IndexError: tuple index out of range by max pooling

Question

I am using keras functional api and I have tensor of shape

phrasing.shape = (?, 553777, 19, 3)

Now I want to take a max feature of unigram and bigram & trigram so the expected shape should be

(?, 553777, 19, 1)

so I added this line

ph = MaxPooling2D ( pool_size=(1,3)  )(phrasing)

and get this error : IndexError: tuple index out of range

Please add all relevant infomation to the question - an error message is not something that should be searched in the comment section — Nico Haase, Feb 19 '18 at 08:26

score 0 · Answer 1 · answered Mar 15 '18 at 18:14

0

Your data is a dimension too high. I had the same issue and I solved it by dropping the layers from Conv2d to Conv1d and MaxPooling2d to MaxPooling1d. Which makes sense depending on your data format which in my case was text classification.

answered Mar 15 '18 at 18:14

iantbutler

431
5
13

IndexError: tuple index out of range by max pooling

1 Answers1