PaddleX/zhch/pdf2md.py

from pathlib import Path
from paddlex import create_pipeline

# 统计耗时
import time
start_time = time.time()

pipeline = create_pipeline(pipeline="./PP-StructureV3-zhch.yaml")

input_file = "./sample_data/300674-财报数据-扫描.pdf"
output_path = Path("./sample_data") / f"{Path(input_file).stem}_output"

output = pipeline.predict(
    input=input_file,
    device="gpu",  # 或者 "gpu" 如果你有 GPU 支持
    enable_mkldnn=False,  # 如果你不需要 MKLDNN 优化，可以设置为 False
    use_doc_orientation_classify=True, # 开启文档方向分类
    use_doc_unwarping=False, # 开启文档去畸变
    # layout_detection_model_name=None, # 如果要禁用版面分析，可以这样设置，或者依赖其默认行为
    use_seal_recognition=True,         # 跳过印章识别
    use_formula_recognition=False,      # 跳过公式识别
    use_chart_recognition=True,         # 跳过图表识别
    use_table_recognition=True,        # 开启表格识别
)
markdown_list = []
markdown_images = []

for res in output:
    res.save_all(save_path=output_path)  # 保存所有结果到指定路径
    md_info = res.markdown
    markdown_list.append(md_info)
    markdown_images.append(md_info.get("markdown_images", {}))

markdown_texts = pipeline.concatenate_markdown_pages(markdown_list)

mkd_file_path = output_path / f"{Path(input_file).stem}.md"
mkd_file_path.parent.mkdir(parents=True, exist_ok=True)

with open(mkd_file_path, "w", encoding="utf-8") as f:
    f.write(markdown_texts)

for item in markdown_images:
    if item:
        for path, image in item.items():
            file_path = output_path / path
            file_path.parent.mkdir(parents=True, exist_ok=True)
            image.save(file_path)

# 统计耗时
end_time = time.time()
elapsed_time = end_time - start_time
print(f"Total time taken: {elapsed_time:.2f} seconds")