singa-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From GitBox <...@apache.org>
Subject [GitHub] [singa] dcslin edited a comment on issue #784: Preparation for V3.1 release
Date Fri, 25 Sep 2020 12:51:39 GMT

dcslin edited a comment on issue #784:
URL: https://github.com/apache/singa/issues/784#issuecomment-698093681


   * BiLSTM model on [InsuranceQA](https://github.com/shuzi/insuranceQA) example training
script. The baseline model is inspired by Tan, Ming, et al. "Lstm-based deep learning models
for non-factoid answer selection." arXiv preprint arXiv:1511.04108 (2015).
   
   * Tensor Refactoring and Enhancement
     - Tensor transformation (reshape, transpose) supports up to 6 dimensions.
     - Implemented traverse_unary_transform in Cuda backend, which is similar to CPP backend
one.
     - Added tensor operation erf, rounde (round even).
     - Fix Tensor operation Mult on Broadcasting use cases.
     - Gaussian function on Tensor now can run on Tensor with odd size.
   
   * autograd
     - Added some sanity check on autograd input to prevent fatal error caused by unexpected
input shape.
     - Updated a testing helper function gradients() in autograd to lookup param gradient
by param python object id for testing purpose.


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to 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



Mime
View raw message