search

ultralytics / yolov5

YOLOv5 🚀 in PyTorch > ONNX > CoreML > TFLite
https://docs.ultralytics.com
GNU Affero General Public License v3.0
50.34k stars 16.25k forks source link

Basic Questions Regarding Yolov5 and need confirmation from @glenn-jocher #3832

Closed akshat-suwalka closed 3 years ago

akshat-suwalka commented 3 years ago

❔Question

Simply using all the default setting of YOLOv5s

  1. What are the loss functions used for box, objectness, classification? Ans => What I found on searching over internet that :- CIOU Loss is used as the loss of bounding box in yolov5 and IoU is used for NMS Binary Cross-Entropy with Logits Loss function from PyTorch for loss calculation of class probability and object score.

Is the above answer fully correct for both training and testing case, Please if there is any correction let me know??

  1. Is precision and recall defined for detection or both detection and classification? Ans => I am not sure @glenn-jocher Can you please enlighten me?

  2. You have reported values for precision and recall. What are the values of recognition accuracy? Ans => Any comments?

  3. How is mean average precision computed? How do you get the P@0.95 values? Ans => I know it but if you can answer then it will increase my understanding more.... Can you please answer?

  4. Was cross validation used? Ans => Not sure....any comments @glenn-jocher ?

  5. Can you please tell computational time results like using default setting how much time the YoloV5small is taking...?

Additional context

glenn-jocher commented 3 years ago

@akshat-suwalka for details on loss function see loss.py: https://github.com/ultralytics/yolov5/blob/master/utils/loss.py

For details on metrics see metrics.py: https://github.com/ultralytics/yolov5/blob/master/utils/metrics.py

Speeds are clearly displayed in README.

github-actions[bot] commented 3 years ago

👋 Hello, this issue has been automatically marked as stale because it has not had recent activity. Please note it will be closed if no further activity occurs.

Access additional YOLOv5 🚀 resources:

Access additional Ultralytics ⚡ resources:

Feel free to inform us of any other issues you discover or feature requests that come to mind in the future. Pull Requests (PRs) are also always welcomed!

Thank you for your contributions to YOLOv5 🚀 and Vision AI ⭐!