BUG: item_cache not cleared on DataFrame.values

[jbrockmendel]

• closes #xxxx
• tests added / passed
• passes black pandas
• passes git diff upstream/master -u -- "*.py" | flake8 --diff
• whatsnew entry

[jreback]

any perf impact? (i would actually expect some small regressions)

[jbrockmendel]

In [1]: import pandas as pd                                                                                                                                                                                        


In [2]:                                                                                                                                                                                                            

In [2]: df = pd.DataFrame([1, 2])                                                                                                                                                                                  

In [3]: %timeit df.values
2.25 µs ± 42.9 ns per loop (mean ± std. dev. of 7 runs, 100000 loops each)  # <-- master
3.82 µs ± 106 ns per loop (mean ± std. dev. of 7 runs, 100000 loops each)  # <-- PR

[jreback]

hmm, do we have any to_numpy() benchmarks? just to see not blocking this PR

[jbrockmendel]

do we have any to_numpy() benchmarks

Looks like we do not

[jreback]

do we have any to_numpy() benchmarks

Looks like we do not

kk happy to merge and can create issue for benchmarks (as a good first - people like to do those)

[jbrockmendel]

issue opened

[jreback]

thanks

[TomAugspurger]

@jbrockmendel caused a 100x slowdown in replace.Convert.time_replace (https://pandas.pydata.org/speed/pandas/index.html#replace.Convert.time_replace?p-constructor=%27Series%27&p-replace_data=%27Timestamp%27&commits=dbfbef7eaea7f23a42d9a65344b5745fc055f5b4). Is that expected?

[jbrockmendel]

caused a 100x slowdown in replace.Convert.time_replace [...] Is that expected?

I'd expect something closer to the 70% slowdown we see in df.values. 100x is surprising.

[TomAugspurger]

Opened #35053 to track this.

[phs-sakshi]

Hey I am new to open source. Can I take up this issue?

[jorisvandenbossche]

@B417037 this is a merged PR, not an issue to take up. Have a look at good first issue