mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2026-04-23 00:17:25 +08:00
MingkunZhang
142 lines
5.6 KiB
Python
142 lines
5.6 KiB
Python
import os
|
|
import unittest
|
|
|
|
from paddleocr import PaddleOCRVL
|
|
|
|
# os.environ["MACA_VISIBLE_DEVICES"] = "0"
|
|
# os.environ["FD_MOE_BACKEND"] = "cutlass"
|
|
# os.environ["PADDLE_XCCL_BACKEND"] = "metax_gpu"
|
|
# os.environ["FLAGS_weight_only_linear_arch"] = "80"
|
|
# os.environ["FD_METAX_KVCACHE_MEM"] = "8"
|
|
# os.environ["ENABLE_V1_KVCACHE_SCHEDULER"] = "1"
|
|
# os.environ["FD_ENC_DEC_BLOCK_NUM"] = "2"
|
|
# os.environ["FD_SAMPLING_CLASS"] = "rejection"
|
|
# os.environ["PADDLE_PDX_DISABLE_DEV_MODEL_WL"] = "true"
|
|
|
|
|
|
class TestPaddleOCR(unittest.TestCase):
|
|
@classmethod
|
|
def setUpClass(cls):
|
|
"""Class-level setup that runs once before all tests."""
|
|
cls.set_config()
|
|
|
|
# ===============================
|
|
# 基于 FastDeploy Server 推理
|
|
# ===============================
|
|
cls.pipeline = PaddleOCRVL(
|
|
vl_rec_backend="fastdeploy-server",
|
|
vl_rec_server_url="http://127.0.0.1:8118/v1",
|
|
# device=os.getenv("PADDLE_XCCL_BACKEND", "metax_gpu"),
|
|
layout_detection_model_name="PP-DocLayoutV3",
|
|
layout_detection_model_dir=cls.model_root_path + "/PP-DocLayoutV3/",
|
|
vl_rec_model_name="PaddleOCR-VL-1.5-0.9B",
|
|
vl_rec_model_dir=cls.model_root_path + "/PaddleOCR-VL-1.5/",
|
|
)
|
|
|
|
# ===============================
|
|
# 可选功能（按需开启）
|
|
# ===============================
|
|
|
|
# 是否启用文档方向分类模型
|
|
# pipeline = PaddleOCRVL(use_doc_orientation_classify=True)
|
|
|
|
# 是否启用文本图像矫正模块
|
|
# pipeline = PaddleOCRVL(use_doc_unwarping=True)
|
|
|
|
# 是否关闭版面区域检测排序模块
|
|
# pipeline = PaddleOCRVL(use_layout_detection=False)
|
|
|
|
@classmethod
|
|
def set_config(cls):
|
|
"""Set the configuration parameters for the test."""
|
|
|
|
cls.model_root_path = os.getenv(
|
|
"MODEL_ROOT_PATH", "/workspace/models/modelscope.hub.metax-tech.com/models/PaddlePaddle"
|
|
)
|
|
cls.input_image_path = os.getenv("INPUT_IMAGE_PATH", "paddleocr_vl_demo.png")
|
|
cls.ocr_outputs_path = os.getenv("OCR_OUTPUTS_PATH", "./paddle_ocr_outputs")
|
|
|
|
cls.image_content_keywords = [
|
|
"助力双方交往 搭建友谊桥梁",
|
|
"学好中文，我们的未来不是梦",
|
|
"在中国学习的经历让我看到更广阔的世界",
|
|
"共同向世界展示非洲和亚洲的灿烂文明",
|
|
"中厄两国人文交流不断深化，互利合作的民意基础日益深厚",
|
|
"这是我人生的重要一步，自此我拥有了一双坚固的鞋子，赋予我穿越荆棘的力量",
|
|
"学习彼此的语言和文化，将帮助厄中两国人民更好地理解彼此，助力双方交往，搭建友谊桥梁",
|
|
"中国文化博大精深，我希望我的学生们能够通过中文歌曲更好地理解中国文化",
|
|
]
|
|
|
|
def test_image_parse(self):
|
|
# ===============================
|
|
# 执行预测
|
|
# ===============================
|
|
outputs = self.pipeline.predict(self.input_image_path)
|
|
image_all_content = ""
|
|
|
|
# def extract_contents(contents: list[str]) -> str:
|
|
# text=""
|
|
# for content in contents:
|
|
# text = text + content.split("content:")[1].split("#")[0]
|
|
# return text
|
|
|
|
# ===============================
|
|
# 保存和打印结果
|
|
# ===============================
|
|
for res in outputs:
|
|
# res.print() # 打印结构化输出
|
|
# image_all_content=extract_contents(str(res["parsing_res_list"]))
|
|
image_all_content = str(res["parsing_res_list"])
|
|
# res.save_to_json(save_path=self.ocr_outputs_path) # 保存 JSON 结果
|
|
res.save_to_markdown(save_path=self.ocr_outputs_path) # 保存 Markdown 结果
|
|
|
|
assert all(keyword in image_all_content for keyword in self.image_content_keywords)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|
|
|
|
# model_root_path=os.getenv("MODEL_ROOT_PATH", "/workspace/models/modelscope.hub.metax-tech.com/models/PaddlePaddle")
|
|
# input_image_path=os.getenv("INPUT_IMAGE_PATH", "paddleocr_vl_demo.png")
|
|
# ocr_outputs_path=os.getenv("OCR_OUTPUTS_PATH", "./paddle_ocr_outputs")
|
|
|
|
# pipeline = PaddleOCRVL(
|
|
# vl_rec_backend="fastdeploy-server",
|
|
# vl_rec_server_url="http://127.0.0.1:8118/v1",
|
|
# device="metax_gpu",
|
|
|
|
# layout_detection_model_name="PP-DocLayoutV3",
|
|
# layout_detection_model_dir=model_root_path+"/PP-DocLayoutV3/",
|
|
|
|
# vl_rec_model_name="PaddleOCR-VL-1.5-0.9B",
|
|
# vl_rec_model_dir=model_root_path+"/PaddleOCR-VL-1.5/",
|
|
# )
|
|
|
|
# # ===============================
|
|
# # 可选功能（按需开启）
|
|
# # ===============================
|
|
|
|
# # 是否启用文档方向分类模型
|
|
# # pipeline = PaddleOCRVL(use_doc_orientation_classify=True)
|
|
|
|
# # 是否启用文本图像矫正模块
|
|
# # pipeline = PaddleOCRVL(use_doc_unwarping=True)
|
|
|
|
# # 是否关闭版面区域检测排序模块
|
|
# # pipeline = PaddleOCRVL(use_layout_detection=False)
|
|
|
|
# # ===============================
|
|
# # 执行预测
|
|
# # ===============================
|
|
|
|
# outputs = pipeline.predict(input_image_path)
|
|
|
|
# # ===============================
|
|
# # 保存和打印结果
|
|
# # ===============================
|
|
|
|
# for res in outputs:
|
|
# # res.print() # 打印结构化输出
|
|
# # res.save_to_json(save_path=ocr_outputs_path) # 保存 JSON 结果
|
|
# res.save_to_markdown(save_path=ocr_outputs_path) # 保存 Markdown 结果
|