implementation of Yolov8 & evaluation metrics

When installing YOLOv8, what package manager is recommended?

• 2. conda

• • pip

• • Both conda and pip are equally suitable.

• • It depends on your operating system and preferences.

Which file in the YOLOv8 repository contains the main training script?

• • train.py

• • main.py

• • yolov8.py

• • It depends on the specific implementation you choose.

To download a pre-trained YOLOv8 model, you can use the command:

• The option to download models directly is not available.

• pip install yolov8_models

• python download_model.py <model_name>

• yolov8 download <model_name>

Which hyperparameter controls the learning rate during training?

• 
  

• • Batch size

• • Momentum

• • Learning rate

• • Epochs

To save checkpoints of your model during training, you can use the --save flag with what value?

• Number of epochs

• Validation loss

• Interval in epochs

• Directory name

The most common metric for object detection in YOLOv8 is:

• Recall

• Accuracy

• Intersection over Union (IoU)

• Mean Average Precision (mAP)

mAP considers object localization and classification accuracy. It is calculated by averaging the Average Precision (AP) across different:

• (a) IoU thresholds

• (b) Object classes

• (c) Image sizes

• (d) Both (a) and (b)

Precision measures the proportion of true positives among all predicted positives for a specific object class.

• True

• False

While mAP is a good overall metric, it might not be suitable for tasks where the detection of small objects is crucial.

• True

• False

Recall measures the proportion of true positives that are correctly identified out of all actual positives for a specific object class.

• True

• False