Check the train images.
For input images with index >= 400, I am getting random pixels on displaying them. Images before that are normal. Similar issue is in the test images.
Kindly confirm this
Pokemon Classification image error
Yes images till 328 are correct. Use them only. After 328 there is random noise only no need to take include them in training data.
Hope this helped you.
what about testing data
Yes in test data images are correct till 61, after that there is random noise only. Try to make predictions for first 62 and after that the predictions may be random.
Or Try to implement your algorithm on your own test and train data.
Hope this helped.
The webapp is unable to calculate the score for the submission
Make sure you are submitting the csv containing same number of rows as in test data and your format must abide with the sample submission file.
Happy Learning.
I hope I’ve cleared your doubt. I ask you to please rate your experience here
Your feedback is very important. It helps us improve our platform and hence provide you
the learning experience you deserve.
On the off chance, you still have some questions or not find the answers satisfactory, you may reopen
the doubt.