kernel, bias = self._get_kernel_bias() File "/dataset/zwq/work/opensource/PaddleOCR/ppocr/modeling/backbones/rec_lcnetv3.py", line 221, in _get_kernel_bias kernel_conv_1x1, bias_conv_1x1 = self._fuse_bn_tensor(self.conv_1x1) File "/dataset/zwq/work/opensource/PaddleOCR/ppocr/modelin...
建议去除GTC策略，单独使用 LCNet_SVTR 进行训练。可以参考这个issue修改配置文件：#6355 ...
acc: 0.339844, norm_edit_dis: 0.936109, loss: 13.633232, avg_reader_cost: 0.63327 s, avg...
algorithm:SVTR_LCNet Transform: Backbone: name:MobileNetV1Enhance scale:0.5 last_conv_stride:[1,2] last_pool_type:avg last_pool_kernel_size:[2,2] Head: name:MultiHead head_list: -CTCHead: Neck: name:svtr dims:64 depth:2 hidden_dims:120 ...
[2024/05/22 14:06:59] ppocr INFO: algorithm : SVTR [2024/05/22 14:06:59] ppocr INFO: freeze_params : False [2024/05/22 14:06:59] ppocr INFO: model_type : rec [2024/05/22 14:06:59] ppocr INFO: pretrained : None
det_pse_box_thresh=0.85, det_pse_min_area=16, det_pse_scale=1, scales=[8, 16, 32], alpha=1.0, beta=1.0, fourier_degree=5, rec_algorithm='SVTR_LCNet', rec_model_dir='/root/.paddleocr/whl/rec/ch/ch_PP-OCRv4_rec_server_infer', rec_image_inverse=True, rec_image_shape='3, ...