支持多模型进行推理吗？

[复制链接] · 发表于 2024-6-28 18:23:28

1.我在部署的时候，想进行多个模型一起推理，但是出现了问题，单个模型推理正常.(6, 1001)
--> Load RKNN model LSTM_CNN_35_bs1.rknn
done
[RKNN] Can not find libdrm.so
librknn_runtime version 1.7.3 (2e55827 build: 2022-08-25 10:45:32 base: 1131)
E [op_optimize:428]CONCAT, uid 8 must have same quantize parameter!
E [op_optimize:428]CONCAT, uid 13 must have same quantize parameter!
E [op_optimize:428]CONCAT, uid 13 must have same quantize parameter!
E [op_optimize:428]CONCAT, uid 8 must have same quantize parameter!

(10:0) : error : Error(0,10) : Cannot find the header file cl_viv_vx_ext.h.
(255:0) : error : Error(0,255) : Cannot find the header file cl_viv_vx_ext.h.
(27:0) : error : undefined identifier: 'COPY'
(55:0) : error : undefined identifier: 'COPY'
(258:0) : error : syntax error at 'VXC_512Bits'

ERROR: Failed to compile vx shader. (error: FFFFFFFF)
E [_gpu_register:476]Build program fail.
E [rknn_kernel_create_node:632]Register client kernel com.vivantecorp.extension.evis.gather_F16toF16 fail with -1.

done
[1]
done
done
[RKNN] Can not find libdrm.so
librknn_runtime version 1.7.3 (2e55827 build: 2022-08-25 10:45:32 base: 1131)
E [rknn_CreateTensorFromData:1275]Create tensor from data fail.
Segmentation fault

只看该作者 · 发表于 2024-6-29 10:13:56

1. 支持多模型推理。
2. 多模型每个都要有一个独立的context
3. 先把rknn版本更新到最新，避免已知问题