openocr-python 0.0.9__py3-none-any.whl → 0.1.0.dev0__py3-none-any.whl

openocr/configs/rec/mdiff4str/svtrv2_mdiffdecoder_small.yml

@@ -0,0 +1,152 @@
+Global:
+  device: gpu
+  epoch_num: 40
+  log_smooth_window: 20
+  print_batch_step: 10
+  output_dir: ./output/rec/u14m_filter/svtrv2_mdiffdecoder_small/
+  save_epoch_step: [30, 1]
+  # evaluation is run every 2000 iterations
+  eval_batch_step: [0, 500]
+  eval_epoch_step: [0, 1]
+  cal_metric_during_train: False
+  pretrained_model:
+  checkpoints:
+  use_tensorboard: false
+  infer_img:
+  # for data or label process
+  character_dict_path: &character_dict_path ./tools/utils/EN_symbol_dict.txt # 96en
+  # ./tools/utils/ppocr_keys_v1.txt  # ch
+  max_text_length: &max_text_length 25
+  use_space_char: &use_space_char False
+  save_res_path: ./output/rec/u14m_filter/predicts_svtrv2_mdiffdecoder_small.txt
+  use_amp: True
+  grad_clip_val: 20.0
+
+Optimizer:
+  name: AdamW
+  lr: 0.0005 # for 8gpus bs128/gpu
+  weight_decay: 0.05
+  filter_bias_and_bn: True
+
+LRScheduler:
+  name: OneCycleLR
+  warmup_epoch: 1.5 # pct_start 0.075*20 = 1.5ep
+  cycle_momentum: False
+
+Architecture:
+  model_type: rec
+  algorithm: MDiff4STR
+  in_channels: 3
+  Transform:
+  Encoder:
+    name: SVTRv2LNConvTwo33
+    use_pos_embed: False
+    dims: [128, 256, 384]
+    depths: [3, 6, 3]
+    num_heads: [4, 8, 12]
+    mixer: [['Conv','Conv','Conv'],['Conv','Conv','Conv','FGlobal','Global','Global'],['Global','Global','Global']]
+    local_k: [[5, 5], [5, 5], [-1, -1]]
+    sub_k: [[1, 1], [2, 1], [-1, -1]]
+    last_stage: false
+    feat2d: False
+  Decoder:
+    name: MDiffDecoder
+    num_decoder_layers: 3
+    nhead: 6
+    max_len: *max_text_length
+    parallel_decoding: False
+    autoregressive_decoding: False
+    low_confidence_decoding: False
+    random_mask_decoding: False
+    semi_autoregressive_decoding: False
+    cloze_mask_decoding: False
+    sampler_step: 3
+    sample_k: &sample_k 3
+    temperature: 1.0
+
+Loss:
+  name: MDiffLoss
+
+PostProcess:
+  name: ARLabelDecode
+  character_dict_path: *character_dict_path
+  use_space_char: *use_space_char
+
+Metric:
+  name: RecMetric
+  main_indicator: acc
+  is_filter: True
+
+Train:
+  dataset:
+    name: RatioDataSetTVResize
+    ds_width: True
+    padding: false
+    data_dir_list: ['../Union14M-L-LMDB-Filtered/filter_train_challenging',
+    '../Union14M-L-LMDB-Filtered/filter_train_hard',
+    '../Union14M-L-LMDB-Filtered/filter_train_medium',
+    '../Union14M-L-LMDB-Filtered/filter_train_normal',
+    '../Union14M-L-LMDB-Filtered/filter_train_easy',
+    ]
+    transforms:
+      - DecodeImagePIL: # load image
+          img_mode: RGB
+      - PARSeqAugPIL:
+      - MDiffLabelEncode: # Class handling label
+          character_dict_path: *character_dict_path
+          use_space_char: *use_space_char
+          max_text_length: *max_text_length
+          sample_num: *sample_k
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'reflect_ids', 'noisy_batch', 'masked_indices', 'p_mask', 'length'] # dataloader will return list in this order
+  sampler:
+    name: RatioSampler
+    scales: [[128, 32]] # w, h
+    # divide_factor: to ensure the width and height dimensions can be devided by downsampling multiple
+    first_bs: &bs 256
+    fix_bs: false
+    divided_factor: [4, 16] # w, h
+    is_training: True
+  loader:
+    shuffle: True
+    batch_size_per_card: *bs
+    drop_last: True
+    max_ratio: &max_ratio 4
+    num_workers: 4
+
+Eval:
+  dataset:
+    name: RatioDataSetTVResize
+    ds_width: True
+    padding: False
+    data_dir_list: [
+      '../evaluation/CUTE80',
+      '../evaluation/IC13_857',
+      '../evaluation/IC15_1811',
+      '../evaluation/IIIT5k_3000',
+      '../evaluation/SVT',
+      '../evaluation/SVTP',
+      ]
+    transforms:
+      - DecodeImagePIL: # load image
+          img_mode: RGB
+      - ARLabelEncode: # Class handling label
+          character_dict_path: *character_dict_path
+          use_space_char: *use_space_char
+          max_text_length: *max_text_length
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
+  sampler:
+    name: RatioSampler
+    scales: [[128, 32]] # w, h
+    # divide_factor: to ensure the width and height dimensions can be devided by downsampling multiple
+    first_bs: *bs
+    fix_bs: false
+    divided_factor: [4, 16] # w, h
+    is_training: False
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: *bs
+    max_ratio: *max_ratio
+    num_workers: 4

openocr/configs/rec/unirec/focalsvtr_ardecoder_unirec.yml

@@ -0,0 +1,114 @@
+Global:
+  device: gpu
+  epoch_num: 10
+  log_smooth_window: 20
+  print_batch_step: 20
+  output_dir: ./output/rec/unirec-0.1b/
+  save_epoch_step: [0, 1] # save every 1 epoch after 7 epochs
+  save_iter_step: [0, 2000]
+  # evaluation is run every 2000 iterations
+  eval_batch_step: [2000000, 4000]
+  eval_epoch_step: [150, 1]
+  cal_metric_during_train: False
+  pretrained_model: ./unirec-0.1b/model.pth
+  checkpoints:
+  resume_from_iter: False
+  use_tensorboard: false
+  infer_img: ../crop_img_hand
+  # for data or label process
+  character_dict_path: &character_dict_path ./tools/utils/EN_symbol_dict.txt # 96en
+  # ./tools/utils/ppocr_keys_v1.txt  # ch
+  max_text_length: &max_text_length 2048
+  use_space_char: &use_space_char False
+  save_res_path: ./output/rec/unirec-0.1b/predicts_unirec-0.1b.txt
+  use_amp: True
+  use_ema: False
+  use_transformers: True
+  grad_clip_val: 1.0
+  vlm_ocr_config: &vlm_ocr_config ./configs/rec/unirec/unirec-0.1b
+
+Optimizer:
+  name: AdamW
+  lr: 0.0001 # for 4gpus bs256/gpu
+  weight_decay: 0.01
+  filter_bias_and_bn: True
+
+LRScheduler:
+  name: OneCycleLR
+  warmup_epoch: 0.3 # pct_start 0.075*20 = 1.5ep
+  cycle_momentum: False
+
+Architecture:
+  model_type: rec
+  algorithm: UniRec
+  in_channels: 3
+  Transform:
+  Encoder:
+  Decoder:
+    out_channels: -1 # for inference, set to -1
+
+Loss:
+  name: UniRecLoss
+
+PostProcess:
+  name: UniRecLabelDecode
+  lower: False
+  tokenizer_path: *vlm_ocr_config
+
+Metric:
+  name: RecMetric
+  main_indicator: acc
+  is_filter: False
+
+Train:
+  dataset:
+    name: NaSizeDataSet
+    divided_factor: &divided_factor [64, 64] # w, h
+    max_side: &max_side [960, 1408] # [64*30, 64*44] # w, h [960, 1408] #
+    root_path: path/to/UniRec40M
+    add_return: True
+    zoom_min_factor: 4
+    use_zoom: True
+    # all_data: True
+    test_data: True
+    use_aug: false
+    use_linedata: False
+    transforms:
+      - UniRecLabelEncode: # Class handling label
+          max_text_length: *max_text_length
+          vlmocr: True
+          tokenizer_path: *vlm_ocr_config # path to tokenizer, e.g. 'vocab.json', 'merges.txt'
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
+  sampler:
+    name: NaSizeSampler
+    # divide_factor: to ensure the width and height dimensions can be devided by downsampling multiple
+    min_bs: 1
+    max_bs: 24
+  loader:
+    shuffle: True
+    batch_size_per_card: 64
+    drop_last: True
+    num_workers: 8
+
+Eval:
+  dataset:
+    name: LMDBDataSet
+    data_dir: ../evaluation
+    transforms:
+      - DecodeImagePIL: # load image
+          img_mode: RGB
+      - UniRecLabelEncode: # Class handling label
+          max_text_length: *max_text_length
+          vlmocr: True
+          tokenizer_path: *vlm_ocr_config # path to tokenizer, e.g. 'vocab.json', 'merges.txt'
+      - NaSizeResize:
+          max_side: *max_side
+          divided_factor: *divided_factor
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 1
+    num_workers: 1

openocr/configs/rec/unirec/opendoc_pipeline.yml

@@ -0,0 +1,105 @@
+
+pipeline_name: PaddleOCR-VL
+
+batch_size: 64
+
+use_queues: False
+
+use_doc_preprocessor: False
+use_layout_detection: True
+use_chart_recognition: False
+format_block_content: False
+merge_layout_blocks: True
+markdown_ignore_labels:
+  - number
+  - footnote
+  - header
+  - header_image
+  - footer
+  - footer_image
+  - aside_text
+
+SubModules:
+  LayoutDetection:
+    module_name: layout_detection
+    model_name: PP-DocLayoutV2
+    model_dir: null
+    batch_size: 8
+    threshold:
+      0: 0.5 # abstract
+      1: 0.5 # algorithm
+      2: 0.5 # aside_text
+      3: 0.5 # chart
+      4: 0.5 # content
+      5: 0.4 # formula
+      6: 0.4 # doc_title
+      7: 0.5 # figure_title
+      8: 0.5 # footer
+      9: 0.5 # footer
+      10: 0.5 # footnote
+      11: 0.5 # formula_number
+      12: 0.5 # header
+      13: 0.5 # header
+      14: 0.5 # image
+      15: 0.4 # formula
+      16: 0.5 # number
+      17: 0.4 # paragraph_title
+      18: 0.5 # reference
+      19: 0.5 # reference_content
+      20: 0.45 # seal
+      21: 0.5 # table
+      22: 0.4 # text
+      23: 0.4 # text
+      24: 0.5 # vision_footnote
+    layout_nms: True
+    layout_unclip_ratio: [1.0, 1.0]
+    layout_merge_bboxes_mode:
+      0: "union" # abstract
+      1: "union" # algorithm
+      2: "union" # aside_text
+      3: "large" # chart
+      4: "union" # content
+      5: "large" # display_formula
+      6: "large" # doc_title
+      7: "union" # figure_title
+      8: "union" # footer
+      9: "union" # footer
+      10: "union" # footnote
+      11: "union" # formula_number
+      12: "union" # header
+      13: "union" # header
+      14: "union" # image
+      15: "large" # inline_formula
+      16: "union" # number
+      17: "large" # paragraph_title
+      18: "union" # reference
+      19: "union" # reference_content
+      20: "union" # seal
+      21: "union" # table
+      22: "union" # text
+      23: "union" # text
+      24: "union" # vision_footnote
+  VLRecognition:
+    module_name: vl_recognition
+    model_name: PaddleOCR-VL-0.9B
+    model_dir: null
+    batch_size: 4096
+    genai_config:
+      backend: native
+
+SubPipelines:
+  DocPreprocessor:
+    pipeline_name: doc_preprocessor
+    batch_size: 8
+    use_doc_orientation_classify: True
+    use_doc_unwarping: True
+    SubModules:
+      DocOrientationClassify:
+        module_name: doc_text_orientation
+        model_name: PP-LCNet_x1_0_doc_ori
+        model_dir: null
+        batch_size: 8
+      DocUnwarping:
+        module_name: image_unwarping
+        model_name: UVDoc
+        model_dir: null

openocr/demo_gradio.py

@@ -9,11 +9,12 @@ import numpy as np
 import json
 import time
 from PIL import Image
-from tools.infer_e2e import OpenOCR, check_and_download_font, draw_ocr_box_txt
+from tools.infer_e2e import OpenOCRE2E, check_and_download_font, draw_ocr_box_txt
+from tools.download_example_images import get_example_images_path
 
 
 def initialize_ocr(model_type, drop_score):
-    return OpenOCR(mode=model_type, drop_score=drop_score)
+    return OpenOCRE2E(mode=model_type, drop_score=drop_score, backend='onnx')
 
 
 # Default model type
@@ -108,12 +109,25 @@ def find_file_in_current_dir_and_subdirs(file_name):
             return relative_path
 
 
-e2e_img_example = list_image_paths('./OCR_e2e_img')
+# Get example images path and download if necessary
+example_img_dir = get_example_images_path(demo_type='ocr')
+e2e_img_example = list_image_paths(example_img_dir)
 
-if __name__ == '__main__':
+
+def launch_demo(share=False, server_port=7860, server_name='0.0.0.0'):
+    """Launch OpenOCR Gradio demo with default configuration.
+
+    Args:
+        share: Whether to create a public share link (default: False)
+        server_port: Server port (default: 7860)
+        server_name: Server name (default: '0.0.0.0')
+
+    Returns:
+        gr.Blocks: Gradio demo instance
+    """
     css = '.image-container img { width: 100%; max-height: 320px;}'
 
-    with gr.Blocks(css=css) as demo:
+    with gr.Blocks(css=css, theme=gr.themes.Soft()) as demo:
         gr.HTML("""
                 <h1 style='text-align: center;'><a href="https://github.com/Topdu/OpenOCR">OpenOCR</a></h1>
                 <p style='text-align: center;'>准确高效的通用 OCR 系统 （由<a href="https://fvl.fudan.edu.cn">FVL实验室</a> <a href="https://github.com/Topdu/OpenOCR">OCR Team</a> 创建） <a href="https://github.com/Topdu/OpenOCR/tree/main?tab=readme-ov-file#quick-start">[本地快速部署]</a></p>"""
@@ -126,7 +140,7 @@ if __name__ == '__main__':
                 examples = gr.Examples(examples=e2e_img_example,
                                        inputs=input_image,
                                        label='Examples')
-                downstream = gr.Button('Run')
+                downstream = gr.Button('🚀 运行识别', variant='primary')
 
                 # 添加参数调节组件
                 with gr.Column():
@@ -139,7 +153,8 @@ if __name__ == '__main__':
                             ['slow', 'fast'],
                             value='slow',
                             label='Detection Score Mode',
-                            info='文本框的置信度计算模式，默认为 slow。slow 模式计算速度较慢，但准确度较高。fast 模式计算速度较快，但准确度较低。'
+                            info=
+                            '文本框的置信度计算模式，默认为 slow。slow 模式计算速度较慢，但准确度较高。fast 模式计算速度较快，但准确度较低。'
                         )
                     with gr.Row():
                         rec_drop_score_slider = gr.Slider(
@@ -204,4 +219,9 @@ if __name__ == '__main__':
                                  img_mask,
                              ])
 
-    demo.launch(share=True)
+    demo.launch(share=share, server_port=server_port, server_name=server_name)
+    return demo
+
+
+if __name__ == '__main__':
+    launch_demo(share=False)