PaddleX/latest/pipeline_usage/tutorials/information_extraction_pipelines/document_scene_information_extraction_v3

[giscus[bot]]

PaddleX/latest/pipeline_usage/tutorials/information_extraction_pipelines/document_scene_information_extraction_v3

https://paddlepaddle.github.io/PaddleX/latest/pipeline_usage/tutorials/information_extraction_pipelines/document_scene_information_extraction_v3.html

[yuanmengwei]

请问，这个怎么使用自己部署的大模型呢

[CrazyDJDave]

visual_predict_res=pipeline.visual_predict(
input="./test.jpg",
use_doc_orientation_classify=False,
use_doc_unwarping=False,
use_common_ocr=True,
use_seal_recognition=False,
use_table_recognition=False,
)
为什么禁用seal_recognition之后，运行代码在输出里面还会看到有使用seal_server_rec模型
Using official model (PP-OCRv4_server_seal_det), the model files will be automatically downloaded
Using official model (PP-OCRv4_server_rec), the model files will be automatically downloaded

[GY13648114184]

V3可以调用本地部署的大模型吗？还有就是ernie3.5可以本地部署吗？如果可以调用本地部署那么它们二者之间怎么使用呢？

[baoyuchen]

Traceback (most recent call last):
File "f:\taitanworkplace\testcode\chatocrv4\demo1.py", line 34, in
for res in visual_predict_res:
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\pipelines\pp_chatocr\pipeline_v3.py", line 275, in visual_predict
self.inintial_visual_predictor(self.config)
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\pipelines\pp_chatocr\pipeline_v3.py", line 103, in inintial_visual_predictor
self.layout_parsing_pipeline = self.create_pipeline(layout_parsing_config)
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\pipelines\base.py", line 134, in create_pipeline
pipeline = create_pipeline(
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\pipelines_init_.py", line 165, in create_pipeline
pipeline = BasePipeline.get(pipeline_name)(
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\utils\deps.py", line 195, in _wrapper
return old_init_func(self, *args, **kwargs)
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\pipelines_parallel.py", line 103, in init
self._pipeline = self._create_internal_pipeline(config, self.device)
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\pipelines_parallel.py", line 158, in _create_internal_pipeline
return self.pipeline_cls(
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\pipelines\layout_parsing\pipeline.py", line 62, in init
self.inintial_predictor(config)
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\pipelines\layout_parsing\pipeline.py", line 91, in inintial_predictor
self.doc_preprocessor_pipeline = self.create_pipeline(
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\pipelines\base.py", line 134, in create_pipeline
pipeline = create_pipeline(
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\pipelines_init.py", line 165, in create_pipeline
pipeline = BasePipeline.get(pipeline_name)(
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\utils\deps.py", line 195, in _wrapper
return old_init_func(self, *args, **kwargs)
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\pipelines_parallel.py", line 103, in init
self._pipeline = self._create_internal_pipeline(config, self.device)
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\pipelines_parallel.py", line 158, in _create_internal_pipeline
return self.pipeline_cls(
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\pipelines\doc_preprocessor\pipeline.py", line 67, in init
self.doc_ori_classify_model = self.create_model(doc_ori_classify_config)
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\pipelines\base.py", line 99, in create_model
model = create_predictor(
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\models_init.py", line 77, in create_predictor
return BasePredictor.get(model_name)(
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\models\image_classification\predictor.py", line 49, in init
self.preprocessors, self.infer, self.postprocessors = self._build()
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\models\image_classification\predictor.py", line 82, in _build
infer = self.create_static_infer()
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\models\base\predictor\base_predictor.py", line 240, in create_static_infer
return PaddleInfer(self.model_dir, self.MODEL_FILE_PREFIX, self._pp_option)
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\models\common\static_infer.py", line 274, in init
self.predictor = self._create()
File "f:\taitanworkplace\testcode\chatocrv4\paddlex\inference\models\common\static_infer.py", line 359, in _create
config.set_optimization_level(3)
AttributeError: 'paddle.fluid.libpaddle.AnalysisConfig' object has no attribute 'set_optimization_level'应该出一个对应的版本说明文档和代码都不一致了

[Echo-Forever]

请问在设计api的时候，为什么考虑是base64 来传递文件，而不是文件流的方式