Explaining food image predictions
You have a model
that classifies food, and your task is to use LIME to identify the regions the model mostly focuses on while making its prediction for the image
below.
The model
responsible for predictions, the model_predict
function and the sample image
shown below are pre-loaded for you.
This exercise is part of the course
Explainable AI in Python
Exercise instructions
- Create a LIME image explainer named
explainer
. - Generate an
explanation
for themodel
's prediction on the givenimage
. - Extract the areas of interest from the
image
based on themodel
's explanation.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
from lime import lime_image
np.random.seed(10)
# Create a LIME explainer
explainer = ____
# Generate the explanation
explanation = explainer.____(____, ____, hide_color=0, num_samples=50)
# Display the explanation
temp, _ = explanation.____(____, ____)
plt.imshow(temp)
plt.title('LIME Explanation')
plt.axis('off')
plt.show()