Error when prune rec model: __init__() missing 1 required positional argument: 'out_channels_list'

[yearzhengcheng]

请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem

• 系统环境/System Environment：
• win11 22000.978
• 版本号/Version：Paddle：2.3.1 PaddleOCR：2.5.0.3 问题相关组件/Related components：
• 运行指令/Command Code：python deploy/slim/prune/sensitivity_anal.py -c project/rec_model/config.yml
• 完整报错/Complete Error Message： [2022/10/11 11:14:09] ppocr INFO: Architecture : [2022/10/11 11:14:09] ppocr INFO: Backbone : [2022/10/11 11:14:09] ppocr INFO: last_conv_stride : [1, 2] [2022/10/11 11:14:09] ppocr INFO: last_pool_type : avg [2022/10/11 11:14:09] ppocr INFO: name : MobileNetV1Enhance [2022/10/11 11:14:09] ppocr INFO: scale : 0.5 [2022/10/11 11:14:09] ppocr INFO: Head : [2022/10/11 11:14:09] ppocr INFO: head_list : [2022/10/11 11:14:09] ppocr INFO: CTCHead : [2022/10/11 11:14:09] ppocr INFO: Head : [2022/10/11 11:14:09] ppocr INFO: fc_decay : 1e-05 [2022/10/11 11:14:09] ppocr INFO: Neck : [2022/10/11 11:14:09] ppocr INFO: depth : 2 [2022/10/11 11:14:09] ppocr INFO: dims : 64 [2022/10/11 11:14:09] ppocr INFO: hidden_dims : 120 [2022/10/11 11:14:09] ppocr INFO: name : svtr [2022/10/11 11:14:09] ppocr INFO: use_guide : True [2022/10/11 11:14:09] ppocr INFO: SARHead : [2022/10/11 11:14:09] ppocr INFO: enc_dim : 512 [2022/10/11 11:14:09] ppocr INFO: max_text_length : 25 [2022/10/11 11:14:09] ppocr INFO: name : MultiHead [2022/10/11 11:14:09] ppocr INFO: Transform : None [2022/10/11 11:14:09] ppocr INFO: algorithm : SVTR [2022/10/11 11:14:09] ppocr INFO: model_type : rec [2022/10/11 11:14:09] ppocr INFO: Eval : [2022/10/11 11:14:09] ppocr INFO: dataset : [2022/10/11 11:14:09] ppocr INFO: data_dir : ./project [2022/10/11 11:14:09] ppocr INFO: label_file_list : ['./project/train_data/rec/val.txt'] [2022/10/11 11:14:09] ppocr INFO: name : SimpleDataSet [2022/10/11 11:14:09] ppocr INFO: transforms : [2022/10/11 11:14:09] ppocr INFO: DecodeImage : [2022/10/11 11:14:09] ppocr INFO: channel_first : False [2022/10/11 11:14:09] ppocr INFO: img_mode : BGR [2022/10/11 11:14:09] ppocr INFO: MultiLabelEncode : None [2022/10/11 11:14:09] ppocr INFO: RecResizeImg : [2022/10/11 11:14:09] ppocr INFO: image_shape : [3, 48, 160] [2022/10/11 11:14:09] ppocr INFO: KeepKeys : [2022/10/11 11:14:09] ppocr INFO: keep_keys : ['image', 'label_ctc', 'label_sar', 'length', 'valid_ratio'] [2022/10/11 11:14:09] ppocr INFO: loader : [2022/10/11 11:14:09] ppocr INFO: batch_size_per_card : 32 [2022/10/11 11:14:09] ppocr INFO: drop_last : False [2022/10/11 11:14:09] ppocr INFO: num_workers : 4 [2022/10/11 11:14:09] ppocr INFO: shuffle : False [2022/10/11 11:14:09] ppocr INFO: Global : [2022/10/11 11:14:09] ppocr INFO: cal_metric_during_train : True [2022/10/11 11:14:09] ppocr INFO: character_dict_path : project/dict.txt [2022/10/11 11:14:09] ppocr INFO: checkpoints : project/rec_model/best_accuracy [2022/10/11 11:14:09] ppocr INFO: debug : True [2022/10/11 11:14:09] ppocr INFO: distributed : False [2022/10/11 11:14:09] ppocr INFO: epoch_num : 500 [2022/10/11 11:14:09] ppocr INFO: eval_batch_step : [0, 2000] [2022/10/11 11:14:09] ppocr INFO: infer_img : project/train_data/rec/train/12626277.bmp [2022/10/11 11:14:09] ppocr INFO: infer_mode : False [2022/10/11 11:14:09] ppocr INFO: log_smooth_window : 20 [2022/10/11 11:14:09] ppocr INFO: max_text_length : 25 [2022/10/11 11:14:09] ppocr INFO: pretrained_model : project/rec_model/best_accuracy [2022/10/11 11:14:09] ppocr INFO: print_batch_step : 10 [2022/10/11 11:14:09] ppocr INFO: save_epoch_step : 100 [2022/10/11 11:14:09] ppocr INFO: save_inference_dir : project/rec_model/inference [2022/10/11 11:14:09] ppocr INFO: save_model_dir : ./project/rec_model/prune_model/ [2022/10/11 11:14:09] ppocr INFO: save_res_path : ./project/output/rec1/predicts_ppocrv3_en.txt [2022/10/11 11:14:09] ppocr INFO: use_gpu : False [2022/10/11 11:14:09] ppocr INFO: use_space_char : True [2022/10/11 11:14:09] ppocr INFO: use_visualdl : False [2022/10/11 11:14:09] ppocr INFO: Loss : [2022/10/11 11:14:09] ppocr INFO: loss_config_list : [2022/10/11 11:14:09] ppocr INFO: CTCLoss : None [2022/10/11 11:14:09] ppocr INFO: SARLoss : None [2022/10/11 11:14:09] ppocr INFO: name : MultiLoss [2022/10/11 11:14:09] ppocr INFO: Metric : [2022/10/11 11:14:09] ppocr INFO: ignore_space : False [2022/10/11 11:14:09] ppocr INFO: main_indicator : acc [2022/10/11 11:14:09] ppocr INFO: name : RecMetric [2022/10/11 11:14:09] ppocr INFO: Optimizer : [2022/10/11 11:14:09] ppocr INFO: beta1 : 0.9 [2022/10/11 11:14:09] ppocr INFO: beta2 : 0.999 [2022/10/11 11:14:09] ppocr INFO: lr : [2022/10/11 11:14:09] ppocr INFO: learning_rate : 0.001 [2022/10/11 11:14:09] ppocr INFO: name : Cosine [2022/10/11 11:14:09] ppocr INFO: warmup_epoch : 5 [2022/10/11 11:14:09] ppocr INFO: name : Adam [2022/10/11 11:14:09] ppocr INFO: regularizer : [2022/10/11 11:14:09] ppocr INFO: factor : 3e-05 [2022/10/11 11:14:09] ppocr INFO: name : L2 [2022/10/11 11:14:09] ppocr INFO: PostProcess : [2022/10/11 11:14:09] ppocr INFO: name : CTCLabelDecode [2022/10/11 11:14:09] ppocr INFO: Train : [2022/10/11 11:14:09] ppocr INFO: dataset : [2022/10/11 11:14:09] ppocr INFO: data_dir : ./project [2022/10/11 11:14:09] ppocr INFO: ext_op_transform_idx : 1 [2022/10/11 11:14:09] ppocr INFO: label_file_list : ['./project/train_data/rec/train.txt'] [2022/10/11 11:14:09] ppocr INFO: name : SimpleDataSet [2022/10/11 11:14:09] ppocr INFO: transforms : [2022/10/11 11:14:09] ppocr INFO: DecodeImage : [2022/10/11 11:14:09] ppocr INFO: channel_first : False [2022/10/11 11:14:09] ppocr INFO: img_mode : BGR [2022/10/11 11:14:09] ppocr INFO: RecConAug : [2022/10/11 11:14:09] ppocr INFO: ext_data_num : 2 [2022/10/11 11:14:09] ppocr INFO: image_shape : [48, 160, 3] [2022/10/11 11:14:09] ppocr INFO: prob : 0.5 [2022/10/11 11:14:09] ppocr INFO: RecAug : None [2022/10/11 11:14:09] ppocr INFO: MultiLabelEncode : None [2022/10/11 11:14:09] ppocr INFO: RecResizeImg : [2022/10/11 11:14:09] ppocr INFO: image_shape : [3, 48, 160] [2022/10/11 11:14:09] ppocr INFO: KeepKeys : [2022/10/11 11:14:09] ppocr INFO: keep_keys : ['image', 'label_ctc', 'label_sar', 'length', 'valid_ratio'] [2022/10/11 11:14:09] ppocr INFO: loader : [2022/10/11 11:14:09] ppocr INFO: batch_size_per_card : 24 [2022/10/11 11:14:09] ppocr INFO: drop_last : True [2022/10/11 11:14:09] ppocr INFO: num_workers : 4 [2022/10/11 11:14:09] ppocr INFO: shuffle : True [2022/10/11 11:14:09] ppocr INFO: profiler_options : None [2022/10/11 11:14:09] ppocr INFO: train with paddle 2.3.1 and device Place(cpu) [2022/10/11 11:14:09] ppocr INFO: Initialize indexs of datasets:['./project/train_data/rec/train.txt'] [2022/10/11 11:14:09] ppocr INFO: Initialize indexs of datasets:['./project/train_data/rec/val.txt'] Traceback (most recent call last): File "deploy/slim/prune/sensitivity_anal.py", line 175, in main(config, device, logger, vdl_writer) File "deploy/slim/prune/sensitivity_anal.py", line 75, in main model = build_model(config['Architecture']) File "E:\WorkPower\Project\px\PaddleOCR\deploy/slim/prune......\ppocr\modeling\architectures_init_.py", line 27, in build_model arch = BaseModel(config) File "E:\WorkPower\Project\px\PaddleOCR\deploy/slim/prune......\ppocr\modeling\architectures\base_model.py", line 71, in init self.head = build_head(config["Head"]) File "E:\WorkPower\Project\px\PaddleOCR\deploy/slim/prune......\ppocr\modeling\heads_init_.py", line 57, in build_head module_class = eval(module_name)(**config) TypeError: init() missing 1 required positional argument: 'out_channels_list'