ibis
ibis copied to clipboard
ibis-project
the portable Python dataframe library
I added an extensive comment. I think we have some follow up work to do around clarifying the difference between `rows=(None, None)` and `rows=(None, 0)`. _Originally posted by @cpcloud in...
edit from @lostmygithubaccount: we'll re-purpose this issue to investigate the q2 performance issue noticed below and the "one billion row challenge" performance issue noticed w/ the Polars backend. I may...
### What happened? ```python t1 = ibis.memtable({"x": [1, 2]}) e = t1.x.isin([1, None]) print(ibis.duckdb.connect().execute(e)) print(ibis.sqlite.connect().execute(e)) print(ibis.polars.connect().execute(e)) print(ibis.pandas.connect().execute(e)) ``` yields 0 True 1 None Name: InValues(x), dtype: object 0 True 1...
### Please describe the issue It would be nice if we had an Ibis tutorial for users familiar with the PySpark DataFrame API. This could be similar in structure to...
### Is your feature request related to a problem? Instead of specifying the fraction, I want to sample an absolute number of rows. This is a nice-to-have, I can get...
### What happened? These are failing after `the-epic-split` ### What version of ibis are you using? `the-epic-split` ### What backend(s) are you using, if any? mssql ### Relevant log output...
### What happened? These are now failing on `the-epic-split` ### What version of ibis are you using? `the-epic-split` ### What backend(s) are you using, if any? mssql ### Relevant log...
### Is your feature request related to a problem? If a table has any string columns, those columns in table.execute() are of dtype object, not "string". ### Describe the solution...
**Acceptance criteria**: With UDFs overhauled in recent Ibis releases, this meta-issue is to track planned improvements in Q2 2024. This may include table-value functions and other improvements requested from the...
**Acceptance criteria**: With streaming backends introduced in Q1 2024 (#8036 and #8038), we will continue to improve these existing streaming backends and look to potentially introduce more (i.e. Spark Streaming)....