CODEWITHSUNDEEP
validationyoloyolov8custom-trainingultralytics
user24192861
user24192861

Reputation: 1

Validation Result is same as Training Result in YOLOV8

I'm a beginner please spare me, I'm running a YOLOV8 model in Google Colaboratory Environment. I've noticed that after training my model and validating it with the pretrained yolov8 model. Then I've noticed that validation result is same as my training result. Is that even possible? or my code has a problem? I have a limited data a very small data. Training Images : 153 images (training images has been augmented importing imaugh library) Validation & Testing Images: 22 Please refer to the images for reference Training result\

Validation result

Well I've tried the validation code using from the YOLO docs result is still the same.

Upvotes: 0

Views: 434

Answers (1)

Surya Giri
Surya Giri

Reputation: 1

You must have different datasets during training and validation. If you train your model on same data and validate in the same data. You will get the same results. Use this command for training results = model.train(data="coco8.yaml", epochs=100, imgsz=640) and validation_results = model.val(data="coco8.yaml", imgsz=640, batch=16, conf=0.25, iou=0.6, device="0") for validation.

Upvotes: 0

Related Questions