- 17 2月, 2017 1 次提交
-
-
由 Arnaud Bergeron 提交于
-
- 16 2月, 2017 4 次提交
-
-
由 Frederic Bastien 提交于
-
由 Frederic Bastien 提交于
-
由 Arnaud Bergeron 提交于
-
由 Arnaud Bergeron 提交于
-
- 15 2月, 2017 8 次提交
-
-
由 Frédéric Bastien 提交于
Consistently use '\n' in outputted code
-
由 Frederic Bastien 提交于
-
由 valtron 提交于
All generated C code uses `\n` line separators, except some in `gof/op.py` when on Windows. This fixes it. Strangely, this also fixes the issue I described in [#5464](/Theano/Theano/issues/5463#issuecomment-279826487).
-
由 Frédéric Bastien 提交于
Convert dos end of line to unix version
-
由 abergeron 提交于
multiple data type support for GPUMultinomialFromUniform output
-
由 Pascal Lamblin 提交于
Improve the example for strict=True in the scan() doc part 2.
-
由 Frederic Bastien 提交于
-
由 Frédéric Bastien 提交于
Fix test for Python 3 on 32-bit
-
- 14 2月, 2017 7 次提交
-
-
由 Frédéric Bastien 提交于
Interface addition and cleanup
-
由 Pascal Lamblin 提交于
-
由 Pascal Lamblin 提交于
replace multinomial_wo_replacement by choice
-
由 Frederic Bastien 提交于
-
由 Frederic Bastien 提交于
Change MRG choice default of a to be None. We can do this as in the old back-end, when a wasn't provided and pvals was, a=2, but it was only valid when that match pvals shapes. So it was duplicate information. As MRG have pvals mandatory, to work it need a matching value for a.
-
由 Frédéric Bastien 提交于
Merge optimizer 5319
-
由 Frederic Bastien 提交于
-
- 13 2月, 2017 3 次提交
-
-
由 Frédéric Bastien 提交于
Add C code for CPU and GPU for tensor.Split op.
-
由 Pascal Lamblin 提交于
Fix gpu integers in scan.
-
由 Pascal Lamblin 提交于
validate numpy1.12 compatibility changes
-
- 12 2月, 2017 1 次提交
-
-
由 Pascal Lamblin 提交于
-
- 11 2月, 2017 16 次提交
-
-
由 Pascal Lamblin 提交于
-
由 Pascal Lamblin 提交于
casting to int could lead to issues in the following case: - 64-bit Windows - python 2 - shape bigger than 2**31 In that case, int would be 32-bit, but shapes could be bigger and stored in long.
-
由 Pascal Lamblin 提交于
-
由 Chinnadhurai Sankar 提交于
-
由 Pascal Lamblin 提交于
Fix CAReduce.c_code_cache_version_apply Calling `append` on a tuple raised AttributeError, which was mistaken for a missing `c_code...` function altogether, and the version was ignored.
-
由 Pascal Lamblin 提交于
-
由 Chinnadhurai Sankar 提交于
-
由 Chinnadhurai Sankar 提交于
-
由 Chinnadhurai Sankar 提交于
-
由 Chinnadhurai Sankar 提交于
-
由 Chinnadhurai Sankar 提交于
-
由 Chinnadhurai Sankar 提交于
-
由 Chinnadhurai Sankar 提交于
-
由 Chinnadhurai Sankar 提交于
-
由 Chinnadhurai Sankar 提交于
-
由 notoraptor 提交于
Done in both CPU and GPU code.
-