I have an error when converting the Yolo model in Ubuntu 22.04 cuda 12.4 environment. Because I have changed the network structure of the Yolo model, onnx is the tool provided by Yolo to complete the conversion. However, the following error is prompted during the conversion as shown in the picture.
Then the yolov11m.yaml of networks has also been modified as follows
base:
- base/yolov8.yaml
postprocessing:
device_pre_post_layers:
nms: true
hpp: true
network:
network_name: yolov11m
paths:
network_path: - /root/Training_data/ultralytics/th0922/yolo11m_peoson_test_2/weights/best.onnx
alls_script: yolov11m.alls
url: https://hailo-model-zoo.s3.eu-west-2.amazonaws.com/ObjectDetection/Detection-COCO/yolo/yolov11m/2024-10-02/yolo11m.zip
parser:
nodes: - null
info:
task: object detection
input_shape: 896x896x3
output_shape: 2x5x100
operations: 93.5G
parameters: 22.3M
framework: pytorch
training_data: /root/split_dataset_th/train/images
validation_data: /root/split_dataset_th/val/images
eval_metric: mAP
full_precision_result: 51.1
source: GitHub - ultralytics/ultralytics: Ultralytics YOLO 🚀
license_url: ultralytics/LICENSE at main · ultralytics/ultralytics · GitHub
license_name: AGPL-3.0
If you need more information to solve my problem, I will do my best.