[
https://issues.apache.org/jira/browse/ARROW-1993?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Uwe L. Korn updated ARROW-1993:
-------------------------------
Description:
Currently the only option is to use {{Table/Array.from_pandas}} which does
significant unnecessary work and allocates memory. If only the schema is of
interest, then we could do less work and not allocate memory.
We should provide the user a function {{pyarrow.Schema.from_pandas}} which
takes a DataFrame as an input and returns the respective Arrow schema. The
functionality for determing the schema is already available in the Python code,
it is at moment just very tightly bound to the conversion infrastructure.
was:
Currently the only option is to use {{Table/Array.from_pandas}} which does
significant unnecessary work and allocates memory. If only the schema is of
interest, then we could do less work and not allocate memory.
We should provide the user a function {{pyarrow.Schema.from_pandas}} which
takes a DataFrame as an input and returns the respective Arrow schema.
> [Python] Add function for determining implied Arrow schema from
> pandas.DataFrame
> --------------------------------------------------------------------------------
>
> Key: ARROW-1993
> URL: https://issues.apache.org/jira/browse/ARROW-1993
> Project: Apache Arrow
> Issue Type: Improvement
> Components: Python
> Reporter: Wes McKinney
> Assignee: Uwe L. Korn
> Priority: Major
> Labels: beginner
> Fix For: 0.10.0
>
>
> Currently the only option is to use {{Table/Array.from_pandas}} which does
> significant unnecessary work and allocates memory. If only the schema is of
> interest, then we could do less work and not allocate memory.
> We should provide the user a function {{pyarrow.Schema.from_pandas}} which
> takes a DataFrame as an input and returns the respective Arrow schema. The
> functionality for determing the schema is already available in the Python
> code, it is at moment just very tightly bound to the conversion
> infrastructure.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
