mineru 2.5.1__py3-none-any.whl → 2.5.3__py3-none-any.whl

mineru/backend/pipeline/batch_analyze.py

@@ -116,9 +116,14 @@ class BatchAnalyze:
                 atom_model_name=AtomicModel.ImgOrientationCls,
             )
             try:
-                img_orientation_cls_model.batch_predict(table_res_list_all_page,
-                                                        det_batch_size=self.batch_ratio * OCR_DET_BASE_BATCH_SIZE,
-                                                        batch_size=TABLE_ORI_CLS_BATCH_SIZE)
+                if self.enable_ocr_det_batch:
+                    img_orientation_cls_model.batch_predict(table_res_list_all_page,
+                                                            det_batch_size=self.batch_ratio * OCR_DET_BASE_BATCH_SIZE,
+                                                            batch_size=TABLE_ORI_CLS_BATCH_SIZE)
+                else:
+                    for table_res in table_res_list_all_page:
+                        rotate_label = img_orientation_cls_model.predict(table_res['table_img'])
+                        img_orientation_cls_model.img_rotate(table_res, rotate_label)
             except Exception as e:
                 logger.warning(
                     f"Image orientation classification failed: {e}, using original image"

mineru/backend/vlm/custom_logits_processors.py

@@ -0,0 +1,41 @@
+import os
+
+from loguru import logger
+from packaging import version
+
+
+def enable_custom_logits_processors():
+    import torch
+    from vllm import __version__ as vllm_version
+
+    if not torch.cuda.is_available():
+        logger.info("CUDA not available, disabling custom_logits_processors")
+        return False
+
+    major, minor = torch.cuda.get_device_capability()
+    # 正确计算Compute Capability
+    compute_capability = f"{major}.{minor}"
+
+    # 安全地处理环境变量
+    vllm_use_v1_str = os.getenv('VLLM_USE_V1', "1")
+    if vllm_use_v1_str.isdigit():
+        vllm_use_v1 = int(vllm_use_v1_str)
+    else:
+        vllm_use_v1 = 1
+
+    if vllm_use_v1 == 0:
+        logger.info("VLLM_USE_V1 is set to 0, disabling custom_logits_processors")
+        return False
+    elif version.parse(vllm_version) < version.parse("0.10.1"):
+        logger.info(f"vllm version: {vllm_version} < 0.10.1, disable custom_logits_processors")
+        return False
+    elif version.parse(compute_capability) < version.parse("8.0"):
+        if version.parse(vllm_version) >= version.parse("0.10.2"):
+            logger.info(f"compute_capability: {compute_capability} < 8.0, but vllm version: {vllm_version} >= 0.10.2, enable custom_logits_processors")
+            return True
+        else:
+            logger.info(f"compute_capability: {compute_capability} < 8.0 and vllm version: {vllm_version} < 0.10.2, disable custom_logits_processors")
+            return False
+    else:
+        logger.info(f"compute_capability: {compute_capability} >= 8.0 and vllm version: {vllm_version} >= 0.10.1, enable custom_logits_processors")
+        return True

mineru/backend/vlm/vlm_analyze.py

@@ -4,6 +4,7 @@ import time
 
 from loguru import logger
 
+from .custom_logits_processors import enable_custom_logits_processors
 from .model_output_to_middle_json import result_to_middle_json
 from ...data.data_reader_writer import DataWriter
 from mineru.utils.pdf_image_tools import load_images_from_pdf
@@ -88,7 +89,6 @@ class ModelSingleton:
                 elif backend == "vllm-engine":
                     try:
                         import vllm
-                        vllm_version = vllm.__version__
                         from mineru_vl_utils import MinerULogitsProcessor
                     except ImportError:
                         raise ImportError("Please install vllm to use the vllm-engine backend.")
@@ -96,7 +96,7 @@ class ModelSingleton:
                         kwargs["gpu_memory_utilization"] = 0.5
                     if "model" not in kwargs:
                         kwargs["model"] = model_path
-                    if version.parse(vllm_version) >= version.parse("0.10.1") and "logits_processors" not in kwargs:
+                    if enable_custom_logits_processors() and ("logits_processors" not in kwargs):
                         kwargs["logits_processors"] = [MinerULogitsProcessor]
                     # 使用kwargs为 vllm初始化参数
                     vllm_llm = vllm.LLM(**kwargs)
@@ -104,7 +104,6 @@ class ModelSingleton:
                     try:
                         from vllm.engine.arg_utils import AsyncEngineArgs
                         from vllm.v1.engine.async_llm import AsyncLLM
-                        from vllm import __version__ as vllm_version
                         from mineru_vl_utils import MinerULogitsProcessor
                     except ImportError:
                         raise ImportError("Please install vllm to use the vllm-async-engine backend.")
@@ -112,7 +111,7 @@ class ModelSingleton:
                         kwargs["gpu_memory_utilization"] = 0.5
                     if "model" not in kwargs:
                         kwargs["model"] = model_path
-                    if version.parse(vllm_version) >= version.parse("0.10.1") and "logits_processors" not in kwargs:
+                    if enable_custom_logits_processors() and ("logits_processors" not in kwargs):
                         kwargs["logits_processors"] = [MinerULogitsProcessor]
                     # 使用kwargs为 vllm初始化参数
                     vllm_async_llm = AsyncLLM.from_engine_args(AsyncEngineArgs(**kwargs))

mineru/backend/vlm/vlm_middle_json_mkcontent.py

@@ -54,7 +54,7 @@ def mk_blocks_to_markdown(para_blocks, make_mode, formula_enable, table_enable,
         elif para_type == BlockType.LIST:
             for block in para_block['blocks']:
                 item_text = merge_para_with_text(block, formula_enable=formula_enable, img_buket_path=img_buket_path)
-                para_text += f"{item_text}\n"
+                para_text += f"{item_text}  \n"
         elif para_type == BlockType.TITLE:
             title_level = get_title_level(para_block)
             para_text = f'{"#" * title_level} {merge_para_with_text(para_block)}'

mineru/model/ori_cls/paddle_ori_cls.py

@@ -255,25 +255,28 @@ class PaddleOrientationClsModel:
                     results = self.sess.run(None, {"x": x})
                     for img_info, res in zip(rotated_imgs, results[0]):
                         label = self.labels[np.argmax(res)]
-                        if label == "270":
-                            img_info["table_img"] = cv2.rotate(
-                                np.asarray(img_info["table_img"]),
-                                cv2.ROTATE_90_CLOCKWISE,
-                            )
-                            img_info["wired_table_img"] = cv2.rotate(
-                                np.asarray(img_info["wired_table_img"]),
-                                cv2.ROTATE_90_CLOCKWISE,
-                            )
-                        elif label == "90":
-                            img_info["table_img"] = cv2.rotate(
-                                np.asarray(img_info["table_img"]),
-                                cv2.ROTATE_90_COUNTERCLOCKWISE,
-                            )
-                            img_info["wired_table_img"] = cv2.rotate(
-                                np.asarray(img_info["wired_table_img"]),
-                                cv2.ROTATE_90_COUNTERCLOCKWISE,
-                            )
-                        else:
-                            # 180度和0度不做处理
-                            pass
+                        self.img_rotate(img_info, label)
                         pbar.update(1)
+
+    def img_rotate(self, img_info, label):
+        if label == "270":
+            img_info["table_img"] = cv2.rotate(
+                np.asarray(img_info["table_img"]),
+                cv2.ROTATE_90_CLOCKWISE,
+            )
+            img_info["wired_table_img"] = cv2.rotate(
+                np.asarray(img_info["wired_table_img"]),
+                cv2.ROTATE_90_CLOCKWISE,
+            )
+        elif label == "90":
+            img_info["table_img"] = cv2.rotate(
+                np.asarray(img_info["table_img"]),
+                cv2.ROTATE_90_COUNTERCLOCKWISE,
+            )
+            img_info["wired_table_img"] = cv2.rotate(
+                np.asarray(img_info["wired_table_img"]),
+                cv2.ROTATE_90_COUNTERCLOCKWISE,
+            )
+        else:
+            # 180度和0度不做处理
+            pass

mineru/model/vlm_vllm_model/server.py

@@ -1,10 +1,9 @@
 import sys
 
+from mineru.backend.vlm.custom_logits_processors import enable_custom_logits_processors
 from mineru.utils.models_download_utils import auto_download_and_get_model_root_path
 
 from vllm.entrypoints.cli.main import main as vllm_main
-from vllm import __version__ as vllm_version
-from packaging import version
 
 
 def main():
@@ -37,6 +36,8 @@ def main():
         for index in sorted(model_arg_indices, reverse=True):
             args.pop(index)
 
+    custom_logits_processors = enable_custom_logits_processors()
+
     # 添加默认参数
     if not has_port_arg:
         args.extend(["--port", "30000"])
@@ -44,7 +45,7 @@ def main():
         args.extend(["--gpu-memory-utilization", "0.5"])
     if not model_path:
         model_path = auto_download_and_get_model_root_path("/", "vlm")
-    if not has_logits_processors_arg and version.parse(vllm_version) >= version.parse("0.10.1"):
+    if (not has_logits_processors_arg) and custom_logits_processors:
         args.extend(["--logits-processors", "mineru_vl_utils:MinerULogitsProcessor"])
 
     # 重构参数，将模型路径作为位置参数

mineru/version.py

@@ -1 +1 @@
-__version__ = "2.5.1"
+__version__ = "2.5.3"

{mineru-2.5.1.dist-info → mineru-2.5.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mineru
-Version: 2.5.1
+Version: 2.5.3
 Summary: A practical tool for converting PDF to Markdown
 License: AGPL-3.0
 Project-URL: homepage, https://mineru.net/
@@ -37,7 +37,7 @@ Requires-Dist: scikit-image<1.0.0,>=0.25.0
 Requires-Dist: openai<2,>=1.70.0
 Requires-Dist: beautifulsoup4<5,>=4.13.5
 Requires-Dist: magika<0.7.0,>=0.6.2
-Requires-Dist: mineru-vl-utils<1,>=0.1.7
+Requires-Dist: mineru-vl-utils<1,>=0.1.8
 Provides-Extra: test
 Requires-Dist: mineru[core]; extra == "test"
 Requires-Dist: pytest; extra == "test"
@@ -45,11 +45,11 @@ Requires-Dist: pytest-cov; extra == "test"
 Requires-Dist: coverage; extra == "test"
 Requires-Dist: fuzzywuzzy; extra == "test"
 Provides-Extra: vlm
-Requires-Dist: torch<2.8.0,>=2.6.0; extra == "vlm"
+Requires-Dist: torch<3,>=2.6.0; extra == "vlm"
 Requires-Dist: transformers<5.0.0,>=4.51.1; extra == "vlm"
 Requires-Dist: accelerate>=1.5.1; extra == "vlm"
 Provides-Extra: vllm
-Requires-Dist: vllm==0.10.1.1; extra == "vllm"
+Requires-Dist: vllm<0.11,>=0.10.1.1; extra == "vllm"
 Provides-Extra: pipeline
 Requires-Dist: matplotlib<4,>=3.10; extra == "pipeline"
 Requires-Dist: ultralytics<9,>=8.3.48; extra == "pipeline"
@@ -60,7 +60,7 @@ Requires-Dist: ftfy<7,>=6.3.1; extra == "pipeline"
 Requires-Dist: shapely<3,>=2.0.7; extra == "pipeline"
 Requires-Dist: pyclipper<2,>=1.3.0; extra == "pipeline"
 Requires-Dist: omegaconf<3,>=2.3.0; extra == "pipeline"
-Requires-Dist: torch<2.8.0,>=2.6.0; extra == "pipeline"
+Requires-Dist: torch<3,>=2.6.0; extra == "pipeline"
 Requires-Dist: torchvision; extra == "pipeline"
 Requires-Dist: transformers!=4.51.0,<5.0.0,>=4.49.0; extra == "pipeline"
 Requires-Dist: onnxruntime>1.17.0; extra == "pipeline"
@@ -127,7 +127,13 @@ Dynamic: license-file
 
 # Changelog
 
-- 2025/09/19 2.5.1 Released
+- 2025/09/20 2.5.3 Released
+  - Dependency version range adjustment to enable Turing and earlier architecture GPUs to use vLLM acceleration for MinerU2.5 model inference.
+  - `pipeline` backend compatibility fixes for torch 2.8.0.
+  - Reduced default concurrency for vLLM async backend to lower server pressure and avoid connection closure issues caused by high load.
+  - More compatibility-related details can be found in the [announcement](https://github.com/opendatalab/MinerU/discussions/3548)
+
+- 2025/09/19 2.5.2 Released
 
   We are officially releasing MinerU2.5, currently the most powerful multimodal large model for document parsing.
   With only 1.2B parameters, MinerU2.5's accuracy on the OmniDocBench benchmark comprehensively surpasses top-tier multimodal models like Gemini 2.5 Pro, GPT-4o, and Qwen2.5-VL-72B. It also significantly outperforms leading specialized models such as dots.ocr, MonkeyOCR, and PP-StructureV3.

{mineru-2.5.1.dist-info → mineru-2.5.3.dist-info}/RECORD

@@ -1,8 +1,8 @@
 mineru/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5c,50
-mineru/version.py,sha256=PfQ9ThOuZlUZhThya-_PpR02LjazRR6LNSivpta03mM,22
+mineru/version.py,sha256=9As6EseYF3Vi8VWUZnZuuTi-kW5fYIv8u6cHK0RidGo,22
 mineru/backend/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5c,50
 mineru/backend/pipeline/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5c,50
-mineru/backend/pipeline/batch_analyze.py,sha256=rp9nHYmuBBytlJIc3oRwqTtgFd5mhRak5UMhQ4mu02Y,21896
+mineru/backend/pipeline/batch_analyze.py,sha256=dOnktvOMjfg84w1H34YlJg6N9_x6Yfvf14NIpOQcZqQ,22221
 mineru/backend/pipeline/model_init.py,sha256=tb0kOqXrSJ9HuS65Z3tr7C84P43fOIc_zlnVY5TEvXI,8440
 mineru/backend/pipeline/model_json_to_middle_json.py,sha256=i0Kmd31gR1kp0-UGtQIS7byiRRQBm2q7G-zpYIL1lwo,10898
 mineru/backend/pipeline/model_list.py,sha256=62_61GQ9ZgS5YmNSpzRkts9kLtbSu8dYrfn-Cs4Hr8I,223
@@ -11,10 +11,11 @@ mineru/backend/pipeline/pipeline_analyze.py,sha256=rbO5AetOdnxR5ctkoDzFCFoElkz7J
 mineru/backend/pipeline/pipeline_magic_model.py,sha256=w8jGx8f6yZN0Wf2yPP3L9rYKc9rogxreZCrUJzJvPO8,14974
 mineru/backend/pipeline/pipeline_middle_json_mkcontent.py,sha256=xWWOFmYL6hB8PLrxQFyRJ72dAmTIDHtqiWV-WFUfR44,14081
 mineru/backend/vlm/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5c,50
+mineru/backend/vlm/custom_logits_processors.py,sha256=ShdWaCjCcoP1NxAq1U5LuyD391NmcnLDdZ6jm5RjDHk,1634
 mineru/backend/vlm/model_output_to_middle_json.py,sha256=e4Yc98_Cth2cjVPybPGehD5cpjGcTka4D2qKKrP_qqo,5121
-mineru/backend/vlm/vlm_analyze.py,sha256=jR5DTHT0qtOLVos5-qSGPvlRCdYREoAshFej98FY5ao,8282
+mineru/backend/vlm/vlm_analyze.py,sha256=ZUXM7LSNlNhEAL7JPu8jMb8X_wurmixnlkHr0buAcI8,8189
 mineru/backend/vlm/vlm_magic_model.py,sha256=o1WKwgArV0f4lp7ufmSq9zRZaM5bnfOdx1AQtKnosro,16379
-mineru/backend/vlm/vlm_middle_json_mkcontent.py,sha256=DX2TRpfSIqpuhWqou5QXNtCW40ddQi0kdQxXi4QgzKs,13375
+mineru/backend/vlm/vlm_middle_json_mkcontent.py,sha256=Ie95XpwTgi7EmidcwE_scvXMRQjE2xASU_Rm_F8EP-I,13377
 mineru/cli/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5c,50
 mineru/cli/client.py,sha256=uo7db9Wqj1Mc11MYuaM-bi54BfKKU3SFB9Urc8md5X4,6641
 mineru/cli/common.py,sha256=yJPdrwSYVidl2hTJ2Hn2YhnfH97GJ-QZi20dGFz7h5c,14025
@@ -120,7 +121,7 @@ mineru/model/ocr/paddleocr2pytorch/tools/infer/predict_rec.py,sha256=q3l32JbBlqN
 mineru/model/ocr/paddleocr2pytorch/tools/infer/predict_system.py,sha256=hkegkn6hq2v2zqHVAP615-k-fkTS8swRYSbZeoqmSI8,3822
 mineru/model/ocr/paddleocr2pytorch/tools/infer/pytorchocr_utility.py,sha256=i1PFN-_kefJUUZ4Vk7igs1TU8gfErTDlDXY6-8Uaurw,9323
 mineru/model/ori_cls/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5c,50
-mineru/model/ori_cls/paddle_ori_cls.py,sha256=Ct0KFo-LEc_GBMYKdiv6c7DW2rxBkrt-i2uyfExmB5M,12178
+mineru/model/ori_cls/paddle_ori_cls.py,sha256=VIS22IerHST7g60AC9r2PEQIG6NQWeQaH1OrXIxNTsg,11943
 mineru/model/reading_order/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5c,50
 mineru/model/reading_order/layout_reader.py,sha256=IVUFcNMDF3-kio-BIxjppHnWS3eHPqvvNihIw2fbIFM,4372
 mineru/model/reading_order/xycut.py,sha256=ezNSq_Y4UXiztB58hbXJsjTJlOBqWIjuW5A2uLSaZSo,7349
@@ -143,7 +144,7 @@ mineru/model/table/rec/unet_table/utils.py,sha256=CYAqJW0wePJk4NAemb8W203N7E32v0
 mineru/model/table/rec/unet_table/utils_table_line_rec.py,sha256=zrCdPwI4M8nu0FEfd7lRJAe0z8kYq3KFbzwElM82USE,11174
 mineru/model/table/rec/unet_table/utils_table_recover.py,sha256=XksJsY82ZS0kqUnNT-jvaYzxJ3V3svMSzj0puwIau1k,10651
 mineru/model/vlm_vllm_model/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-mineru/model/vlm_vllm_model/server.py,sha256=v07x1esggP7Wbw0r8NeAbqG2kuJN9x5Xl2CmE2x0qzk,2003
+mineru/model/vlm_vllm_model/server.py,sha256=0hlM50Mrz9IhhCte1ZDu0hTxKQguxMm7-rtpm6Z0LLM,2054
 mineru/resources/header.html,sha256=NO8ZZdCYLqu_E72AtNcuRnA2NbFBamScjjGhtg9PKiM,4409
 mineru/resources/fasttext-langdetect/lid.176.ftz,sha256=jzRyz-hzintgmejpmcPL-uDc0VaWqsfXc4qAOdtgPoM,938013
 mineru/utils/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5c,50
@@ -172,9 +173,9 @@ mineru/utils/run_async.py,sha256=rPeP4BCZerR8VByRDhiYzfZiahLVqoZEBVAS54dAjNg,128
 mineru/utils/span_block_fix.py,sha256=0eVQjJCrT03woRt9hoh6Uu42Tp1dacfGTv2x3B9qq94,8797
 mineru/utils/span_pre_proc.py,sha256=h41q2uQajI0xQbc_30hqaju1dv3oVYxBAlKgURl8HIc,13692
 mineru/utils/table_merge.py,sha256=zYUpYLrfhBCnbHCYZi6rG8-s38NDnTbiNTObvLdYwJk,11494
-mineru-2.5.1.dist-info/licenses/LICENSE.md,sha256=jVa0BUaKrRH4erV2P5AeJ24I2WRv9chIGxditreJ6e0,34524
-mineru-2.5.1.dist-info/METADATA,sha256=BnhjCe7hE1EMEZdmz-YQgX0Mq5LANyNWHayKFaJztw0,64460
-mineru-2.5.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-mineru-2.5.1.dist-info/entry_points.txt,sha256=luXmbhPiZK_tKlRgWuYOaW_V6EFpG-yJcAevVv9MEqE,252
-mineru-2.5.1.dist-info/top_level.txt,sha256=zuGQfZcbsHv4I4oKI9gaKPqEWBFm6xJroKuug2LnKP8,7
-mineru-2.5.1.dist-info/RECORD,,
+mineru-2.5.3.dist-info/licenses/LICENSE.md,sha256=jVa0BUaKrRH4erV2P5AeJ24I2WRv9chIGxditreJ6e0,34524
+mineru-2.5.3.dist-info/METADATA,sha256=T45OlRf2MarpDXHk36RO4Hx8EN2A15V2UJmmjTd8YCE,64957
+mineru-2.5.3.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+mineru-2.5.3.dist-info/entry_points.txt,sha256=luXmbhPiZK_tKlRgWuYOaW_V6EFpG-yJcAevVv9MEqE,252
+mineru-2.5.3.dist-info/top_level.txt,sha256=zuGQfZcbsHv4I4oKI9gaKPqEWBFm6xJroKuug2LnKP8,7
+mineru-2.5.3.dist-info/RECORD,,