[
https://issues.apache.org/jira/browse/ARROW-1614?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17210489#comment-17210489
]
Bryan Cutler commented on ARROW-1614:
-------------------------------------
I just wanted to let you all know I have been working on a similar Tensor
extension type. I currently have a Pandas extension type for a tensor with
conversion to/from an Arrow extension type, just for Python/PyArrow right now,
and zero-copy conversion with numpy.ndarrays. It's part of the project [Text
Extensions for Pandas|https://github.com/CODAIT/text-extensions-for-pandas]
where we use it for NLP feature vectors, but it's really general purpose. You
can check it out at
[https://github.com/CODAIT/text-extensions-for-pandas/blob/master/text_extensions_for_pandas/array/tensor.py]
[https://github.com/CODAIT/text-extensions-for-pandas/blob/master/text_extensions_for_pandas/array/arrow_conversion.py]
Or install the package if you like via {{pip install
text-extensions-for-pandas}} (it's currently in alpha)
We would love to help out with this effort and contribute what we have to
Arrow, if it fits the bill!
> [C++] Add a Tensor logical value type with constant dimensions, implemented
> using ExtensionType
> -----------------------------------------------------------------------------------------------
>
> Key: ARROW-1614
> URL: https://issues.apache.org/jira/browse/ARROW-1614
> Project: Apache Arrow
> Issue Type: New Feature
> Components: C++, Format
> Reporter: Wes McKinney
> Priority: Major
>
> In an Arrow table, we would like to add support for a column that has values
> cells each containing a tensor value, with all tensors having the same
> dimensions. These would be stored as a binary value, plus some metadata to
> store type and shape/strides.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
