Error about load pnfs-mobilenet-v2_nfx3_384x288_coco_ckpt.tar for testing

[wmy10128]

python test.py --cfg configs/mobilenet_v2_CNFx3_coco_384x288.yaml --exp_name o/mobile_cnf3_coco/ --test_model weights/pnfs-mobilenet-v2_nfx3_384x288_coco_ckpt.tar --flip_test the error is： RuntimeError: Error(s) in loading state_dict for Body_Part_Representation: Missing key(s) in state_dict: "total_ops", "total_params", "backbone.total_ops", "backbone.total_params", "backbone.features.total_ops", "backbone.features.total_para ms", "backbone.features.1.total_ops", "backbone.features.1.total_params".......

The mobilenet_v2_CNFx3_coco_384x288.yaml file is: model: use_pretrained: false pretrained: '' use_backbone: True frozen_resnet: False frozen_mobilenet: False backbone_net_name: 'mobilenet_v2' # or 'meta_arch' or 'mobilenet_v2' #backbone_layers: 50 #resnet backbone_feature_num: 4 # resnet backbone_pretrained_path: './weights/mobilenet_v2.pth.tar' init_weights: True dataset_name: 'coco' keypoints_num: 17 input_size: w: 288 h: 384 heatmap_size: w: 72 h: 96 margin_to_border: 1.25 # >=1

#Only when using an extra mask module branch, NOT USED IN THE PAPER
use_mask: false
mask_num: 0

backbone:
    frozen_backbone: false #
    search_alpha: True
    search_beta: True
    operators: ["Zero", "skip_connect", "Sep_Conv_3x3", "Atr_Conv_3x3","avg_pool_3x3","max_pool_3x3"]

    reserve_layers_num: 4  # only use first 4 layers
    depth: 12
    size_types: [4,8,16,32]
    hidden_states_num: 1
    factor: 8
    input_nodes_num: 1

subnetwork_config:
    dataset_name: 'coco'
    parts_num : 3
    cell_config:
        vector_in_pixel : True
        vector_dim: 8
        convolution_mode: '2D'
        one-shot-search: True
        search_alpha: true
        search_beta: true
        operators: [ "skip_connect", "Sep_Conv_3x3", "Atr_Conv_3x3",]
        depth: 6
        cut_layers_num: 3  # first several layers
        size_types: [4,8,16,32]
        hidden_states_num: 1
        factor: 16
        input_nodes_num: 1

Thanks!