[GitHub] yuxiangw opened a new pull request #10779: An example of differentially private deep learning using mxnet gluon

[GitBox <gi...@apache.org>]

yuxiangw opened a new pull request #10779: An example of differentially private deep learning using mxnet gluon
URL: https://github.com/apache/incubator-mxnet/pull/10779
 
 
   ## Description ##
   
   An example of differentially private deep learning using `mxnet.gluon` and `pydiffpriv`.
   The module `dpdl_utils.py` might have some independent interest, when one needs to work with parameters and gradients of a model directly as ndarrays. 
   
   ## Checklist ##
   ### Essentials ###
   - [x] Code is well-documented: 
   - For new examples, README.md is added to explain the what the example does.
   - [x] To the best of my knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change
   
   ## Comments ##
   - More examples will be added to it in the next update.
   - The module `dpdl_utils.py` might be of independent interest. It bridges the gap between python the gluon-style model parameters: `mxnet.gluon.ParameterDict`  and the *from scratch* version: that simply uses a list of `mxnet.ndarray`s.  `dpdl_utils.py` contains utility functions necessary to manipulate the parameters and the corresponding gradients in ParameterDict using generic ndarray operations. 
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services