[jira] [Updated] (MXNET-807) Some mxnet ctc_loss bug
[ https://issues.apache.org/jira/browse/MXNET-807?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lin Yuan updated MXNET-807: --- Status: In Progress (was: To Do) > Some mxnet ctc_loss bug > --- > > Key: MXNET-807 > URL: https://issues.apache.org/jira/browse/MXNET-807 > Project: Apache MXNet > Issue Type: Bug > Components: Apache MXNet Backend >Reporter: Lin Yuan >Assignee: Lin Yuan >Priority: Major > -- 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
[jira] [Created] (MXNET-884) NDArray, Symbol and Others require explicit Disposing in Clojure
CARIN M MEIER created MXNET-884: --- Summary: NDArray, Symbol and Others require explicit Disposing in Clojure Key: MXNET-884 URL: https://issues.apache.org/jira/browse/MXNET-884 Project: Apache MXNet Issue Type: Improvement Reporter: CARIN M MEIER Creating an NDArray or Symbol requires explicit disposing of the resource in the Clojure/ Scala package. It would be nice to have a macro that would provide an idiomatic way for Clojure users to dispose the resources automatically -- 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
[jira] [Updated] (MXNET-883) Implementation of einsum operator
[ https://issues.apache.org/jira/browse/MXNET-883?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jason Yu updated MXNET-883: --- Description: Einstein summation convention is an expressive and flexible notation for conducting summation along specified axes of a list of tensors. It would greatly simplify some computational operation that would be tricky or complicated to perform without it. ([https://en.wikipedia.org/wiki/Einstein_notation]) As far as I know, numpy, Tensorflow and PyTorch all provide support for it. [https://www.numpy.org/devdocs/reference/generated/numpy.einsum.html] [https://pytorch.org/docs/stable/torch.html#torch.einsum] > Implementation of einsum operator > - > > Key: MXNET-883 > URL: https://issues.apache.org/jira/browse/MXNET-883 > Project: Apache MXNet > Issue Type: New Feature >Reporter: Jason Yu >Priority: Major > > Einstein summation convention is an expressive and flexible notation for > conducting summation along specified axes of a list of tensors. It would > greatly simplify some computational operation that would be tricky or > complicated to perform without it. > ([https://en.wikipedia.org/wiki/Einstein_notation]) > As far as I know, numpy, Tensorflow and PyTorch all provide support for it. > [https://www.numpy.org/devdocs/reference/generated/numpy.einsum.html] > [https://pytorch.org/docs/stable/torch.html#torch.einsum] -- 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
[jira] [Created] (MXNET-883) Implementation of einsum operator
Jason Yu created MXNET-883: -- Summary: Implementation of einsum operator Key: MXNET-883 URL: https://issues.apache.org/jira/browse/MXNET-883 Project: Apache MXNet Issue Type: New Feature Reporter: Jason Yu -- 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