Note that I only went through the files and fix whatever the PEP8 script for
vim reported as incorrect, but I had not actually paid attention to the
text, so there might still be inconsistencies (like for example strange
variable names, etc.). Also only scan_opt.py is done completely, scan_op.py
still has a bunch of PEP8 inconsistency.

After the fix I only run the tests in scan_module/tests.py and they passed
so I assume I had not introduced any bug in my PEP8 fixes.

Note also a test was added for this case.

Fix test

Fix test in DebugMode. The test op inherit from a class that implement the __eq__ and __hash__ and those implementation was wrong.

Added note in Windows GotoBLAS compilation

This is to make sure the Perl executable is properly found.

Added doc about R_op.

Implement the grad for inner dimensions in MaxAndArgMax.
The pull request is fine. 

There is one question about gradient of max, that I'm not sure how to tackle, but the changes introduced in this code do not change the behaviour of Theano in this respect. 

The problem is how should the grad of max look if there are several instances of max ? Should all get the full gradient (like it is happening now), should only one of them receive the gradinet ? 

det OP

removed dependency to scipy and used numpy instead
fixed some convention discrepencies

Added some documentation on OutputGuard for new c types.

fixed infer_shape
and
added dome tests

reviewed code... everything seems OK
added infer_shape
added a bit of doc

Typos

Up/downcasted is correct though. English is weird.

Clarified when to register a type with OutputGuard

Better error message when compilation fails

Typo fixes

This may catch more errors.

1. The g++ error message is now displayed in the exception text rather
than hidden before the file content is printed (which made it very hard
to notice).

2. If the call to Popen fails for some reason (e.g. because g++ is not
found), the command line is printed to stderr so that the user has a
better idea of what the problem may be.

Fix subtensor2

Add test for the IntDiv element wise operation.

Fix test