Why YOLOv5 not training?

Help
#1

Question/Issue: Hi, I don’t understand why my YOLOv5 AI cannot train. The target device is OpenMV Cam H7 Plus. Below I write the training error:

Creating job… OK (ID: 21106135)

Scheduling job in cluster…
Container image pulled!
Job started
Scheduling job in cluster…
Container image pulled!
Job started
Application exited with code 1
Transforming Edge Impulse data format into something compatible with YOLOv5
Traceback (most recent call last):
File “/usr/local/lib/python3.8/dist-packages/PIL/Image.py”, line 2813, in fromarray
mode, rawmode = _fromarray_typemap[typekey]
KeyError: ((1, 1, 1), ‘|u1’)

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File “extract_dataset.py”, line 80, in
convert(X=X_train, Y=Y_train, category=‘train’)
File “extract_dataset.py”, line 51, in convert
im = Image.fromarray(raw_img_data)
File “/usr/local/lib/python3.8/dist-packages/PIL/Image.py”, line 2815, in fromarray
raise TypeError(“Cannot handle this data type: %s, %s” % typekey) from e
TypeError: Cannot handle this data type: (1, 1, 1), |u1
Job failed (see above)

**Project ID:**423785

#2

Hi @Pieru98

Why are you using the Renesas specific Yolo v5 block when the target device is OpenMV H7 Plus?

I think you used the incorrect image processing yolo v5 block.

Best

Eoin

#3

I’m running into the same issue, it trains fine with FOMO, but with YOLO V5 it throws the same error. I want to use YOLO for the bounding boxes. I plan on deploying it to the Seeed Vision AI

#4

Hi @asbecka,

Could you provide your project ID so we could try to replicate the issue?