, |
| dataset=dict( |
| ann_file='splits//train.txt', |
| data_prefix=dict(img_path='images', seg_map_path='labels'), |
| data_root='StomataPy400K_filtered_train/', |
| pipeline=[ |
| dict(type='LoadImageFromFile'), |
| dict(reduce_zero_label=False, type='LoadAnnotations'), |
| dict( |
| max_size=2048, |
| resize_type='ResizeShortestEdge', |
| scales=[ |
| 358, |
| 409, |
| 460, |
| 512, |
| 563, |
| 614, |
| 665, |
| 716, |
| ], |
| type='RandomChoiceResize'), |
| dict( |
| cat_max_ratio=0.75, crop_size=( |
| 512, |
| 512, |
| ), type='RandomCrop'), |
| dict( |
| keymap=dict(gt_seg_map='mask', img='image'), |
| transforms=[ |
| dict( |
| always_apply=True, |
| border_mode=0, |
| min_height=512, |
| min_width=512, |
| type='PadIfNeeded'), |
| dict(always_apply=True, type='Flip'), |
| dict( |
| always_apply=True, |
| interpolation=4, |
| limit=( |
| -180, |
| 180, |
| ), |
| type='Rotate'), |
| dict( |
| p=0.5, |
| transforms=[ |
| dict( |
| alpha=20, |
| approximate=True, |
| border_mode=0, |
| interpolation=4, |
| mask_value=( |
| 0, |
| 0, |
| 0, |
| ), |
| p=0.5, |
| same_dxdy=True, |
| sigma=15, |
| type='ElasticTransform'), |
| dict( |
| alpha=40, |
| approximate=True, |
| border_mode=0, |
| interpolation=4, |
| mask_value=( |
| 0, |
| 0, |
| 0, |
| ), |
| p=0.5, |
| same_dxdy=False, |
| sigma=15, |
| type='ElasticTransform'), |
| ], |
| type='OneOf'), |
| dict( |
| always_apply=True, |
| brightness=0.2, |
| contrast=0.1, |
| hue=0.2, |
| saturation=0.2, |
| type='ColorJitter'), |
| dict(p=0.5, type='AdvancedBlur'), |
| dict( |
| always_apply=True, |
| height=512, |
| type='CenterCrop', |
| width=512), |
| ], |
| type='Albu'), |
| dict( |
| cutout_ratio=( |
| 0.02, |
| 0.05, |
| ), |
| n_holes=10, |
| prob=0.5, |
| type='RandomCutOut'), |
| dict( |
| interpolation='lanczos', |
| keep_ratio=True, |
| scale=( |
| 512, |
| 512, |
| ), |
| type='Resize'), |
| dict( |
| meta_keys=( |
| 'img_path', |
| 'img_shape', |
| 'img', |
| 'gt_seg_map', |
| ), |
| type='PackSegInputs'), |
| ], |
| type='StomataDataset'), |
| num_workers=16) |
| train_pipeline = [ |
| dict(type='LoadImageFromFile'), |
| dict(reduce_zero_label=False, type='LoadAnnotations'), |
| dict( |
| max_size=2048, |
| resize_type='ResizeShortestEdge', |
| scales=[ |
| 358, |
| 409, |
| 460, |
| 512, |
| 563, |
| 614, |
| 665, |
| 716, |
| ], |
| type='RandomChoiceResize'), |
| dict(cat_max_ratio=0.75, crop_size=( |
| 512, |
| 512, |
| ), type='RandomCrop'), |
| dict( |
| keymap=dict(gt_seg_map='mask', img='image'), |
| transforms=[ |
| dict( |
| always_apply=True, |
| border_mode=0, |
| min_height=512, |
| min_width=512, |
| type='PadIfNeeded'), |
| dict(always_apply=True, type='Flip'), |
| dict( |
| always_apply=True, |
| interpolation=4, |
| limit=( |
| -180, |
| 180, |
| ), |
| type='Rotate'), |
| dict( |
| p=0.5, |
| transforms=[ |
| dict( |
| alpha=20, |
| approximate=True, |
| border_mode=0, |
| interpolation=4, |
| mask_value=( |
| 0, |
| 0, |
| 0, |
| ), |
| p=0.5, |
| same_dxdy=True, |
| sigma=15, |
| type='ElasticTransform'), |
| dict( |
| alpha=40, |
| approximate=True, |
| border_mode=0, |
| interpolation=4, |
| mask_value=( |
| 0, |
| 0, |
| 0, |
| ), |
| p=0.5, |
| same_dxdy=False, |
| sigma=15, |
| type='ElasticTransform'), |
| ], |
| type='OneOf'), |
| dict( |
| always_apply=True, |
| brightness=0.2, |
| contrast=0.1, |
| hue=0.2, |
| saturation=0.2, |
| type='ColorJitter'), |
| dict(p=0.5, type='AdvancedBlur'), |
| dict(always_apply=True, height=512, type='CenterCrop', width=512), |
| ], |
| type='Albu'), |
| dict( |
| cutout_ratio=( |
| 0.02, |
| 0.05, |
| ), n_holes=10, prob=0.5, type='RandomCutOut'), |
| dict( |
| interpolation='lanczos', |
| keep_ratio=True, |
| scale=( |
| 512, |
| 512, |
| ), |
| type='Resize'), |
| dict( |
| meta_keys=( |
| 'img_path', |
| 'img_shape', |
| 'img', |
| 'gt_seg_map', |
| ), |
| type='PackSegInputs'), |
| ] |
| tta_model = dict(type='SegTTAModel') |
| val_cfg = dict(type='ValLoop') |
| val_dataloader = dict( |
| batch_size=1, |
| dataset=dict( |
| ann_file='splits//val.txt', |
| data_prefix=dict(img_path='images', seg_map_path='labels'), |
| data_root='StomataPy400K_filtered_train/', |
| pipeline=[ |
| dict(type='LoadImageFromFile'), |
| dict(prob=0.5, type='RandomFlip'), |
| dict(keep_ratio=False, scale=( |
| 512, |
| 512, |
| ), type='Resize'), |
| dict(reduce_zero_label=False, type='LoadAnnotations'), |
| dict(type='PackSegInputs'), |
| ], |
| type='StomataDataset'), |
| num_workers=16) |
| val_evaluator = dict( |
| iou_metrics=[ |
| 'mIoU', |
| ], type='IoUMetric') |
| val_interval = 1 |
| visualizer = dict( |
| name='visualizer', |
| type='SegLocalVisualizer', |
| vis_backends=[ |
| dict(type='LocalVisBackend'), |
| dict( |
| init_kwargs=dict( |
| name='StomataPy400K_aperture_512', project='StomataPy'), |
| type='WandbVisBackend'), |
| ]) |
| wandb_project = 'StomataPy' |
| warmup_epochs = 30 |
| work_dir = '../Models/StomataPy400K_aperture_512' |
|
|
