Control: forwarded -1 https://github.com/Theano/Theano/issues/5498 Control: tags -1 patch
I don't think this is a regression - it's Python 3 specific (numpy.array(list of longs, which this test uses on Python 2)=int64 array, but numpy(list of Python 3 ints)=int_nativesize array; see above link for longer discussion) and 0.8.2-2 appears to be the first time the tests were run with Python 3.
Fix (though I've only tried these particular tests, not a full build): --- a/theano/tensor/tests/test_basic.py +++ b/theano/tensor/tests/test_basic.py @@ -6672,11 +6672,11 @@ class T_long_tensor(unittest.TestCase): assert scalar_ct.value == val vector_ct = constant([val, val]) - assert vector_ct.dtype == 'int64' + assert vector_ct.dtype in ('int32','int64') assert numpy.all(vector_ct.value == val) matrix_ct = constant([[val, val]]) - assert matrix_ct.dtype == 'int64' + assert matrix_ct.dtype in ('int32','int64') assert numpy.all(matrix_ct.value == val) def test_too_big(self):