latest/version3.x/pipeline_usage/PaddleOCR-VL-Huawei-Ascend-NPU #17806
Replies: 4 comments 1 reply
-
|
NPU下不支持PaddlePaddle推理方式是指CLI和python代码两者都不支持吗。 |
Beta Was this translation helpful? Give feedback.
-
|
使用python代码直接调用PaddleOCR-VL的时候batch_size总是为1,但是没有发现地方能够调整(使用vllm调用方式没问题) |
Beta Was this translation helpful? Give feedback.
-
|
参照高性能服务化部署方案,发现paddleocr-vl-tritonserver没有调PP-DocLayoutV3这个模型,只是做了一层请求的转发,看起来这个容器意义不大了,还是只想利用tritonserver的批合并这种请求调度能力? |
Beta Was this translation helpful? Give feedback.
-
|
请问怎么在离线设备上进行高性能服务化部署??给的这个教程应该是在线设备上部署的https://github.com/PaddlePaddle/PaddleOCR/blob/main/deploy/paddleocr_vl_docker/hps/README.md |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
latest/version3.x/pipeline_usage/PaddleOCR-VL-Huawei-Ascend-NPU
Awesome multilingual OCR toolkits based on PaddlePaddle (practical ultra lightweight OCR system, support 80+ languages recognition, provide data annotation and synthesis tools, support training and deployment among server, mobile, embedded and IoT devices)
https://www.paddleocr.ai/latest/version3.x/pipeline_usage/PaddleOCR-VL-Huawei-Ascend-NPU.html
Beta Was this translation helpful? Give feedback.
All reactions