PaddleX/latest/pipeline_usage/pipeline_develop_guide

[giscus[bot]]

PaddleX/latest/pipeline_usage/pipeline_develop_guide

https://paddlepaddle.github.io/PaddleX/latest/pipeline_usage/pipeline_develop_guide.html

[lyj201644070230]

paddlex 加载yaml文件没有layout_parsing_v2这个产线名:
用的官方镜像paddlex:paddlex3.0.0rc0-paddlepaddle3.0.0rc0-gpu-cuda11.8-cudnn8.6-trt8.5.
运行报错：
没有layout_parsing_v2is not registered on BasePipeline，如何解决： Traceback (most recent call last): File "/data8T/yajieliu/seal-PaddleOCR/manage.py", line 5, in from app.api.api_handler import RecommendHandler File "/data8T/yajieliu/seal-PaddleOCR/app/api/api_handler.py", line 3, in from app.main import sealpredict File "/data8T/yajieliu/seal-PaddleOCR/app/main.py", line 6, in from app.conf_ocr.conf import * File "/data8T/yajieliu/seal-PaddleOCR/app/conf_ocr/conf.py", line 7, in pipeline = create_pipeline(pipeline="./app/weights/layout_parsing_v2.yaml") File "/home/hnxt/anaconda3/envs/ppocrx/lib/python3.9/site-packages/paddlex/inference/pipelines/init.py", line 155, in create_pipeline pipeline = BasePipeline.get(pipeline_name)( File "/home/hnxt/anaconda3/envs/ppocrx/lib/python3.9/site-packages/paddlex/utils/subclass_register.py", line 94, in get raise_class_not_found_error(name, cls, all_entities) File "/home/hnxt/anaconda3/envs/ppocrx/lib/python3.9/site-packages/paddlex/utils/errors/others.py", line 126, in raise_class_not_found_error raise ClassNotFoundException(msg) paddlex.utils.errors.others.ClassNotFoundException:layout_parsing_v2` is not registered on BasePipeline.
The registied entities: [doc_preprocessor, OCR, layout_parsing, PP-StructureV3, PP_ChatOCR_Pipeline, PP-ChatOCRv3-doc, PP-ChatOCRv4-doc, image_classification, object_detection, seal_recognition, table_recognition, table_recognition_v2, multilingual_speech_recognition, formula_recognition, image_multilabel_classification, video_classification, video_detection, anomaly_detection, ts_forecast, ts_anomaly_detection, ts_classification, PP-ShiTuV2, face_recognition, AttributeRecPipeline, pedestrian_attribute_recognition, vehicle_attribute_recognition, semantic_segmentation, instance_segmentation, small_object_detection, rotated_object_detection, human_keypoint_detection, open_vocabulary_detection, open_vocabulary_segmentation, 3d_bev_detection]

[songmingjun3]

是否支持自己设计产线和对应的模型？有标准的开发流程吗