Mean Average Precision (mAP) | Lecture 34 (Part 1) | Applied Deep Learning

Поділитися
Вставка
  • Опубліковано 17 гру 2024

КОМЕНТАРІ •

  • @andrewmoskovtsev3222
    @andrewmoskovtsev3222 Рік тому

    I think it's one of the best explanations of mAP which I have seen on the internet. thanks a lot!

  • @rafaelsqueiroz
    @rafaelsqueiroz 2 роки тому +1

    Very explanatory class. Thank you!
    Just a subtle contribution: in the video, the mAP is calculated considering a single threshold (lets say, 0.5), so it would be mAP@0.5. In papers it's common to see also a metric called mAP@0.5:0.05:0.95, which is the average of the calculations of mAP@0.5, mAP@0.55, mAP@0.60, ... , mAP@0.95.
    One more time thank you for this awesome content!

  • @Cristyioan2012
    @Cristyioan2012 3 роки тому +1

    Great work!

  • @aramkarakhanyan4224
    @aramkarakhanyan4224 Рік тому

    What is the method of selecting these red bounding boxes ? Is there any step between selective search which suggested 2000 boxes and counting mean average precision ?As i guess IOU is counted after suggestion of bounding boxes .Which algorithm is responsible for confidence scores,how it assessed ? Thanks

  • @alvaro-tech3449
    @alvaro-tech3449 2 роки тому +1

    Thank you