Where communities thrive

  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
Repo info
  • 23:11
    alexreg commented #43949
  • 23:03
    mroeschke commented #43900
  • 23:03

    mroeschke on master

    ENH: Add support for more place… (compare)

  • 23:03
    mroeschke closed #43900
  • 23:03
    mroeschke closed #43901
  • 22:59
    pep8speaks commented #43949
  • 22:59
    alexreg synchronize #43949
  • 22:50
    jreback commented #44041
  • 22:42

    jreback on master

    REF: share ExtensionIndex astyp… (compare)

  • 22:42
    jreback closed #44059
  • 22:39
    jreback commented #44063
  • 22:39
    jreback labeled #44063
  • 22:35
    jreback milestoned #43909
  • 22:34
    jreback milestoned #44068
  • 22:34
    jreback labeled #44068
  • 22:04
    gabrieldi95 synchronize #44046
  • 22:00

    jreback on master

    REF: share Index subclass forma… (compare)

  • 22:00
    jreback closed #44070
  • 22:00
    jreback milestoned #44070
  • 22:00
    jreback labeled #44070
Michael Sarrazin
I try to modify the docstring of a function in Pandas. The "script/validate_docstrings.py" doensn't take my modification into account. Any idea why ?
Guneet Singh
I am new to open source and am having difficulty understanding the issues marked as "good first issue" and the ones that I can understand, are already done could anyone maybe help me with finding an issue to solve?
hey I found a bug 7 years ago when i was n00b. :)
sorry for not fixing it
3 replies
Varun Shrivastava
@MarcoGorelli Can you please assign me issue #21749, I tried using "take" but the bot didn't assign it to me automatically. Also, I have opened a PR for issue #28724 it will close the issue most probably.
1 reply
his everyone, is there a specific room for help with UDF functions and performance. I need to find if a list_of_words are in every dataframe row and extract the column names if exist. but it's taking 10min to check one word against single a row with 200 elements using .apply(UDF, args[]).
2 replies
Mr. Noob
Hey !!!
Is there any app or something where all developers talk and solve each other's problem and having fun
Shoham Debnath
Hi everyone.. what's the process to get triaging rights?
Suyash Gupta

Hi everyone, I'm facing this error while trying to run a pytest test. The test ran fine a few days ago; however, for a long time now, I'm facing this error. Not sure what's causing it and I've been unable to resolve it for hours now.

Here's the command I'm giving:

cd /workspaces/pandas/pandas/tests/indexes/datetimes
pytest -vv test_indexing.py::TestIndexerBetweenTime::test_indexer_between_time

Here's the error:

ImportError while loading conftest '/workspaces/pandas/pandas/conftest.py'.
../../../__init__.py:137: in <module>
    from pandas.io.api import (
../../../io/api.py:8: in <module>
    from pandas.io.excel import (
../../../io/excel/__init__.py:1: in <module>
    from pandas.io.excel._base import (
../../../io/excel/_base.py:18: in <module>
    from pandas._libs.parsers import STR_NA_VALUES
E   ImportError: /workspaces/pandas/pandas/_libs/parsers.cpython-38-x86_64-linux-gnu.so: undefined symbol: str_to_int64
Irv Lustig

@suyashgupta01 Try this:

cd /workspaces/pandas
pytest -vv pandas/tests/indexes/datetimes/test_indexing.py

I believe you need to have your current working directory the parent of the location of pandas/__init__.py

Suyash Gupta
@Dr-Irv It gives the same error.
I ran the command just fine a few days ago, the only difference is a fetch from master and merge along with a few more commits from my side
Irv Lustig
@suyashgupta01 If you did a merge from master, you probably need to do python setup.py build_ext -j 4 again to get the Cpython stuff rebuilt
Suyash Gupta
@Dr-Irv I actually did do that. However, that doesn't seem to resolve the error. Not sure what direction should I go in now.
Irv Lustig
@suyashgupta01 Try doing a python setup.py clean then python setup.py build_ext -j 4
Suyash Gupta
Oh this works! Thanks a lot @Dr-Irv
Noah Wöhler
Hi, can I post a call for participants in an interview study on open source projects here? If any mod wants more details via DM first, then I'm happy to oblige 🙂
guru kiran

Reposting again, I deleted my post by mistake.

As of now there's no support for drop duplicate columns. API currently as .drop_duplicated, which removes duplicate rows. Can we add axis parameter to .drop_duplicated where 0 -> drop rows and 1 -> drop columns.

  X1   X2  Y1   Y2
 0.0  0.0  6.0  6.0
 3.0  3.0  7.1  7.1
 7.6  7.6  1.2  1.2

 # Desired Output

    X1   Y1
0  0.0  6.0
1  3.0  7.1
2  7.6  1.2

There are a few overaround. One, transpose, drop_duplicate and transpose back.
Second, using np.unique over axis 1

def drop_duplicate_cols(df):
    uniq, idxs = np.unique(df, return_index=True, axis=1)
    return pd.DataFrame(uniq, index=df.index, columns=df.columns[idxs])

    X1   Y1
0  0.0  6.0
1  3.0  7.1
2  7.6  1.2

Is there some reason to not add drop_duplicates over axis 1?

Irv Lustig
@gurukiran07 We have an open issue for this: pandas-dev/pandas#16868 Feel free to create a PR
guru kiran
@Dr-Irv Thank you for finding the relevant issue. I'll write a PR once I have some free time. I agree with @jreback that double .T would mess dtypes(I wrote an answer on StackOverflow: https://stackoverflow.com/a/69323395/12416453). If the implementation I proposed is ok I can go ahead and write a PR.
Alex Lim
Hi, is there a way to run debug mode on Cython files while running pytests? I'm currently developing with a conda environment in VSCode
hi guys, i have task related with pandas in my university, but i dont love python and dont have time for learn it, so could you please help me with it. i think it easy
2 replies
ah v1.2.5 has the old behaviour
7 replies
(not a with a mind to merge it of course, just if it's the correct fix or not)
Rommel Silva
Hey everyone, is there any help that I can provide to help review or make necessary updates to the following PR? pandas-dev/pandas#29636
I desperately need this feature live
Jonathan Rousseau

I'm trying to get pandas to work on Ubuntu server 20.04.3 LTS installed on a Raspberry pi, with python 3.8. I've tried everything I could find on the forums to get this to work.
using pip, the 1.3.3 installation hangs. With apt python3-pandas, it installs v0.25. I'm using 1.1.0 functionalities so that doesn't work.

I decided to try building it from source. I'm stuck at this point where it seems to hang again:

ubuntu@raspi001:~/pandasrepo/pandas$ python3 setup.py install --user
running install
running bdist_egg
running egg_info
writing pandas.egg-info/PKG-INFO
writing dependency_links to pandas.egg-info/dependency_links.txt
writing entry points to pandas.egg-info/entry_points.txt
writing requirements to pandas.egg-info/requires.txt
writing top-level names to pandas.egg-info/top_level.txt
reading manifest file 'pandas.egg-info/SOURCES.txt'
reading manifest template 'MANIFEST.in'
no previously-included directories found matching 'doc/build'
warning: no previously-included files matching '*.so' found anywhere in distribution
warning: no previously-included files matching '*~' found anywhere in distribution
warning: no previously-included files matching '.DS_Store' found anywhere in distribution
warning: no previously-included files matching '#*' found anywhere in distribution
warning: no previously-included files matching '*.py[ocd]' found anywhere in distribution
writing manifest file 'pandas.egg-info/SOURCES.txt'
installing library code to build/bdist.linux-armv7l/egg
running install_lib
running build_py
UPDATING build/lib.linux-armv7l-3.8/pandas/_version.py
set build/lib.linux-armv7l-3.8/pandas/_version.py to '1.4.0.dev0+857.gd30aeeba0c'
running build_ext
building 'pandas._libs.algos' extension
arm-linux-gnueabihf-gcc -pthread -Wno-unused-result -Wsign-compare -DNDEBUG -g -fwrapv -O2 -Wall -g -fstack-protector-strong -Wformat -Werror=format-security -g -fwrapv -O2 -g -fstack-protector-strong -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2 -fPIC -DNPY_NO_DEPRECATED_API=0 -I./pandas/_libs -Ipandas/_libs/src/klib -I/usr/lib/python3/dist-packages/numpy/core/include -I/usr/include/python3.8 -c pandas/_libs/algos.c -o build/temp.linux-armv7l-3.8/pandas/_libs/algos.o

I'm at a point where I have no clue what to try next. I figured you guys might have an idea of what's going on?

4 replies
Da Li
I am wondering why not provide Pandas.DataFrame.apply() with engine='numba' arguments, but provide for window and groupby contexts? Any reason?
Hi, I published a package called piso yesterday which brings set operations to the interval classes in pandas. Feedback or feature requests are more than welcome. https://github.com/staircase-dev/piso.
Erfan Nariman

Anyone else ran in the following issue with the newer versions of pip while building a dev environment? INFO: pip is looking at multiple versions of <package name> to determine which version is compatible with other requirements. After an hour I cancelled it.

Using --no-deps does not solve it, since none of the dependencies of the packages will be installed.

Bijay Regmi
Hello everyone, has anyone configured pandas-dev-flaker with flake8 and black with vscode before? I have very absurd interactions with black and when I modify and save a file, it reformats almost everything altering lines that my change has nothing to do with. I would appreciate any help.
3 replies
Makes sense based on the solutions mentioned in related issues #42423 and #42888