[
https://issues.apache.org/jira/browse/SINGA-370?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16528701#comment-16528701
]
ASF subversion and git services commented on SINGA-370:
-------------------------------------------------------
Commit f9e7caaf5bf91bd3236b8d1532b192d4416999e6 in incubator-singa's branch
refs/heads/master from Vaan Ng
[ https://git-wip-us.apache.org/repos/asf?p=incubator-singa.git;h=f9e7caa ]
SINGA-370 Improvement to tensor reshape and various misc. changes related to
SINGA-341 and 351
converted some upper-case functions to lower-case
fixed return types for some non-void cuda functions
added check_cudnn to check for CUDNN_STATUS_SUCCESS for all cudnn functions
removal of set_strides usage in cuda file and replacement with additional cudnn
transform functions
updated all unary and binary cuda functions for transform
updated reshape function to return tensor
fixed FromProto function by removing original reshape
change to reshape to support in-place operations as well as return new tensor
added constructors for Transform function (GenUnaryTensorFn) similar to cudnn's
transform function
updated tensor_math.h for transform
added Transform function for cpp and cuda
added Transform to reshape instead of add
> Improvement to tensor reshape and various misc. changes related to SINGA-341
> and 351
> ------------------------------------------------------------------------------------
>
> Key: SINGA-370
> URL: https://issues.apache.org/jira/browse/SINGA-370
> Project: Singa
> Issue Type: Improvement
> Reporter: Ng Yi Sen
> Priority: Major
>
> Reshape function should be modified to 'Tensor Reshape' instead of 'void
> Reshape', to allow the function to return a new tensor object with new block
> when encountering a transposed tensor.
> Various misc. changes:
> * Naming conventions (standardization of lower-case characters) in tensor.cc
> related files
> * CUDNN error checks for all CUDNN functions
> * Fixing of certain non-void functions with no return values
> * Strides should not be modified in CUDA or CPP files (check input tensors
> for same strides, else transform out)
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
