Week 2 Exercise 8 : where's the m_train data?

The comments indicate:
X_train – training set represented by a numpy array of shape (num_px * num_px * 3, m_train)
X_test – test set represented by a numpy array of shape (num_px * num_px * 3, m_test)

if I add print statements before any of my code:
print (str(m_train) + ’ training examples’)
print (str(m_test) + ’ test examples’)
print (str(num_px) + ’ image size’)

print (str(X_train.shape) + ' X_train shape')
print (str(X_test.shape) + ' X_test shape')

I get:
209 training examples
50 test examples
64 image size
(4, 7) X_train shape
(4, 3) X_test shape

So I am told my training data is [64 * 64 * 3, 209] and instead of images, I am given a 4x7 array. Where’s my data? This looks seriously messed up before it even gets to my code, but I see a lot of other questions with people having issues with 4x7 arrays, so I guess I am reading it wrong but I still have no idea. Please help.

The point is that the code we are writing here is supposed to be completely general: it should handle inputs with any number of “features” or “samples”. When they are writing test cases (“unit tests”) for the various “worker” functions in the notebook, it is way easier to debug if they use small examples like 4 x 7. Trust me, debugging with vectors with 12288 entries is no fun.

So there are different kinds and sizes of data and your code needs to handle all of them. You’ll see the full 12288 x 209 X_train matrix soon enough, once you’ve written and debugged all the subroutines you need. Keep in mind as you work on the various functions that it’s a mistake to “hardcode” or make any fixed assumptions about the various dimensions of the input objects.

Thanks! Easy enough once I realized I really only had 7 training examples of an image size 4, but I thought the comments were really misleading.