Bounding boxes prediction
You have trained an object recognition model. Now, you want to use to generate bounding boxes and classifications for a test image.
The model is available as model and it's already in the evaluation mode. The test_image is also available, and torch and torch.nn as nn have been imported.
This exercise is part of the course
Deep Learning for Images with PyTorch
Exercise instructions
- Get the model's output for the
test_imageand assign it tooutput. - Extract the predicted bounding boxes from the output and assign them to
boxes. - Extract the predicted scores from the output and assign them to
scores.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Get model's prediction
with torch.no_grad():
output = ____
# Extract boxes from the output
boxes = ____
# Extract scores from the output
scores = ____
print(boxes, scores)