Merge branch 'dev' of https://github.com/WenmuZhou/PytorchOCR into dev

# Conflicts: # readme.md
whutbd · Aug 30, 2023 · 4a507ac · 4a507ac
2 parents 34e66c7 + 1fdba01
commit 4a507ac
Show file tree

Hide file tree

Showing 44 changed files with 749 additions and 1,571 deletions.
diff --git a/configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml b/configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml
@@ -3,7 +3,7 @@ Global:
   epoch_num: 500
   log_smooth_window: 20
   print_batch_step: 10
-  output_dir: ./output/ch_PP-OCR_V3_det/
+  output_dir: ./output/det/ch_PP-OCR_V3_det_cml/
   eval_epoch_step: [0, 1]
   cal_metric_during_train: false
   pretrained_model:

diff --git a/configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_dml.yml b/configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_dml.yml
@@ -3,7 +3,7 @@ Global:
   epoch_num: 1200
   log_smooth_window: 20
   print_batch_step: 10
-  output_dir: ./output/ch_PP-OCR_V3_det/
+  output_dir: ./output/det/ch_PP-OCR_V3_det_dml/
   eval_epoch_step: [0, 1]
   cal_metric_during_train: false
   pretrained_model:

diff --git a/configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_student.yml b/configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_student.yml
@@ -3,7 +3,7 @@ Global:
   epoch_num: 500
   log_smooth_window: 20
   print_batch_step: 10
-  output_dir: ./output/ch_PP-OCR_V3_det/
+  output_dir: ./output/det/ch_PP-OCR_V3_det/
   eval_epoch_step: [0, 1]
   cal_metric_during_train: false
   pretrained_model:

diff --git a/configs/det/ch_PP-OCRv4/ch_PP-OCRv4_det_cml.yml b/configs/det/ch_PP-OCRv4/ch_PP-OCRv4_det_cml.yml
@@ -3,7 +3,7 @@ Global:
   epoch_num: &epoch_num 500
   log_smooth_window: 20
   print_batch_step: 10
-  output_dir: ./output/ch_PP-OCRv4_cml
+  output_dir: ./output/det/ch_PP-OCRv4_cml
   eval_epoch_step: [0, 1]
   cal_metric_during_train: false
   pretrained_model:

diff --git a/configs/det/ch_PP-OCRv4/ch_PP-OCRv4_det_student.yml b/configs/det/ch_PP-OCRv4/ch_PP-OCRv4_det_student.yml
@@ -3,7 +3,7 @@ Global:
   epoch_num: &epoch_num 500
   log_smooth_window: 20
   print_batch_step: 10
-  output_dir: ./output/ch_PP-OCRv4/
+  output_dir: ./output/det/ch_PP-OCRv4/
   eval_epoch_step: [0, 1]
   cal_metric_during_train: false
   pretrained_model:

diff --git a/configs/det/ch_PP-OCRv4/ch_PP-OCRv4_det_teacher.yml b/configs/det/ch_PP-OCRv4/ch_PP-OCRv4_det_teacher.yml
@@ -3,7 +3,7 @@ Global:
   epoch_num: &epoch_num 500
   log_smooth_window: 20
   print_batch_step: 10
-  output_dir: ./output/ch_PP-OCRv4/
+  output_dir: ./output/det/h_PP-OCRv4_hgnet/
   eval_epoch_step: [0, 1]
   cal_metric_during_train: false
   pretrained_model:

diff --git a/configs/rec/PP-OCRv3/ch_PP-OCRv3_rec.yml b/configs/rec/PP-OCRv3/ch_PP-OCRv3_rec.yml
@@ -3,16 +3,16 @@ Global:
   epoch_num: 500
   log_smooth_window: 20
   print_batch_step: 10
-  output_dir: ./output/rec_ppocr_v3
+  output_dir: ./output/rec/rec_ppocr_v3
   eval_epoch_step: [0, 1]
   cal_metric_during_train: true
   pretrained_model:
   checkpoints:
   use_tensorboard: false
+  infer_mode: false
   infer_img: doc/imgs_words/ch/word_1.jpg
   character_dict_path: &character_dict_path ppocr/utils/ppocr_keys_v1.txt
   max_text_length: &max_text_length 25
-  infer_mode: false
   use_space_char: &use_space_char true
 
 Export:
@@ -49,8 +49,6 @@ Architecture:
             depth: 2
             hidden_dims: 120
             use_guide: True
-          Head:
-            fc_decay: 0.00001
       - SARHead:
           enc_dim: 512
           max_text_length: *max_text_length

diff --git a/configs/rec/PP-OCRv3/ch_PP-OCRv3_rec_distillation.yml b/configs/rec/PP-OCRv3/ch_PP-OCRv3_rec_distillation.yml
@@ -3,16 +3,16 @@ Global:
   epoch_num: 800
   log_smooth_window: 20
   print_batch_step: 10
-  output_dir: ./output/rec_ppocr_v3_distillation
+  output_dir: ./output/rec/rec_ppocr_v3_distillation
   eval_epoch_step: [0, 1]
   cal_metric_during_train: true
   pretrained_model:
   checkpoints:
   use_tensorboard: false
+  infer_mode: false
   infer_img: doc/imgs_words/ch/word_1.jpg
   character_dict_path: &character_dict_path ppocr/utils/ppocr_keys_v1.txt
   max_text_length: &max_text_length 25
-  infer_mode: false
   use_space_char: &use_space_char true
 
 Export:
@@ -59,8 +59,6 @@ Architecture:
                 depth: 2
                 hidden_dims: 120
                 use_guide: True
-              Head:
-                fc_decay: 0.00001
           - SARHead:
               enc_dim: 512
               max_text_length: *max_text_length

diff --git a/configs/rec/PP-OCRv3/multi_language/.gitkeep b/configs/rec/PP-OCRv3/multi_language/.gitkeep
diff --git a/configs/rec/PP-OCRv3/multi_language/arabic_PP-OCRv3_rec.yml b/configs/rec/PP-OCRv3/multi_language/arabic_PP-OCRv3_rec.yml
diff --git a/configs/rec/PP-OCRv3/multi_language/chinese_cht_PP-OCRv3_rec.yml b/configs/rec/PP-OCRv3/multi_language/chinese_cht_PP-OCRv3_rec.yml