PaddleX/latest/pipeline_usage/tutorials/ocr_pipelines/PaddleOCR-VL

[giscus[bot]]

PaddleX/latest/pipeline_usage/tutorials/ocr_pipelines/PaddleOCR-VL

https://paddlepaddle.github.io/PaddleX/latest/pipeline_usage/tutorials/ocr_pipelines/PaddleOCR-VL.html

[HuangShiKaiKai]

PaddleOCR-VL产线需要多少显存呢？

[ccyycc1994]

https://aistudio.baidu.com/application/detail/98365 相同图片无法对齐官方应用结果，是否可以提供应用部署paddlex yaml和请求paddlex server的参数

[oujiamingya]

请问有没有什么方法能够把识别出来的一行行文本智能的归为段落级别的？

[sfeng1996]

paddlex --serve --pipeline ./PaddleOCR-VL.yaml --port 8080 --device dcu
Creating model: ('PP-DocLayoutV2', '/mnt/model/PaddleOCR-VL/PP-DocLayoutV2')
Skip checking if model is supported on device because the flag PADDLE_PDX_DISABLE_DEV_MODEL_WL has been set.
Creating model: ('PaddleOCR-VL-0.9B', '/mnt/model/PaddleOCR-VL')
Failed to create the pipeline
Traceback (most recent call last):
File "/usr/local/lib/python3.10/dist-packages/paddlex/paddlex_cli.py", line 488, in serve
pipeline = create_pipeline(
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/pipelines/init.py", line 167, in create_pipeline
pipeline = BasePipeline.get(pipeline_name)(
File "/usr/local/lib/python3.10/dist-packages/paddlex/utils/deps.py", line 206, in _wrapper
return old_init_func(self, *args, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/pipelines/_parallel.py", line 103, in init
self._pipeline = self._create_internal_pipeline(config, self.device)
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/pipelines/_parallel.py", line 158, in _create_internal_pipeline
return self._pipeline_cls(
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/pipelines/paddleocr_vl/pipeline.py", line 127, in init
self.vl_rec_model = self.create_model(vl_rec_config)
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/pipelines/base.py", line 106, in create_model
model = create_predictor(
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/init.py", line 87, in create_predictor
return BasePredictor.get(model_name)(
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/doc_vlm/predictor.py", line 59, in init
self.infer, self.processor = self._build(**kwargs)
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/doc_vlm/predictor.py", line 98, in _build
from .modeling import (
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/doc_vlm/modeling/init.py", line 16, in
from .paddleocr_vl import PaddleOCRVLForConditionalGeneration
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/doc_vlm/modeling/paddleocr_vl/init.py", line 15, in
from ._paddleocr_vl import PaddleOCRVLForConditionalGeneration
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/doc_vlm/modeling/paddleocr_vl/_paddleocr_vl.py", line 50, in
from ._ernie import Ernie4_5Model, Ernie4_5PretrainedModel
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/doc_vlm/modeling/paddleocr_vl/_ernie.py", line 60, in
from ._fusion_ops import (
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/doc_vlm/modeling/paddleocr_vl/_fusion_ops/init.py", line 19, in
from paddle.incubate.nn.functional import fused_rms_norm_ext
ImportError: cannot import name 'fused_rms_norm_ext' from 'paddle.incubate.nn.functional' (/usr/local/lib/python3.10/dist-packages/paddle/incubate/nn/functional/init.py)
root@dcu-k100-paddle-857b554754-qv9xt:/my_configs# export PADDLE_PDX_DISABLE_DEV_MODEL_WL=true
root@dcu-k100-paddle-857b554754-qv9xt:/my_configs# paddlex --serve --pipeline ./PaddleOCR-VL.yaml --port 8080 --device dcu
Creating model: ('PP-DocLayoutV2', '/mnt/model/PaddleOCR-VL/PP-DocLayoutV2')
Skip checking if model is supported on device because the flag PADDLE_PDX_DISABLE_DEV_MODEL_WL has been set.
Creating model: ('PaddleOCR-VL-0.9B', '/mnt/model/PaddleOCR-VL')
Failed to create the pipeline
Traceback (most recent call last):
File "/usr/local/lib/python3.10/dist-packages/paddlex/paddlex_cli.py", line 488, in serve
pipeline = create_pipeline(
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/pipelines/init.py", line 167, in create_pipeline
pipeline = BasePipeline.get(pipeline_name)(
File "/usr/local/lib/python3.10/dist-packages/paddlex/utils/deps.py", line 206, in _wrapper
return old_init_func(self, *args, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/pipelines/_parallel.py", line 103, in init
self._pipeline = self._create_internal_pipeline(config, self.device)
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/pipelines/_parallel.py", line 158, in _create_internal_pipeline
return self._pipeline_cls(
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/pipelines/paddleocr_vl/pipeline.py", line 127, in init
self.vl_rec_model = self.create_model(vl_rec_config)
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/pipelines/base.py", line 106, in create_model
model = create_predictor(
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/init.py", line 87, in create_predictor
return BasePredictor.get(model_name)(
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/doc_vlm/predictor.py", line 59, in init
self.infer, self.processor = self._build(**kwargs)
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/doc_vlm/predictor.py", line 98, in _build
from .modeling import (
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/doc_vlm/modeling/init.py", line 16, in
from .paddleocr_vl import PaddleOCRVLForConditionalGeneration
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/doc_vlm/modeling/paddleocr_vl/init.py", line 15, in
from ._paddleocr_vl import PaddleOCRVLForConditionalGeneration
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/doc_vlm/modeling/paddleocr_vl/_paddleocr_vl.py", line 50, in
from ._ernie import Ernie4_5Model, Ernie4_5PretrainedModel
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/doc_vlm/modeling/paddleocr_vl/_ernie.py", line 60, in
from ._fusion_ops import (
File "/usr/local/lib/python3.10/dist-packages/paddlex/inference/models/doc_vlm/modeling/paddleocr_vl/_fusion_ops/init.py", line 19, in
from paddle.incubate.nn.functional import fused_rms_norm_ext
ImportError: cannot import name 'fused_rms_norm_ext' from 'paddle.incubate.nn.functional' (/usr/local/lib/python3.10/dist-packages/paddle/incubate/nn/functional/init.py)
在海光 DCU 上使用 paddlex 运行 PaddleOCR-VL 报错，请问怎么解决

[lilong1988]

如何对PaddleOCR-VL进行微调，是否有相关教程

[F-people]

您好，我看此次VL模型舍弃后续阅读顺序逻辑，是因为PP-DocLayoutV2模型在训练时包含了文档元素的空间布局和阅读顺序关系么？

[Echo-Forever]

PaddleOCR-VL 对手写体的识别 效果不佳，请问如何达到 OCR产线手写体的识别效果