Replies: 2 comments 2 replies
-
@mo-atiff hi there! 👋 Thank you for reaching out and providing the details of your issue. Let's address this step-by-step:
If the issue persists after these steps, please provide a minimum reproducible example that includes your Looking forward to your update! 😊 |
Beta Was this translation helpful? Give feedback.
-
hey @glenn-jocher The same code works now no issues, I actually trained the v8 model on coco.yaml dataset which got completed with no issues, then I tried the same script with data.yaml, Interestingly it worked no such errors. I actually trained v8 model yesterday but then today I saw a an upgrade of ultralytics and did upgrade(v8.2.39) soon after which I had ran into such issues ! I did remove the "device" parameter which doesn't matter ig but anyway I did Thanks for your reply really appreciate it |
Beta Was this translation helpful? Give feedback.
-
Just updated ultralytics to v8.2.39 and ran into errors while training a simple custom object detection model :
`
from ultralytics import YOLO
import cv2
model = YOLO("yolov8n.pt")
results = model.train(data = "data.yaml", epochs=10, resume=True, device='cpu') # I'm using absolute path for data parameter
`
ERROR :
SyntaxError: 'v5loader' is not a valid YOLO argument.
'fl_gamma' is not a valid YOLO argument.
'image_weights' is not a valid YOLO argument.
I do also get such warnings :
WARNING⚠️ 'hide_labels' is deprecated and will be removed in 'ultralytics 8.399999999999999' in the future. Please use 'show_labels' instead.⚠️ 'hide_conf' is deprecated and will be removed in 'ultralytics 8.399999999999999' in the future. Please use 'show_conf' instead.⚠️ 'line_thickness' is deprecated and will be removed in 'ultralytics 8.399999999999999' in the future. Please use 'line_width' instead.
WARNING
WARNING
Beta Was this translation helpful? Give feedback.
All reactions