
In your wed-mar4-catvsdog kaggle notebook, do the following:
# hw: # Ensure your individual custom dataset images have capitalization of the first letter of the file name # run prediction on each image in ur custom dataset then see if there are any errors, and print out the number of correct guesses # you can use the labeling function to test correctness. so something like: if predicted_ans == true_ans: corrects += 1 # hint: run a for loop over the list of your custom images
You can reach me at ddjapri@ayclogic.com.
All class notes can be found here.