magic-pdf 1.1.0__py3-none-any.whl → 1.2.0__py3-none-any.whl

magic_pdf/pdf_parse_union_core_v2.py

@@ -6,8 +6,10 @@ import statistics
 import time
 from typing import List
 
+import cv2
 import fitz
 import torch
+import numpy as np
 from loguru import logger
 
 from magic_pdf.config.enums import SupportedPdfParseMethod
@@ -127,16 +129,15 @@ def fill_char_in_spans(spans, all_chars):
                 span['chars'].append(char)
                 break
 
-    empty_spans = []
-
+    need_ocr_spans = []
     for span in spans:
         chars_to_content(span)
         # 有的span中虽然没有字但有一两个空的占位符，用宽高和content长度过滤
         if len(span['content']) * span['height'] < span['width'] * 0.5:
             # logger.info(f"maybe empty span: {len(span['content'])}, {span['height']}, {span['width']}")
-            empty_spans.append(span)
+            need_ocr_spans.append(span)
         del span['height'], span['width']
-    return empty_spans
+    return need_ocr_spans
 
 
 # 使用鲁棒性更强的中心点坐标判断
@@ -190,6 +191,31 @@ def remove_tilted_line(text_blocks):
             block['lines'].remove(line)
 
 
+def calculate_contrast(img, img_mode) -> float:
+    """
+    计算给定图像的对比度。
+    :param img: 图像，类型为numpy.ndarray
+    :Param img_mode = 图像的色彩通道，'rgb' 或 'bgr'
+    :return: 图像的对比度值
+    """
+    if img_mode == 'rgb':
+        # 将RGB图像转换为灰度图
+        gray_img = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
+    elif img_mode == 'bgr':
+        # 将BGR图像转换为灰度图
+        gray_img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
+    else:
+        raise ValueError("Invalid image mode. Please provide 'rgb' or 'bgr'.")
+
+    # 计算均值和标准差
+    mean_value = np.mean(gray_img)
+    std_dev = np.std(gray_img)
+    # 对比度定义为标准差除以平均值（加上小常数避免除零错误）
+    contrast = std_dev / (mean_value + 1e-6)
+    # logger.info(f"contrast: {contrast}")
+    return round(contrast, 2)
+
+
 def txt_spans_extract_v2(pdf_page, spans, all_bboxes, all_discarded_blocks, lang):
     # cid用0xfffd表示，连字符拆开
     # text_blocks_raw = pdf_page.get_text('rawdict', flags=fitz.TEXT_PRESERVE_WHITESPACE | fitz.TEXT_MEDIABOX_CLIP)['blocks']
@@ -274,9 +300,9 @@ def txt_spans_extract_v2(pdf_page, spans, all_bboxes, all_discarded_blocks, lang
             span['chars'] = []
             new_spans.append(span)
 
-    empty_spans = fill_char_in_spans(new_spans, all_pymu_chars)
+    need_ocr_spans = fill_char_in_spans(new_spans, all_pymu_chars)
 
-    if len(empty_spans) > 0:
+    if len(need_ocr_spans) > 0:
 
         # 初始化ocr模型
         atom_model_manager = AtomModelSingleton()
@@ -287,9 +313,15 @@ def txt_spans_extract_v2(pdf_page, spans, all_bboxes, all_discarded_blocks, lang
             lang=lang
         )
 
-        for span in empty_spans:
+        for span in need_ocr_spans:
             # 对span的bbox截图再ocr
             span_img = cut_image_to_pil_image(span['bbox'], pdf_page, mode='cv2')
+
+            # 计算span的对比度，低于0.20的span不进行ocr
+            if calculate_contrast(span_img, img_mode='bgr') <= 0.20:
+                spans.remove(span)
+                continue
+
             ocr_res = ocr_model.ocr(span_img, det=False)
             if ocr_res and len(ocr_res) > 0:
                 if len(ocr_res[0]) > 0:
@@ -306,24 +338,7 @@ def txt_spans_extract_v2(pdf_page, spans, all_bboxes, all_discarded_blocks, lang
 
 def model_init(model_name: str):
     from transformers import LayoutLMv3ForTokenClassification
-    device = get_device()
-    if torch.cuda.is_available():
-        device = torch.device('cuda')
-        if torch.cuda.is_bf16_supported():
-            supports_bfloat16 = True
-        else:
-            supports_bfloat16 = False
-    elif str(device).startswith("npu"):
-        import torch_npu
-        if torch_npu.npu.is_available():
-            device = torch.device('npu')
-            supports_bfloat16 = False
-        else:
-            device = torch.device('cpu')
-            supports_bfloat16 = False
-    else:
-        device = torch.device('cpu')
-        supports_bfloat16 = False
+    device = torch.device(get_device())
 
     if model_name == 'layoutreader':
         # 检测modelscope的缓存目录是否存在
@@ -339,9 +354,6 @@ def model_init(model_name: str):
             model = LayoutLMv3ForTokenClassification.from_pretrained(
                 'hantian/layoutreader'
             )
-        # 检查设备是否支持 bfloat16
-        if supports_bfloat16:
-            model.bfloat16()
         model.to(device).eval()
     else:
         logger.error('model name not allow')
@@ -404,10 +416,11 @@ def cal_block_index(fix_blocks, sorted_bboxes):
             block_bboxes.append(block['bbox'])
 
             # 删除图表body block中的虚拟line信息, 并用real_lines信息回填
-            if block['type'] in [BlockType.ImageBody, BlockType.TableBody]:
-                block['virtual_lines'] = copy.deepcopy(block['lines'])
-                block['lines'] = copy.deepcopy(block['real_lines'])
-                del block['real_lines']
+            if block['type'] in [BlockType.ImageBody, BlockType.TableBody, BlockType.Title, BlockType.InterlineEquation]:
+                if 'real_lines' in block:
+                    block['virtual_lines'] = copy.deepcopy(block['lines'])
+                    block['lines'] = copy.deepcopy(block['real_lines'])
+                    del block['real_lines']
 
         import numpy as np
 

magic_pdf/post_proc/llm_aided.py

@@ -3,6 +3,7 @@ import json
 from loguru import logger
 from magic_pdf.dict2md.ocr_mkcontent import merge_para_with_text
 from openai import OpenAI
+import ast
 
 
 #@todo: 有的公式以"\"结尾，这样会导致尾部拼接的"$"被转义，也需要修复
@@ -119,11 +120,12 @@ def llm_aided_title(pdf_info_dict, title_aided_config):
     - 在完成初步分级后，仔细检查分级结果的合理性
     - 根据上下文关系和逻辑顺序，对不合理的分级进行微调
     - 确保最终的分级结果符合文档的实际结构和逻辑
+    - 字典中可能包含被误当成标题的正文，你可以通过将其层级标记为 0 来排除它们
     
 IMPORTANT: 
-请直接返回优化过的由标题层级组成的json，格式如下：
-{{"0":1,"1":2,"2":2,"3":3}}
-返回的json不需要格式化。
+请直接返回优化过的由标题层级组成的字典，格式为{{标题id:标题层级}}，如下：
+{{0:1,1:2,2:2,3:3}}
+不需要对字典格式化，不需要返回任何其他信息。
 
 Input title list:
 {title_dict}
@@ -133,7 +135,7 @@ Corrected title list:
 
     retry_count = 0
     max_retries = 3
-    json_completion = None
+    dict_completion = None
 
     while retry_count < max_retries:
         try:
@@ -143,24 +145,20 @@ Corrected title list:
                     {'role': 'user', 'content': title_optimize_prompt}],
                 temperature=0.7,
             )
-            json_completion = json.loads(completion.choices[0].message.content)
+            # logger.info(f"Title completion: {completion.choices[0].message.content}")
+            dict_completion = ast.literal_eval(completion.choices[0].message.content)
+            # logger.info(f"len(dict_completion): {len(dict_completion)}, len(title_dict): {len(title_dict)}")
 
-            # logger.info(f"Title completion: {json_completion}")
-            # logger.info(f"len(json_completion): {len(json_completion)}, len(title_dict): {len(title_dict)}")
-
-            if len(json_completion) == len(title_dict):
+            if len(dict_completion) == len(title_dict):
                 for i, origin_title_block in enumerate(origin_title_list):
-                    origin_title_block["level"] = int(json_completion[str(i)])
+                    origin_title_block["level"] = int(dict_completion[i])
                 break
             else:
                 logger.warning("The number of titles in the optimized result is not equal to the number of titles in the input.")
                 retry_count += 1
         except Exception as e:
-            if isinstance(e, json.decoder.JSONDecodeError):
-                logger.warning(f"JSON decode error on attempt {retry_count + 1}: {e}")
-            else:
-                logger.exception(e)
+            logger.exception(e)
             retry_count += 1
 
-    if json_completion is None:
-        logger.error("Failed to decode JSON after maximum retries.")
+    if dict_completion is None:
+        logger.error("Failed to decode dict after maximum retries.")

magic_pdf/pre_proc/ocr_dict_merge.py

@@ -60,6 +60,19 @@ def merge_spans_to_line(spans, threshold=0.6):
         return lines
 
 
+def span_block_type_compatible(span_type, block_type):
+    if span_type in [ContentType.Text, ContentType.InlineEquation]:
+        return block_type in [BlockType.Text, BlockType.Title, BlockType.ImageCaption, BlockType.ImageFootnote, BlockType.TableCaption, BlockType.TableFootnote]
+    elif span_type == ContentType.InterlineEquation:
+        return block_type in [BlockType.InterlineEquation]
+    elif span_type == ContentType.Image:
+        return block_type in [BlockType.ImageBody]
+    elif span_type == ContentType.Table:
+        return block_type in [BlockType.TableBody]
+    else:
+        return False
+
+
 def fill_spans_in_blocks(blocks, spans, radio):
     """将allspans中的span按位置关系，放入blocks中."""
     block_with_spans = []
@@ -78,8 +91,7 @@ def fill_spans_in_blocks(blocks, spans, radio):
         block_spans = []
         for span in spans:
             span_bbox = span['bbox']
-            if calculate_overlap_area_in_bbox1_area_ratio(
-                    span_bbox, block_bbox) > radio:
+            if calculate_overlap_area_in_bbox1_area_ratio(span_bbox, block_bbox) > radio and span_block_type_compatible(span['type'], block_type):
                 block_spans.append(span)
 
         block_dict['spans'] = block_spans

{magic_pdf-1.1.0.dist-info → magic_pdf-1.2.0.dist-info}/METADATA

@@ -1,49 +1,49 @@
 Metadata-Version: 2.1
 Name: magic-pdf
-Version: 1.1.0
+Version: 1.2.0
 Summary: A practical tool for converting PDF to Markdown
 Home-page: https://github.com/opendatalab/MinerU
 Requires-Python: >=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE.md
-Requires-Dist: boto3 >=1.28.43
-Requires-Dist: Brotli >=1.1.0
-Requires-Dist: click >=8.1.7
-Requires-Dist: fast-langdetect >=0.2.3
-Requires-Dist: loguru >=0.6.0
-Requires-Dist: numpy <2.0.0,>=1.21.6
-Requires-Dist: pydantic >=2.7.2
-Requires-Dist: PyMuPDF <=1.24.14,>=1.24.9
-Requires-Dist: scikit-learn >=1.0.2
-Requires-Dist: torch >=2.2.2
+Requires-Dist: boto3>=1.28.43
+Requires-Dist: Brotli>=1.1.0
+Requires-Dist: click>=8.1.7
+Requires-Dist: fast-langdetect>=0.2.3
+Requires-Dist: loguru>=0.6.0
+Requires-Dist: numpy<2.0.0,>=1.21.6
+Requires-Dist: pydantic>=2.7.2
+Requires-Dist: PyMuPDF<=1.24.14,>=1.24.9
+Requires-Dist: scikit-learn>=1.0.2
+Requires-Dist: torch>=2.2.2
 Requires-Dist: transformers
-Requires-Dist: pdfminer.six ==20231228
+Requires-Dist: pdfminer.six==20231228
 Provides-Extra: full
-Requires-Dist: unimernet ==0.2.3 ; extra == 'full'
-Requires-Dist: torch <=2.3.1,>=2.2.2 ; extra == 'full'
-Requires-Dist: torchvision <=0.18.1,>=0.17.2 ; extra == 'full'
-Requires-Dist: ultralytics >=8.3.48 ; extra == 'full'
-Requires-Dist: paddleocr ==2.7.3 ; extra == 'full'
-Requires-Dist: struct-eqtable ==0.3.2 ; extra == 'full'
-Requires-Dist: einops ; extra == 'full'
-Requires-Dist: accelerate ; extra == 'full'
-Requires-Dist: doclayout-yolo ==0.0.2b1 ; extra == 'full'
-Requires-Dist: rapidocr-paddle ; extra == 'full'
-Requires-Dist: rapidocr-onnxruntime ; extra == 'full'
-Requires-Dist: rapid-table <2.0.0,>=1.0.3 ; extra == 'full'
-Requires-Dist: PyYAML ; extra == 'full'
-Requires-Dist: openai ; extra == 'full'
-Requires-Dist: detectron2 ; extra == 'full'
-Requires-Dist: paddlepaddle ==3.0.0b1 ; (platform_system == "Linux") and extra == 'full'
-Requires-Dist: matplotlib ; (platform_system == "Linux" or platform_system == "Darwin") and extra == 'full'
-Requires-Dist: matplotlib <=3.9.0 ; (platform_system == "Windows") and extra == 'full'
-Requires-Dist: paddlepaddle ==2.6.1 ; (platform_system == "Windows" or platform_system == "Darwin") and extra == 'full'
+Requires-Dist: unimernet==0.2.3; extra == "full"
+Requires-Dist: torch<=2.3.1,>=2.2.2; extra == "full"
+Requires-Dist: torchvision<=0.18.1,>=0.17.2; extra == "full"
+Requires-Dist: ultralytics>=8.3.48; extra == "full"
+Requires-Dist: paddleocr==2.7.3; extra == "full"
+Requires-Dist: struct-eqtable==0.3.2; extra == "full"
+Requires-Dist: einops; extra == "full"
+Requires-Dist: accelerate; extra == "full"
+Requires-Dist: doclayout-yolo==0.0.2b1; extra == "full"
+Requires-Dist: rapidocr-paddle<2.0.0,>=1.4.5; extra == "full"
+Requires-Dist: rapidocr-onnxruntime<2.0.0,>=1.4.4; extra == "full"
+Requires-Dist: rapid-table<2.0.0,>=1.0.3; extra == "full"
+Requires-Dist: PyYAML; extra == "full"
+Requires-Dist: openai; extra == "full"
+Requires-Dist: detectron2; extra == "full"
+Requires-Dist: matplotlib; (platform_system == "Linux" or platform_system == "Darwin") and extra == "full"
+Requires-Dist: paddlepaddle==3.0.0rc1; (platform_system == "Linux" or platform_system == "Darwin") and extra == "full"
+Requires-Dist: matplotlib<=3.9.0; platform_system == "Windows" and extra == "full"
+Requires-Dist: paddlepaddle==2.6.1; platform_system == "Windows" and extra == "full"
 Provides-Extra: lite
-Requires-Dist: paddleocr ==2.7.3 ; extra == 'lite'
-Requires-Dist: paddlepaddle ==3.0.0b1 ; (platform_system == "Linux") and extra == 'lite'
-Requires-Dist: paddlepaddle ==2.6.1 ; (platform_system == "Windows" or platform_system == "Darwin") and extra == 'lite'
+Requires-Dist: paddleocr==2.7.3; extra == "lite"
+Requires-Dist: paddlepaddle==3.0.0b1; platform_system == "Linux" and extra == "lite"
+Requires-Dist: paddlepaddle==2.6.1; (platform_system == "Windows" or platform_system == "Darwin") and extra == "lite"
 Provides-Extra: old_linux
-Requires-Dist: albumentations <=1.4.20 ; extra == 'old_linux'
+Requires-Dist: albumentations<=1.4.20; extra == "old-linux"
 
 <div align="center" xmlns="http://www.w3.org/1999/html">
 <!-- logo -->
@@ -94,6 +94,15 @@ Easier to use: Just grab MinerU Desktop. No coding, no login, just a simple inte
 </div>
 
 # Changelog
+- 2025/02/24 1.2.0 released. This version includes several fixes and improvements to enhance parsing efficiency and accuracy:
+  - Performance Optimization
+    - Increased classification speed for PDF documents in auto mode.
+  - Parsing Optimization
+    - Improved parsing logic for documents containing watermarks, significantly enhancing the parsing results for such documents.
+    - Enhanced the matching logic for multiple images/tables and captions within a single page, improving the accuracy of image-text matching in complex layouts.
+  - Bug Fixes
+    - Fixed an issue where image/table spans were incorrectly filled into text blocks under certain conditions.
+    - Resolved an issue where title blocks were empty in some cases.
 - 2025/01/22 1.1.0 released. In this version we have focused on improving parsing accuracy and efficiency:
   - Model capability upgrade (requires re-executing the [model download process](docs/how_to_download_models_en.md) to obtain incremental updates of model files)
     - The layout recognition model has been upgraded to the latest `doclayout_yolo(2501)` model, improving layout recognition accuracy.
@@ -280,10 +289,9 @@ There are three different ways to experience MinerU:
 
 ### Online Demo
 
-Stable Version (Stable version verified by QA):  
-[![OpenDataLab](https://img.shields.io/badge/Demo_on_OpenDataLab-blue?logo=data:image/svg+xml;base64,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&labelColor=white)](https://mineru.net/OpenSourceTools/Extractor?source=github)
+Synced with dev branch updates:
 
-Test Version (Synced with dev branch updates, testing new features):  
+[![OpenDataLab](https://img.shields.io/badge/Demo_on_OpenDataLab-blue?logo=data:image/svg+xml;base64,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&labelColor=white)](https://mineru.net/OpenSourceTools/Extractor?source=github)
 [![HuggingFace](https://img.shields.io/badge/Demo_on_HuggingFace-yellow.svg?logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAF8AAABYCAMAAACkl9t/AAAAk1BMVEVHcEz/nQv/nQv/nQr/nQv/nQr/nQv/nQv/nQr/wRf/txT/pg7/yRr/rBD/zRz/ngv/oAz/zhz/nwv/txT/ngv/0B3+zBz/nQv/0h7/wxn/vRb/thXkuiT/rxH/pxD/ogzcqyf/nQvTlSz/czCxky7/SjifdjT/Mj3+Mj3wMj15aTnDNz+DSD9RTUBsP0FRO0Q6O0WyIxEIAAAAGHRSTlMADB8zSWF3krDDw8TJ1NbX5efv8ff9/fxKDJ9uAAAGKklEQVR42u2Z63qjOAyGC4RwCOfB2JAGqrSb2WnTw/1f3UaWcSGYNKTdf/P+mOkTrE+yJBulvfvLT2A5ruenaVHyIks33npl/6C4s/ZLAM45SOi/1FtZPyFur1OYofBX3w7d54Bxm+E8db+nDr12ttmESZ4zludJEG5S7TO72YPlKZFyE+YCYUJTBZsMiNS5Sd7NlDmKM2Eg2JQg8awbglfqgbhArjxkS7dgp2RH6hc9AMLdZYUtZN5DJr4molC8BfKrEkPKEnEVjLbgW1fLy77ZVOJagoIcLIl+IxaQZGjiX597HopF5CkaXVMDO9Pyix3AFV3kw4lQLCbHuMovz8FallbcQIJ5Ta0vks9RnolbCK84BtjKRS5uA43hYoZcOBGIG2Epbv6CvFVQ8m8loh66WNySsnN7htL58LNp+NXT8/PhXiBXPMjLSxtwp8W9f/1AngRierBkA+kk/IpUSOeKByzn8y3kAAAfh//0oXgV4roHm/kz4E2z//zRc3/lgwBzbM2mJxQEa5pqgX7d1L0htrhx7LKxOZlKbwcAWyEOWqYSI8YPtgDQVjpB5nvaHaSnBaQSD6hweDi8PosxD6/PT09YY3xQA7LTCTKfYX+QHpA0GCcqmEHvr/cyfKQTEuwgbs2kPxJEB0iNjfJcCTPyocx+A0griHSmADiC91oNGVwJ69RudYe65vJmoqfpul0lrqXadW0jFKH5BKwAeCq+Den7s+3zfRJzA61/Uj/9H/VzLKTx9jFPPdXeeP+L7WEvDLAKAIoF8bPTKT0+TM7W8ePj3Rz/Yn3kOAp2f1Kf0Weony7pn/cPydvhQYV+eFOfmOu7VB/ViPe34/EN3RFHY/yRuT8ddCtMPH/McBAT5s+vRde/gf2c/sPsjLK+m5IBQF5tO+h2tTlBGnP6693JdsvofjOPnnEHkh2TnV/X1fBl9S5zrwuwF8NFrAVJVwCAPTe8gaJlomqlp0pv4Pjn98tJ/t/fL++6unpR1YGC2n/KCoa0tTLoKiEeUPDl94nj+5/Tv3/eT5vBQ60X1S0oZr+IWRR8Ldhu7AlLjPISlJcO9vrFotky9SpzDequlwEir5beYAc0R7D9KS1DXva0jhYRDXoExPdc6yw5GShkZXe9QdO/uOvHofxjrV/TNS6iMJS+4TcSTgk9n5agJdBQbB//IfF/HpvPt3Tbi7b6I6K0R72p6ajryEJrENW2bbeVUGjfgoals4L443c7BEE4mJO2SpbRngxQrAKRudRzGQ8jVOL2qDVjjI8K1gc3TIJ5KiFZ1q+gdsARPB4NQS4AjwVSt72DSoXNyOWUrU5mQ9nRYyjp89Xo7oRI6Bga9QNT1mQ/ptaJq5T/7WcgAZywR/XlPGAUDdet3LE+qS0TI+g+aJU8MIqjo0Kx8Ly+maxLjJmjQ18rA0YCkxLQbUZP1WqdmyQGJLUm7VnQFqodmXSqmRrdVpqdzk5LvmvgtEcW8PMGdaS23EOWyDVbACZzUJPaqMbjDxpA3Qrgl0AikimGDbqmyT8P8NOYiqrldF8rX+YN7TopX4UoHuSCYY7cgX4gHwclQKl1zhx0THf+tCAUValzjI7Wg9EhptrkIcfIJjA94evOn8B2eHaVzvBrnl2ig0So6hvPaz0IGcOvTHvUIlE2+prqAxLSQxZlU2stql1NqCCLdIiIN/i1DBEHUoElM9dBravbiAnKqgpi4IBkw+utSPIoBijDXJipSVV7MpOEJUAc5Qmm3BnUN+w3hteEieYKfRZSIUcXKMVf0u5wD4EwsUNVvZOtUT7A2GkffHjByWpHqvRBYrTV72a6j8zZ6W0DTE86Hn04bmyWX3Ri9WH7ZU6Q7h+ZHo0nHUAcsQvVhXRDZHChwiyi/hnPuOsSEF6Exk3o6Y9DT1eZ+6cASXk2Y9k+6EOQMDGm6WBK10wOQJCBwren86cPPWUcRAnTVjGcU1LBgs9FURiX/e6479yZcLwCBmTxiawEwrOcleuu12t3tbLv/N4RLYIBhYexm7Fcn4OJcn0+zc+s8/VfPeddZHAGN6TT8eGczHdR/Gts1/MzDkThr23zqrVfAMFT33Nx1RJsx1k5zuWILLnG/vsH+Fv5D4NTVcp1Gzo8AAAAAElFTkSuQmCC&labelColor=white)](https://huggingface.co/spaces/opendatalab/MinerU)
 [![ModelScope](https://img.shields.io/badge/Demo_on_ModelScope-purple?logo=data:image/svg+xml;base64,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&labelColor=white)](https://www.modelscope.cn/studios/OpenDataLab/MinerU)
 
@@ -292,8 +300,8 @@ Test Version (Synced with dev branch updates, testing new features):
 #### 1. Install magic-pdf
 
 ```bash
-conda create -n MinerU python=3.10
-conda activate MinerU
+conda create -n mineru python=3.10
+conda activate mineru
 pip install -U "magic-pdf[full]" --extra-index-url https://wheels.myhloli.com
 ```
 
@@ -353,7 +361,7 @@ If your device supports CUDA and meets the GPU requirements of the mainline envi
   ```bash
   wget https://github.com/opendatalab/MinerU/raw/master/docker/global/Dockerfile -O Dockerfile
   docker build -t mineru:latest .
-  docker run --rm -it --gpus=all mineru:latest /bin/bash -c "echo 'source /opt/mineru_venv/bin/activate' >> ~/.bashrc && exec bash"
+  docker run -it --name mineru --gpus=all mineru:latest /bin/bash -c "echo 'source /opt/mineru_venv/bin/activate' >> ~/.bashrc && exec bash"
   magic-pdf --help
   ```
 

{magic_pdf-1.1.0.dist-info → magic_pdf-1.2.0.dist-info}/RECORD

@@ -1,5 +1,5 @@
 magic_pdf/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-magic_pdf/pdf_parse_union_core_v2.py,sha256=qh-Vj7v8EenC_f_MNMa76i1DVuckulQo1QC1IOw8LRE,37723
+magic_pdf/pdf_parse_union_core_v2.py,sha256=jIrXgU_gKL4toJ6GsCoDxByszaN8mAr5vrEy_c63ewk,38310
 magic_pdf/config/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/config/constants.py,sha256=fXGzANULnJWLPxwYp3BEFWx-rnorzpySMx63ffyEyq4,1272
 magic_pdf/config/drop_reason.py,sha256=CqjMzBE96Qo8OeFvhhhItY8WhyqsKhE3DmyJLoQZNCc,2248
@@ -24,10 +24,10 @@ magic_pdf/data/io/base.py,sha256=SqNQqe30ZvoVvg7GVv-hLMCjN6yBgDyQQWeLgGsTfhQ,111
 magic_pdf/data/io/http.py,sha256=XlKB0DNf4a_uUnfgcclvaaOtmE7lmddx0DnK8A-emAM,958
 magic_pdf/data/io/s3.py,sha256=hyA7sbNriQy64xd_uyJ7acN_oneQ1Pdmoc7_xcvkue8,3606
 magic_pdf/dict2md/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-magic_pdf/dict2md/ocr_mkcontent.py,sha256=RQ47F2CT0Zgmg1rZoqYj5IW5msqoCTEF6GEHi3mVd8U,12989
-magic_pdf/filter/__init__.py,sha256=rV4dvUxfKyVErDx9ZbUp8DVq_fRIlv0lfSXp1ND4STc,1503
-magic_pdf/filter/pdf_classify_by_type.py,sha256=spmDO-f2ihAl1d6-EP-j271Yi50oyu6mw4X2kRd_m0s,42320
-magic_pdf/filter/pdf_meta_scan.py,sha256=rqTuStW2_ICr3HmV_9IQ5jnsl4JnSh7-bL11vbtH3i0,17470
+magic_pdf/dict2md/ocr_mkcontent.py,sha256=ZZTaiIn18OWuWKGbDdpoOZ3VMhe_3_JKwrKCfzDiSk0,13715
+magic_pdf/filter/__init__.py,sha256=_7lSez_myu4b6cdzPpQ-NfREuqeBSq_QdyBPKVLyq2U,1505
+magic_pdf/filter/pdf_classify_by_type.py,sha256=YNYXamxYgEiSujwilCNHOtrwpgJGDiQ597qJfardDVc,42354
+magic_pdf/filter/pdf_meta_scan.py,sha256=eOuM0-JgaXvHolSgepGoNDJDmv_uITWLQpH_0MfnVQw,17478
 magic_pdf/integrations/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/integrations/rag/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/integrations/rag/api.py,sha256=t38wvIBzLje4_JzTP3dewMLqV-tQJ-A3B92Sj2oyrfs,2507
@@ -47,23 +47,23 @@ magic_pdf/libs/language.py,sha256=7RT3mxSa7jdpoC5ySd7ZddHA7TO7UsnmDOWiYZAxuyg,14
 magic_pdf/libs/local_math.py,sha256=tqljQOgqh3fZc146HYhO88JXJaiXMVwArBkk_CSGICc,177
 magic_pdf/libs/markdown_utils.py,sha256=86v2BmsSV4NkoRZrH4uQD1youJhYFF3vIKr_vDeg3z0,270
 magic_pdf/libs/path_utils.py,sha256=Hykw_l5CU736b2egHV9P7B-qh3QNKO4nZSGCbsi0Z8E,1043
-magic_pdf/libs/pdf_check.py,sha256=zBwUThKKBtnrNPmgE10lYsTy1Kq7j_6IejO7JR0J4pA,3118
+magic_pdf/libs/pdf_check.py,sha256=7GWWvDR6g_rj_fE6XJlbTq5AFVX11ngRIzT0N18F214,3396
 magic_pdf/libs/pdf_image_tools.py,sha256=kjzSEbm7K0yiHv8kJ4VbZ9HHktM8qvAv3LhxRyDZEQk,1987
 magic_pdf/libs/safe_filename.py,sha256=ckwcM_eqoysTb5id8czp-tXq2G9da0-l3pshZDCHQtE,236
-magic_pdf/libs/version.py,sha256=LGVQyDsWifdACo7qztwb8RWWHds1E7uQ-ZqD8SAjyw4,22
+magic_pdf/libs/version.py,sha256=MpAT5hgNoHnTtG1XRD_GV_A7QrHVU6vJjGSw_8qMGA4,22
 magic_pdf/model/__init__.py,sha256=sa-dO2k-TLy25I2gRrzjm_cQeYfzMf-pLwBJHkIxGo0,51
 magic_pdf/model/batch_analyze.py,sha256=sbrgOJWycb1Ep6e62CPi6jEyG6VSeklIxc4PmrqaLhM,11933
-magic_pdf/model/doc_analyze_by_custom_model.py,sha256=zryHy3ljcEvDqOWivXZQrpau_jPtt6x1lLOZaOkk_tI,8153
-magic_pdf/model/magic_model.py,sha256=Nt74oZGYUcbm4qdOQtN-hbKhXxlWO2LVv3K9yXvteWY,25204
+magic_pdf/model/doc_analyze_by_custom_model.py,sha256=wma0aq6RyxAepEqnaiTJ9_pWWKLVBj39c6xWA85dxzA,8068
+magic_pdf/model/magic_model.py,sha256=OcKhSJ_PyAAldgpKPiPxi2uuvnj3Sf4SvXi_5Rv0a6Q,30667
 magic_pdf/model/model_list.py,sha256=aqfEJlEfbib3D3ISrxc0Coh6SbffYh8Yq2FlQN35_zA,213
-magic_pdf/model/pdf_extract_kit.py,sha256=S-UVZQroUe-eEAJzuOucdCh9FCAWy2DVAZow3dGUiWI,12520
+magic_pdf/model/pdf_extract_kit.py,sha256=Rd51VNZPKRA_tUbDss-b44d84K6WDG2S87a37Ax7HUA,12224
 magic_pdf/model/pp_structure_v2.py,sha256=NcqFWL4nUtjl82MFak8HX_8V3i4Aw_fK4dATrIp5uGs,3840
 magic_pdf/model/sub_modules/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-magic_pdf/model/sub_modules/model_init.py,sha256=RCv6BkRLEFBKrfVReRvIvbRQ21BZLz8jj-AKQhwHkhw,6520
+magic_pdf/model/sub_modules/model_init.py,sha256=Ltwi3Nd5PdVVXRF9fto5nImFVg6w-twAMzOLV_F-c3g,7693
 magic_pdf/model/sub_modules/model_utils.py,sha256=2pI1Xcr2zCF3b64e4WoFtIbjSmTVYBE4zjyHB23gvmE,2488
 magic_pdf/model/sub_modules/language_detection/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5c,50
 magic_pdf/model/sub_modules/language_detection/utils.py,sha256=5nec_loLyYCJ5o6n38AYLz2SKmRvHDCBdt6ka84EaGM,3096
-magic_pdf/model/sub_modules/language_detection/yolov11/YOLOv11.py,sha256=GW_9WkqIzpJm1MFJexZ2ZvA6AjoqM-6yh8p4LupJhas,4762
+magic_pdf/model/sub_modules/language_detection/yolov11/YOLOv11.py,sha256=bl2i7kweoJNdj47FlE9h0B_-nNQrMcW9mCLQ1puMEH8,4893
 magic_pdf/model/sub_modules/language_detection/yolov11/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5c,50
 magic_pdf/model/sub_modules/layout/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/model/sub_modules/layout/doclayout_yolo/DocLayoutYOLO.py,sha256=gy7rc8poO-Zr8511NJjuBV8Uryq5k3JKrstLtCONg0c,2237
@@ -115,14 +115,13 @@ magic_pdf/operators/__init__.py,sha256=liU2-WYUvsQ1G4PYBppyvokS9z5IjrnlVMtoBAC1R
 magic_pdf/operators/models.py,sha256=mRqbCVrxxaUVDpEBAsXaK7EL1M-goICkE1W0FYgewio,5305
 magic_pdf/operators/pipes.py,sha256=XgBgisKQd_ruW-3Tw4v5LhqloZUHgn2aFcpi_q8LbCs,6767
 magic_pdf/post_proc/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5c,50
-magic_pdf/post_proc/llm_aided.py,sha256=p-XwDObLkDv5rPxsdI7092MP-rHCMr1uAUq3fs7Zc-E,6334
-magic_pdf/post_proc/llm_aided_ocr.py,sha256=89kxzEQVqNGSUtmvgcg2AVDDmgb43bamdRxXbwS2FxQ,33557
+magic_pdf/post_proc/llm_aided.py,sha256=yzhu2cCpUZjdwf3v0swYDgSs9VWIfMAoXepYIP1EMZs,6367
 magic_pdf/post_proc/para_split_v3.py,sha256=v4SdQn4OZdHRXpWQMfQ-FGJz_tglQ88uFUqpwY542Fo,16922
 magic_pdf/pre_proc/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/pre_proc/construct_page_dict.py,sha256=OFmq5XRKi6fYIo-lmGlL-NB16Sf0egzsfEx-fT2uYrc,660
 magic_pdf/pre_proc/cut_image.py,sha256=NDzbxwD7z7Tb4uAxL4KR6LzURFdN1Tzr4nPvj-VmEqc,1225
 magic_pdf/pre_proc/ocr_detect_all_bboxes.py,sha256=nt88ttXCEI_1ihAF7HU15SQjwM69V-iJmk-L_nyzA6o,9328
-magic_pdf/pre_proc/ocr_dict_merge.py,sha256=Ycgz2whzotL7kwl0-mHNV48QOQ2j4tRXqLSQrJRojYg,4847
+magic_pdf/pre_proc/ocr_dict_merge.py,sha256=vrbLIzNIjxrm7PonfHaFdY6qaicc0uIly62SJwgZ5UM,5496
 magic_pdf/pre_proc/ocr_span_list_modify.py,sha256=xrgC9vR0poklZuY4Og41pZVdXzuaGFg3BnQ01X60dpo,3102
 magic_pdf/pre_proc/remove_bbox_overlap.py,sha256=mcdxAh4P56NZ3Ij8h3vW8qC_SrszfXflVWuWUuUiTNg,3089
 magic_pdf/resources/fasttext-langdetect/lid.176.ftz,sha256=jzRyz-hzintgmejpmcPL-uDc0VaWqsfXc4qAOdtgPoM,938013
@@ -139,9 +138,9 @@ magic_pdf/tools/common.py,sha256=1LfMeXBBsb3WlGeNAze_pPOYXQ8Qbfh-JgRXweojHKo,838
 magic_pdf/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/utils/annotations.py,sha256=82ou3uELNbQWa9hOFFkVt0gsIskAKf5msCv5J2IJ5V0,211
 magic_pdf/utils/office_to_pdf.py,sha256=7aj-Ls2v8saD-Rgu_t3FIc-J3Ka9wnmiEH5zY-H1Vxs,729
-magic_pdf-1.1.0.dist-info/LICENSE.md,sha256=jVa0BUaKrRH4erV2P5AeJ24I2WRv9chIGxditreJ6e0,34524
-magic_pdf-1.1.0.dist-info/METADATA,sha256=Ud48caL9BHS-ZuLN-3VpswLJFPqao7KqY0aqfF0ApOo,40958
-magic_pdf-1.1.0.dist-info/WHEEL,sha256=yQN5g4mg4AybRjkgi-9yy4iQEFibGQmlz78Pik5Or-A,92
-magic_pdf-1.1.0.dist-info/entry_points.txt,sha256=wXwYke3j8fqDQTocUspL-CqDUEv3Tfcwp09fM8dZAhA,98
-magic_pdf-1.1.0.dist-info/top_level.txt,sha256=J9I0AzmHWGkp9c6DL8Oe4mEx3yYphLzkRn4H25Lg1rE,10
-magic_pdf-1.1.0.dist-info/RECORD,,
+magic_pdf-1.2.0.dist-info/LICENSE.md,sha256=jVa0BUaKrRH4erV2P5AeJ24I2WRv9chIGxditreJ6e0,34524
+magic_pdf-1.2.0.dist-info/METADATA,sha256=7iel3MItxKhJc1Bbfh_NMbDp8a23k9G1vA8LYEw2k_U,40720
+magic_pdf-1.2.0.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
+magic_pdf-1.2.0.dist-info/entry_points.txt,sha256=wXwYke3j8fqDQTocUspL-CqDUEv3Tfcwp09fM8dZAhA,98
+magic_pdf-1.2.0.dist-info/top_level.txt,sha256=J9I0AzmHWGkp9c6DL8Oe4mEx3yYphLzkRn4H25Lg1rE,10
+magic_pdf-1.2.0.dist-info/RECORD,,

{magic_pdf-1.1.0.dist-info → magic_pdf-1.2.0.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: bdist_wheel (0.41.2)
+Generator: bdist_wheel (0.45.1)
 Root-Is-Purelib: true
 Tag: py3-none-any