Hello, I faced an error when running the Yolov6 model in the host-decoding program. The error that happened on my local is looking like this
for 6 shaves likely will yield in better performance
Traceback (most recent call last):
File "main.py", line 179, in <module>
boxes = non_max_suppression(output, conf_thres=conf_thresh, iou_thres=iou_thresh)
File "/home/ubuntu/OAK-D model/gen2-yolo/host-decoding/util/functions.py", line 27, in non_max_suppression
xc = prediction[..., 4] > conf_thres # candidates
IndexError: index 4 is out of bounds for dimension 1 with size 0
Previously, i trained the YoloV6 model on my custom dataset based on this tutorial: luxonis/depthai-ml-trainingblob/master/colab-notebooks/YoloV6_training.ipynb.
After I finished my training process, I got the.pt model and converted it using these tools. https://tools.luxonis.com/
Then, I got zip files with these kinds of files:
Then I tried to run the.blob model with host decoding and got the error above. I can't find the solution for this error even when i refer to this post: https://discuss.luxonis.com/d/2484-problem-using-yolov5-detectionsegmentation-custom-model-on-host/5.
When i check the Onnx model, there are no output only these layers
Can you help me for this problem? Actually, I got this error for the YoloV8 files too.
