pytorch/caffe2/python/modeling
Aapo Kyrola 86dc6e0837 Added inverted FP16 Initializer
Summary: Added initializer which sets up the ParameterInfo object in the opposite format as the pFP16Initializer. This is needed for when the op requires the initialized blob to be FP32 but a FP16 copy of the weights is needed.

Reviewed By: wesolwsk

Differential Revision: D5840832

fbshipit-source-id: 439e87f41a1dbc58bf63a5c0e7f7fc4cb00b4d65
2017-10-27 10:20:04 -07:00
..
initializers_test.py Re-license to Apache 2017-09-28 16:22:00 -07:00
initializers.py Added inverted FP16 Initializer 2017-10-27 10:20:04 -07:00
parameter_info.py Re-license to Apache 2017-09-28 16:22:00 -07:00
parameter_sharing_test.py Re-license to Apache 2017-09-28 16:22:00 -07:00
parameter_sharing.py Add shape checks and print more info in parameter sharing 2017-10-27 01:22:06 -07:00