| commit | 3bfabb4d5fc836639eab631d288b5d6f125c1e8a | [log] [tgz] |
|---|---|---|
| author | Dong Li <dongli@fb.com> | Mon Oct 30 19:30:27 2017 -0700 |
| committer | Facebook Github Bot <facebook-github-bot@users.noreply.github.com> | Mon Oct 30 19:32:30 2017 -0700 |
| tree | 6144226418c12a47f8354743504ef1a462b2b377 | |
| parent | 669ec0ccbaf139b2feb9579308280d1140232a11 [diff] |
support float16 input for operator SparseAdagrad
Summary:
Implemented new CUDA class for operator SparseAdagrad. The param and moment inputs now can be float or float16.
The functions for mixed-precision add/mult/store are defined in a separate head file ("caffe2/core/float16_util.h") for reuse purpose.
Reviewed By: azzolini
Differential Revision: D5880200
fbshipit-source-id: dca227f38629a03a9d771f42efe2c0b673075c4d
Caffe2 is a lightweight, modular, and scalable deep learning framework. Building on the original Caffe, Caffe2 is designed with expression, speed, and modularity in mind.
Caffe2 research award competition request for proposals
Please use Github issues (https://github.com/caffe2/caffe2/issues) to ask questions, report bugs, and request new features.
Please participate in our survey (https://www.surveymonkey.com/r/caffe2). We will send you information about new releases and special developer events/webinars.
Caffe2 is released under the Apache 2.0 license. See the NOTICE file for details.