[WIP] Add failing test case for image dataset transforms

[kyleabeauchamp]

Initial attempt at getting test coverage for #243

[googlebot]

So there's good news and bad news.

👍 The good news is that everyone that needs to sign a CLA (the pull request submitter and all commit authors) have done so. Everything is all good there.

😕 The bad news is that it appears that one or more commits were authored or co-authored by someone other than the pull request submitter. We need to confirm that all authors are ok with their commits being contributed to this project. Please have them confirm that here in the pull request.

Note to project maintainer: This is a terminal state, meaning the cla/google commit status will not change from this state. It's up to you to confirm consent of all the commit author(s), set the cla label to yes (if enabled on your project), and then merge this pull request when appropriate.

ℹ️ Googlers: Go here for more info.

[kyleabeauchamp]

OK, so I've rebased this PR off of #330. I can confirm that the failing test case is now passing.

[kyleabeauchamp]

For sanity, we probably don't want to review this until #330 is merged.

[facaiy]

Looks good, Kyle. As we add test case for unknown shape tensor, is it necessary to retain
tf.data.Dataset test case? What do you think?

[kyleabeauchamp]

Yeah it's probably a bit redundant, but the main value is adding a test that covers the reported use case.

…

On Thu, Jul 4, 2019, 11:03 PM Yan Facai (颜发才) ***@***.***> wrote: Looks good, Kyle. As we add test case for unknown shape tensor, is it necessary to retain tf.data.Dataset test case? What do you think? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#244?email_source=notifications&email_token=AAOP7K7YACK3MXR4UBPDOJLP53P3FA5CNFSM4HMWQNG2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODZIT2GA#issuecomment-508640536>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAOP7K4F3NJEPNQIHFDAHG3P53P3FANCNFSM4HMWQNGQ> .

[facaiy]

Make sense to me as well, but I prefer to make unit test simple according to Occam's Razor. How about letting us add test case with tf.data.Dataset later when we think it necessary?

[kyleabeauchamp]

Sgtm, I'm fine with closing this ticket.

…

On Fri, Jul 5, 2019, 4:27 PM Yan Facai (颜发才) ***@***.***> wrote: Make sense to me as well, but I prefer to make unit test simple according to Occam's Razor. How about letting us add test case with tf.data.Dataset later when we think it necessary? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#244?email_source=notifications&email_token=AAOP7K6JD3GOC4ZFSRWCDCLP57KENA5CNFSM4HMWQNG2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODZKODQI#issuecomment-508879297>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAOP7K23JU6AZWEDQK3GAHDP57KENANCNFSM4HMWQNGQ> .