mmdeploy/tests/test_codebase/test_mmdet/data/detr_model.json

130 lines
3.2 KiB
JSON

{
"type": "DETR",
"num_queries": 100,
"data_preprocessor": {
"type": "DetDataPreprocessor",
"mean": [123.675, 116.28, 103.53],
"std": [58.395, 57.12, 57.375],
"bgr_to_rgb": true,
"pad_size_divisor": 1
},
"backbone": {
"type": "ResNet",
"depth": 50,
"num_stages": 4,
"out_indices": [3],
"frozen_stages": 1,
"norm_cfg": {
"type": "BN",
"requires_grad": false
},
"norm_eval": true,
"style": "pytorch",
"init_cfg": {
"type": "Pretrained",
"checkpoint": "torchvision://resnet50"
}
},
"neck": {
"type": "ChannelMapper",
"in_channels": [2048],
"kernel_size": 1,
"out_channels": 256,
"num_outs": 1
},
"encoder": {
"num_layers": 6,
"layer_cfg": {
"self_attn_cfg": {
"embed_dims": 256,
"num_heads": 8,
"dropout": 0.1,
"batch_first": true
},
"ffn_cfg": {
"embed_dims": 256,
"feedforward_channels": 2048,
"num_fcs": 2,
"ffn_drop": 0.1,
"act_cfg": {
"type": "ReLU",
"inplace": true
}
}
}
},
"decoder": {
"num_layers": 6,
"layer_cfg": {
"self_attn_cfg": {
"embed_dims": 256,
"num_heads": 8,
"dropout": 0.1,
"batch_first": true
},
"cross_attn_cfg": {
"embed_dims": 256,
"num_heads": 8,
"dropout": 0.1,
"batch_first": true
},
"ffn_cfg": {
"embed_dims": 256,
"feedforward_channels": 2048,
"num_fcs": 2,
"ffn_drop": 0.1,
"act_cfg": {
"type": "ReLU",
"inplace": true
}
}
},
"return_intermediate": true
},
"positional_encoding": {
"num_feats": 128,
"normalize": true
},
"bbox_head": {
"type": "DETRHead",
"num_classes": 80,
"embed_dims": 256,
"loss_cls": {
"type": "CrossEntropyLoss",
"bg_cls_weight": 0.1,
"use_sigmoid": false,
"loss_weight": 1.0,
"class_weight": 1.0
},
"loss_bbox": {
"type": "L1Loss",
"loss_weight": 5.0
},
"loss_iou": {
"type": "GIoULoss",
"loss_weight": 2.0
}
},
"train_cfg": {
"assigner": {
"type":
"HungarianAssigner",
"match_costs": [{
"type": "ClassificationCost",
"weight": 1.0
}, {
"type": "BBoxL1Cost",
"weight": 5.0,
"box_format": "xywh"
}, {
"type": "IoUCost",
"iou_mode": "giou",
"weight": 2.0
}]
}
},
"test_cfg": {
"max_per_img": 100
}
}