Dear @paulinpaloalto ,
Please your advice to cross this issue PFB:
It’s not clear that it’s a size issue, but you’re right that it seems to be an issue with the last minibatch. That assertion is checking some of the actual values in the last minibatch. So check carefully all the index values you are using in that section of the code. You passed the value check for the first minibatch.
1 Like
Also note that you already passed the assertions for the shapes of all the minibatches, including the last one.
1 Like
Hi Paul,
Can you please guiding me on how to troubleshooting to find where the issue located,
and how to check the indexed used ? I tried to print the matrix to see if can find any abnormality but no luck .PFB:
One approach would be to try printing the exact values that the assertion is comparing and see if you can find them in your complete printout. That might be a clue. One common mistake is to use the wrong starting index on the first axis for the last minibatch.
Actually looking at the values you show, notice that the last two entries shown are the same. In other words the last one is a duplicate of the second to last one. I don’t think that’s the way it supposed look. So how could that happen.
1 Like
Hi Paul,
because i used print(mini_batches[0][0][0]), could you please share the instruction on private msg ! this exercise took a lot of time,Thank you for your advanced support