Predict batch size default

[mheilman]

This changes the default batch size for prediction to be the same as for training, enabling efficient grid search. Previously GridSearchCV would try to make predictions in a single batch, which could take a lot of memory.

This also adds a simple example of using skflow with GridSearchCV.

[googlebot]

Thanks for your pull request. It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

📝 Please visit https://cla.developers.google.com/ to sign.

Once you've signed, please reply here (e.g. I signed it!) and we'll verify. Thanks.

---

• If you've already signed a CLA, it's possible we don't have your GitHub username or you're using a different email address. Check your existing CLA data and verify that your email is set on your git commits.
• If you signed the CLA as a corporation, please let us know the company's name.

[terrytangyuan]

Thanks. @mheilman You have to sign the CLA first and then we can merge.

[mheilman]

Yeah, will do. It might take a day or two for me to get an official OK to sign that. I'm sure it'll be fine since the CLA seems very reasonable.

[codecov-io]

Current coverage is 90.31%

Merging #126 into master will decrease coverage by -0.07% as of a48a313

@@            master    #126   diff @@
======================================
  Files           27      27       
  Stmts         1061    1063     +2
  Branches       164     165     +1
  Methods          0       0       
======================================
+ Hit            959     960     +1
- Partial         50      51     +1
  Missed          52      52       

Review entire Coverage Diff as of a48a313

Powered by Codecov. Updated on successful CI builds.

[terrytangyuan]

Any update on this? @mheilman

[mheilman]

Sorry for the delay. I still don't think this will be a problem, but the person who needs to OK the CLA has been busy.

[terrytangyuan]

Sure no worries.

[terrytangyuan]

@mheilman skflow has just been moved to TensorFlow here. Could you submit a PR there instead? Examples will be located here. Thanks!

[mheilman]

OK. By the way, the google CLA website linked above (and here) seems to have a bug in it. It has two buttons for submitting on behalf of only myself or on behalf of my employer, but both forward to the same page. Is that correct?

[terrytangyuan]

Yeah I just noticed that too. I have no idea why. Could you ask TF people when you send a PR there? They may know the right person to take care of this issue.

[mheilman]

OK, will do.