amd / ryzen-ai-documentation Goto Github PK

I have tried to run the example\resnet50 on my 7840HS notebook. I followed this tutorial to install and check the voe environment. However, I have this ERROR MESS when trying to run

python predict.py --ep ipu

C:\Dev\miniconda3\envs\voe\lib\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py:65: UserWarning: Specified provider 'VitisAIExecutionProvider' is not in available provider names.Available providers: 'CPUExecutionProvider'

I thought it was the ort version problem. But I have no idea how to solve that.

(I didn't install VS2019, instead, I installed the VC++ 2015-2022 redistribution and I think this should not be wrong).

Hi! Thanks for your help I successfully ran the example resnet50. However, it failed when I wanted to change the batch size from 1 to another number.

Here is my log:

F20231016 17:39:21.145711  1928 schedule.cpp:272] Check failed: to_tensor->get_shape()[0] >= onnx_batch (1 vs. 16) to data batch size must be greater than or equal to onnx model dpu subgraph batch size.
2023-10-16 17:39:21.1604509 [E:onnxruntime:, sequential_executor.cc:514 onnxruntime::ExecuteKernel] Non-zero status code returned while running vitis_ai_ep_1 node. Name:'VitisAIExecutionProvider_vitis_ai_ep_1_0' Status Message: attrs_create_cif

I tried to remove the dynamic_axes and export a static onnx model with a fixed batch, but nothing changed.

https://github.com/amd/ryzen-ai-documentation/blob/main/docs/inst.rst#install-quantizer
It shows pip install vai_q_onnx-1.14.0-py2.py3-none-any.whl
It should be 1.15.0?