8 PM – Advanced AI/ML – Joel

Today we did:

1. We continued working on the combined YOLO detection and discussed the datasets for the personal projects.

Homework:

1. If you have already locked in your dataset with me last lesson (Russell and Adi), go ahead and train your model in Colab (or fix your dataset format if necessary), and make sure to evaluate it using this code:
2. metrics = model.val()

   print(metrics.box.map50) #mAP@0.5

   print(metrics.box.map) #mAP@0.5:0.95

   print(metrics.box.precision)

   print(metrics.box.recall)

   #mAP@0.5 (0.6 and up is already good)

   #mAP@0.5:0.95 (0.5 and up is already good)

   #Precision (0.7-0.8 is already good)

   #Recall (0.7-0.8)
3. If you have not locked in your dataset (Joshua and Aiden), make sure to check it in with me early in the week via email, so I can help you confirm whether to use it or not. Once it’s confirmed, go ahead to training the model in Colab as well.
4. If you have any questions, feel free to ask me at joel@ayclogic.com.