diff --git a/configs/det/ch_PP-OCRv2/ch_PP-OCR_det_cml.yml b/configs/det/ch_PP-OCRv2/ch_PP-OCR_det_cml.yml index 0f08909add17d8c73ad6e1b00e17d4c351def7e5..ab484a44833a405513d7f2b4079a4da4c2e403c8 100644 --- a/configs/det/ch_PP-OCRv2/ch_PP-OCR_det_cml.yml +++ b/configs/det/ch_PP-OCRv2/ch_PP-OCR_det_cml.yml @@ -141,6 +141,7 @@ Train: img_mode: BGR channel_first: False - DetLabelEncode: # Class handling label + - CopyPaste: - IaaAugment: augmenter_args: - { 'type': Fliplr, 'args': { 'p': 0.5 } } diff --git a/configs/det/ch_PP-OCRv2/ch_PP-OCR_det_distill.yml b/configs/det/ch_PP-OCRv2/ch_PP-OCR_det_distill.yml index 1159d71bf94c330e26c3009b38c5c2b4a9c96f52..46daeeb86d004772a6fb964d602369dcd53b3a01 100644 --- a/configs/det/ch_PP-OCRv2/ch_PP-OCR_det_distill.yml +++ b/configs/det/ch_PP-OCRv2/ch_PP-OCR_det_distill.yml @@ -68,8 +68,7 @@ Loss: ohem_ratio: 3 - DistillationDBLoss: weight: 1.0 - model_name_list: ["Student", "Teacher"] - # key: maps + model_name_list: ["Student"] name: DBLoss balance_loss: true main_loss_type: DiceLoss @@ -116,6 +115,7 @@ Train: img_mode: BGR channel_first: False - DetLabelEncode: # Class handling label + - CopyPaste: - IaaAugment: augmenter_args: - { 'type': Fliplr, 'args': { 'p': 0.5 } } diff --git a/configs/det/ch_PP-OCRv2/ch_PP-OCR_det_dml.yml b/configs/det/ch_PP-OCRv2/ch_PP-OCR_det_dml.yml index 7fe2d2e1a065b54d0e2479475f5f67ac5e38a166..bfbc3b6268cf521acb035be33ced9141046fc430 100644 --- a/configs/det/ch_PP-OCRv2/ch_PP-OCR_det_dml.yml +++ b/configs/det/ch_PP-OCRv2/ch_PP-OCR_det_dml.yml @@ -118,6 +118,7 @@ Train: img_mode: BGR channel_first: False - DetLabelEncode: # Class handling label + - CopyPaste: - IaaAugment: augmenter_args: - { 'type': Fliplr, 'args': { 'p': 0.5 } } diff --git a/configs/e2e/e2e_r50_vd_pg.yml b/configs/e2e/e2e_r50_vd_pg.yml index 4a6e19f4461c7236f3a9a5253437eff97fa72f67..c4c5226e796a42db723ce78ef65473e357c25dc6 100644 --- a/configs/e2e/e2e_r50_vd_pg.yml +++ b/configs/e2e/e2e_r50_vd_pg.yml @@ -94,7 +94,7 @@ Eval: label_file_list: [./train_data/total_text/test/test.txt] transforms: - DecodeImage: # load image - img_mode: RGB + img_mode: BGR channel_first: False - E2ELabelEncodeTest: - E2EResizeForTest: @@ -111,4 +111,4 @@ Eval: shuffle: False drop_last: False batch_size_per_card: 1 # must be 1 - num_workers: 2 \ No newline at end of file + num_workers: 2 diff --git a/ppocr/modeling/necks/rnn.py b/ppocr/modeling/necks/rnn.py index de87b3d9895168657f8c9722177c026b992c2966..86e649028f8fbb76cb5a1fd85381bd361277c6ee 100644 --- a/ppocr/modeling/necks/rnn.py +++ b/ppocr/modeling/necks/rnn.py @@ -51,7 +51,7 @@ class EncoderWithFC(nn.Layer): super(EncoderWithFC, self).__init__() self.out_channels = hidden_size weight_attr, bias_attr = get_para_bias_attr( - l2_decay=0.00001, k=in_channels, name='reduce_encoder_fea') + l2_decay=0.00001, k=in_channels) self.fc = nn.Linear( in_channels, hidden_size, diff --git a/tools/infer/predict_e2e.py b/tools/infer/predict_e2e.py index 8ff279d7437965f725082a9eb1c83e05a7ffc8a8..5029d6059346a00062418d8d1b6cb029b0110643 100755 --- a/tools/infer/predict_e2e.py +++ b/tools/infer/predict_e2e.py @@ -141,7 +141,6 @@ if __name__ == "__main__": img, flag = check_and_read_gif(image_file) if not flag: img = cv2.imread(image_file) - img = img[:, :, ::-1] if img is None: logger.info("error in loading image:{}".format(image_file)) continue