atlas 300i pro跑yolov5.onnx模型推理出错

[PeterFive]

环境

• 【FastDeploy版本】： fastdeploy-python ==0.0.0
• 【系统平台】: Linux aarch64(Ubuntu 18.04)
• 【硬件】： atlas 300i pro推理卡 cann6.0.1
• 【编译语言】： Python3.7.5

出现问题的操作流程

• 【模型跑不通】
• • 先执行examples下的部署示例，使用examples提供的模型，可以执行成功
• • examples下的示例python infer.py --model yolov5s_infer --image 000000014439.jpg --device ascend可以运行，换成了paddle提供的yolov5的onnx文件(yolov5n.onnx)，代码报错
• • 这是测试代码

import fastdeploy as fd
import cv2

option = fd.RuntimeOption()
option.use_ascend()
model = fd.vision.detection.YOLOv5(
    "./yolov5n.onnx",
    params_file="",
    runtime_option=option,
    model_format=fd.ModelFormat.ONNX)
image = './1.jpg'
im = cv2.imread(image)
result = model.predict(im, conf_threshold=0.25, nms_iou_threshold=0.5)
vis_im = fd.vision.vis_detection(im, result)
cv2.imwrite("visualized_result_monkey.jpg", vis_im)
print("Visualized result save in ./visualized_result_monkey.jpg")

--报错信息

[ERROR] fastdeploy/fastdeploy_model.cc(385)::CreateASCENDBackend        There's no valid ascend backends for model: yolov5
[ERROR] fastdeploy/vision/detection/contrib/yolov5/yolov5.cc(45)::Initialize    Failed to initialize fastdeploy backend.
Traceback (most recent call last):
  File "test.py", line 10, in <module>
    model_format=fd.ModelFormat.ONNX)
  File "/usr/local/python3.7.5/lib/python3.7/site-packages/fastdeploy/vision/detection/contrib/yolov5.py", line 184, in __init__
    assert self.initialized, "YOLOv5 initialize failed."
AssertionError: YOLOv5 initialize failed.

[juncaipeng]

这个后端只支持yolov5s_infer模型，没有支持yolov5n

[PeterFive]

这个后端只支持yolov5s_infer模型，没有支持yolov5n

抱歉我不是很明白，是不支持onnx模型(我需要将ONNX转成paddle inference model才可用)，还是说支持yolov5s.onnx不支持yolov5n.onnx