Comments (16)
TomAugspurger
commented on July 17, 2024
1
More details would be helpful, otherwise it seems more likely that it's something specific to your machine than the package itself.
from pandas-feedstock.
mjbeyeler
commented on July 17, 2024
1
Sorry for the late response.
The problem turned out to be deeper. I don't understand what exactly it was, but it was somehow related to other python distributions that had previously been installed to the VM (not by conda).
In the end, I switched to a new VM, and now everything works fine.
So, you guys were right, the my machine was the problem. Thanks!
from pandas-feedstock.
mjbeyeler
commented on July 17, 2024
Hmm, well I'm running the default miniconda installation, and my channel hierarchy, from highest to lowest priority, is:
conda-forge
bioconda
defaults
This should be all information required to reproduce my error, as far as I can tell, but let me know what else you'd need to know in case.
To reproduce the error, after having created and activated the environment via conda create -n myenv pandas and conda activate myenv,
type python in the Terminal, and then import pandas as pd, which returns the error
Illegal construction (core dumped)
from pandas-feedstock.
TomAugspurger
commented on July 17, 2024
See https://raw.githubusercontent.com/conda/conda/master/.github/ISSUE_TEMPLATE/Bug_report.md for the types of things that conda typically asks for.
from pandas-feedstock.
mjbeyeler
commented on July 17, 2024
What do you mean?
It's not a conda issue as far as I can tell.
When I create the environment via conda create -n myenv -c defaults myenv, then importing pandas works just fine.
So, I think we're looking at a specific conda-forge issue. Did you try reproducing the error?
from pandas-feedstock.
ocefpaf
commented on July 17, 2024
You are also using 3 channels that are not 100% compatible with each other. See https://conda-forge.org/docs/user/tipsandtricks.html#how-to-fix-it
from pandas-feedstock.
TomAugspurger
commented on July 17, 2024
Yes, I created a new env with conda-forge's pandas and things are fine.
…On Wed, Jun 26, 2019 at 10:55 AM Filipe ***@***.***> wrote:
You are also using 3 channels that are not 100% compatible with each
other. See
https://conda-forge.org/docs/user/tipsandtricks.html#how-to-fix-it
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#63?email_source=notifications&email_token=AAKAOIV4ZD5PRUVR5PMEYZDP4OGN3A5CNFSM4H3TEFD2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODYT7VVA#issuecomment-505936596>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AAKAOIWIPYMMHDUYAC6UTXLP4OGN3ANCNFSM4H3TEFDQ>
.
from pandas-feedstock.
ocefpaf
commented on July 17, 2024
So, I think we're looking at a specific conda-forge issue. Did you try reproducing the error?
Nope. It is a mixing channel issue. Please check https://conda-forge.org/docs/user/tipsandtricks.html#using-multiple-channels
Closing this b/c it is not an actionable item on the pandas packages or conda-forge. conda-forge does not guarantee compatibility with third party channel and you should always use strict channel option for stability.
from pandas-feedstock.
mjbeyeler
commented on July 17, 2024
@ocefpaf I am using the strict channel hierarchy option.
So, when I am installing using the commands described above, AND having strict channel option, I still have libgcc-ng and libgfortran-ng and libstdcxx-ng installed from pkgs/main when installing via conda-forge.
Also conda-forge doesn't install mkl-related packages nor numpy-base, chich defaults provides.
Do you have an idea what might be causing the problem then?
from pandas-feedstock.
ocefpaf
commented on July 17, 2024
The following .condarc is what you descrived, core=rect?
channel_priority: strict
channels:
- conda-forge
- bioconda
- defaultsand I just tested:
conda create --name TEST pandas --yes
conda activate TEST
python -c "import pandas; print(pandas.__version__)"
0.24.2and here is the conda list:
> conda list
# packages in environment at /home/filipe/miniconda3/envs/TEST:
#
# Name Version Build Channel
bzip2 1.0.6 h14c3975_1002 conda-forge
ca-certificates 2019.6.16 hecc5488_0 conda-forge
certifi 2019.6.16 py37_0 conda-forge
libblas 3.8.0 7_openblas conda-forge
libcblas 3.8.0 7_openblas conda-forge
libffi 3.2.1 he1b5a44_1006 conda-forge
libgcc-ng 9.1.0 hdf63c60_0 defaults
libgfortran-ng 7.3.0 hdf63c60_0 defaults
liblapack 3.8.0 7_openblas conda-forge
libstdcxx-ng 9.1.0 hdf63c60_0 defaults
ncurses 6.1 hf484d3e_1002 conda-forge
numpy 1.16.4 py37h95a1406_0 conda-forge
openblas 0.3.5 h9ac9557_1001 conda-forge
openssl 1.1.1b h14c3975_1 conda-forge
pandas 0.24.2 py37hb3f55d8_0 conda-forge
pip 19.1.1 py37_0 conda-forge
python 3.7.3 h5b0a415_0 conda-forge
python-dateutil 2.8.0 py_0 conda-forge
pytz 2019.1 py_0 conda-forge
readline 7.0 hf8c457e_1001 conda-forge
setuptools 41.0.1 py37_0 conda-forge
six 1.12.0 py37_1000 conda-forge
sqlite 3.28.0 h8b20d00_0 conda-forge
tk 8.6.9 hed695b0_1002 conda-forge
wheel 0.33.4 py37_0 conda-forge
xz 5.2.4 h14c3975_1001 conda-forge
zlib 1.2.11 h14c3975_1004 conda-forge
Also conda-forge doesn't install mkl-related packages nor numpy-base, chich defaults provides.
Yep. B/c they are incompatible with conda-forge's openblas numpy.
Do you have an idea what might be causing the problem then?
If you are forcing numpy-base and/or mkl, that is the problem.
from pandas-feedstock.
mjbeyeler
commented on July 17, 2024
My ~/.condarc is
channels:
- conda-forge
- bioconda
- defaults
channel_priority: strict
I copy-pasted your exact code, and I get the same error as described above.
My conda list:
# packages in environment at /home/bioinf/miniconda3/envs/TEST:
#
# Name Version Build Channel
bzip2 1.0.6 h14c3975_1002 conda-forge
ca-certificates 2019.6.16 hecc5488_0 conda-forge
certifi 2019.6.16 py37_0 conda-forge
libblas 3.8.0 7_openblas conda-forge
libcblas 3.8.0 7_openblas conda-forge
libffi 3.2.1 he1b5a44_1006 conda-forge
libgcc-ng 9.1.0 hdf63c60_0
libgfortran-ng 7.3.0 hdf63c60_0
liblapack 3.8.0 7_openblas conda-forge
libstdcxx-ng 9.1.0 hdf63c60_0
ncurses 6.1 hf484d3e_1002 conda-forge
numpy 1.16.4 py37h95a1406_0 conda-forge
openblas 0.3.5 h9ac9557_1001 conda-forge
openssl 1.1.1b h14c3975_1 conda-forge
pandas 0.24.2 py37hb3f55d8_0 conda-forge
pip 19.1.1 py37_0 conda-forge
python 3.7.3 h5b0a415_0 conda-forge
python-dateutil 2.8.0 py_0 conda-forge
pytz 2019.1 py_0 conda-forge
readline 7.0 hf8c457e_1001 conda-forge
setuptools 41.0.1 py37_0 conda-forge
six 1.12.0 py37_1000 conda-forge
sqlite 3.28.0 h8b20d00_0 conda-forge
tk 8.6.9 hed695b0_1002 conda-forge
wheel 0.33.4 py37_0 conda-forge
xz 5.2.4 h14c3975_1001 conda-forge
zlib 1.2.11 h14c3975_1004 conda-forge
They seem almost identical to me, except that the three non-conda-forge packages don't have the defaults tag for me. Very weird.
I'll get home soon and check with my home PC.
from pandas-feedstock.
ocefpaf
commented on July 17, 2024
They seem almost identical to me, except that the three non-conda-forge packages don't have the defaults tag for me. Very weird.
Not weird, you can use show_channel_urls: true for that. Also, you did not mentioned if that worked or not or why you mentioned mkl and numpy-base. It is hard to help you with fragmented information. I'm assuming it is resolved and "filing this" as usage question.
from pandas-feedstock.
mjbeyeler
commented on July 17, 2024
Sorry if I haven't been clear, I have mentioned mkl and numpy-base because those packages are installed when building a pandas environment from the defaults channel.
Also, you were right that it's kind of has to be a machine-dependent issue. I ran your code snippet on my personal PC yesterday, and it worked. But it's still not working on my work PC.
using show_channel_urls: true, I get the following conda list, which, according to Word comparison tool is identical to your list... yet my environment's not working. Now that's weird no?
# packages in environment at /home/bioinf/miniconda3/envs/TEST:
#
# Name Version Build Channel
bzip2 1.0.6 h14c3975_1002 conda-forge
ca-certificates 2019.6.16 hecc5488_0 conda-forge
certifi 2019.6.16 py37_0 conda-forge
libblas 3.8.0 7_openblas conda-forge
libcblas 3.8.0 7_openblas conda-forge
libffi 3.2.1 he1b5a44_1006 conda-forge
libgcc-ng 9.1.0 hdf63c60_0 defaults
libgfortran-ng 7.3.0 hdf63c60_0 defaults
liblapack 3.8.0 7_openblas conda-forge
libstdcxx-ng 9.1.0 hdf63c60_0 defaults
ncurses 6.1 hf484d3e_1002 conda-forge
numpy 1.16.4 py37h95a1406_0 conda-forge
openblas 0.3.5 h9ac9557_1001 conda-forge
openssl 1.1.1b h14c3975_1 conda-forge
pandas 0.24.2 py37hb3f55d8_0 conda-forge
pip 19.1.1 py37_0 conda-forge
python 3.7.3 h5b0a415_0 conda-forge
python-dateutil 2.8.0 py_0 conda-forge
pytz 2019.1 py_0 conda-forge
readline 7.0 hf8c457e_1001 conda-forge
setuptools 41.0.1 py37_0 conda-forge
six 1.12.0 py37_1000 conda-forge
sqlite 3.28.0 h8b20d00_0 conda-forge
tk 8.6.9 hed695b0_1002 conda-forge
wheel 0.33.4 py37_0 conda-forge
xz 5.2.4 h14c3975_1001 conda-forge
zlib 1.2.11 h14c3975_1004 conda-forge
from pandas-feedstock.
mjbeyeler
commented on July 17, 2024
from pandas-feedstock.
TomAugspurger
commented on July 17, 2024
You might try re-installing conda, or installing it in a different prefix,
on that other machine.
…On Thu, Jun 27, 2019 at 3:07 AM Michael Beyeler ***@***.***> wrote:
Sorry if I haven't been clear, I have mentioned mkl and numpy-base because
those packages are installed when building a pandas environment from the
defaults channel.
Also, you were right that it's kind of has to be a machine-dependent
issue. I ran your code snippet on my personal PC yesterday, and it worked.
But it's still not working on my work PC.
using show_channel_urls: true, I get the following conda list, which,
according to Word comparison tool is identical to your list... yet my
environment's not working. Now that's weird no?
# packages in environment at /home/bioinf/miniconda3/envs/TEST:
#
# Name Version Build Channel
bzip2 1.0.6 h14c3975_1002 conda-forge
ca-certificates 2019.6.16 hecc5488_0 conda-forge
certifi 2019.6.16 py37_0 conda-forge
libblas 3.8.0 7_openblas conda-forge
libcblas 3.8.0 7_openblas conda-forge
libffi 3.2.1 he1b5a44_1006 conda-forge
libgcc-ng 9.1.0 hdf63c60_0 defaults
libgfortran-ng 7.3.0 hdf63c60_0 defaults
liblapack 3.8.0 7_openblas conda-forge
libstdcxx-ng 9.1.0 hdf63c60_0 defaults
ncurses 6.1 hf484d3e_1002 conda-forge
numpy 1.16.4 py37h95a1406_0 conda-forge
openblas 0.3.5 h9ac9557_1001 conda-forge
openssl 1.1.1b h14c3975_1 conda-forge
pandas 0.24.2 py37hb3f55d8_0 conda-forge
pip 19.1.1 py37_0 conda-forge
python 3.7.3 h5b0a415_0 conda-forge
python-dateutil 2.8.0 py_0 conda-forge
pytz 2019.1 py_0 conda-forge
readline 7.0 hf8c457e_1001 conda-forge
setuptools 41.0.1 py37_0 conda-forge
six 1.12.0 py37_1000 conda-forge
sqlite 3.28.0 h8b20d00_0 conda-forge
tk 8.6.9 hed695b0_1002 conda-forge
wheel 0.33.4 py37_0 conda-forge
xz 5.2.4 h14c3975_1001 conda-forge
zlib 1.2.11 h14c3975_1004 conda-forge
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#63?email_source=notifications&email_token=AAKAOIWHAEWIOMTPHNQ5RA3P4RYNFA5CNFSM4H3TEFD2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODYWJ4GQ#issuecomment-506240538>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AAKAOIU75ZYLU36RFAEWVG3P4RYNFANCNFSM4H3TEFDQ>
.
from pandas-feedstock.
ocefpaf
commented on July 17, 2024
You might try re-installing conda, or installing it in a different prefix, on that other machine.
Also clean your package cache. You may have something corrupted there.
conda clean --all --yes -f
from pandas-feedstock.
Related Issues (20)
- Running test suite on PyPy segfaults HOT 10
- Test failures with `test_rolling_var_numerical_issues` on Linux `ppc64le` HOT 1
- Test failures with NumPy 1.24 HOT 2
- [warning] failed package validation and/or copy for commit d5427e6d3fa3a233598c491a333e6e508327fb70 HOT 11
- @conda-forge-admin, update version HOT 1
- @conda-forge-admin, please update version HOT 1
- Failure with PyPy 7.3.12 HOT 6
- @conda-forge-admin, please update version HOT 1
- add pandas-all
- Pandas 2.0.3 crashing on all 3 major platforms HOT 1
- @conda-forge-admin, please add user @anmyachev HOT 1
- Is it possible to create a v1.5.x branch for Python 3.12 compatibility? HOT 1
- Add run_constrained for optional dependencies HOT 2
- @conda-forge-admin, please update version HOT 1
- @conda-forge-admin, please update version HOT 1
- @conda-forge-admin, please update version HOT 1
- Should we consider removing the tests files from the final package? HOT 2
- @conda-forge-admin, please update version HOT 1
- @conda-forge-admin, please update version HOT 1
- @conda-forge-admin, please update version HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from pandas-feedstock.