[
https://issues.apache.org/jira/browse/MXNET-1262?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16718065#comment-16718065
]
Piyush Ghai commented on MXNET-1262:
------------------------------------
* empty()
* Zeros()
* Ones()
* Full()
* Power()
* maximum()
* minimum()
* equal()
* notEqual()
* greater()
* greaterEqual()
* lesser()
* lesserEqual()
* array()
* arange()
* syncCopyfrom() (Corresponding changes in JNI code as well → jFloat to
jDouble, getFloatArray to getDoubleArray)
* asType ( Should it allow downcasting ? )
* set() (Overloaded method accepting Float)
* set() (Overloaded method accepting an Array[Float])
* + (scalar plus) (Also check the behavior when trying to add NDArrays with
different DTypes)
* += (_plus_) (Also check the behavior when trying to add NDArrays with
different DTypes)
* - (Scalar Minus) (Also check the behavior when trying to subtract NDArrays
with different DTypes)
* -= (_minus_)
* * (multiplication)
* unary_-
* *= (multiplication)
* / & /= (division)
* ** (power operation)
* Greater (>, >=)
* Lesser (<, <=)
* Modulo (%, %=)
* toArray()
* toScalar()
* copyTo()
* hashCode may need to be changed ?
* NDArrayConversions (Needs to take in a Double parameter as well)
* NDArrayFunctionReturn (needs methods with Double). (Seems to be mostly
unused. Needs cleanup ? )
> Compile a list of NDArray functions which needs to be modified to support
> Float64/Double
> ----------------------------------------------------------------------------------------
>
> Key: MXNET-1262
> URL: https://issues.apache.org/jira/browse/MXNET-1262
> Project: Apache MXNet
> Issue Type: Sub-task
> Reporter: Piyush Ghai
> Priority: Minor
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@mxnet.apache.org
For additional commands, e-mail: issues-h...@mxnet.apache.org
