Prototyping numpy arrays with named axes for data management.
jonathan-taylor/stats-lecture-notes 26
Statistics lecture notes
Draft of NumPy paper
Bay area fMRI course description
Example notebooks for basic tasks in neuroimaging and statistcs with nipy
Automated compilation of Python wheel binary packages for Biopython
Research: representation of functions on the sphere and diffusion-weighted imaging
Datarray Documentation
push eventmatthew-brett/cfd2020-notes
commit sha ec69253215bf6a6e3b102a80bc556bcce8aff39d
Policies
commit sha 073347e08286496ddde86551e0622c8d80c44690
Credit for policies
push time in 13 hours
push eventmatthew-brett/uob-jupyterhub
commit sha 81d68ed439f1400ff10c2fe12a313c004e6e2aba
Typo
push time in a day
push eventmatthew-brett/uob-jupyterhub
commit sha 1f6395e919286762d5511991cbbc56cb24ddecb0
Fill out CILogon etc auth instructions.
push time in a day
push eventmatthew-brett/uob-jupyterhub
commit sha 777c701c75478b386d5631968c2ee0416e804159
Better comments for cleaned config.
push time in a day
issue commentjupyterhub/oauthenticator
Stripping domains only for case of single allowed domain?
Ah - thanks - I knew there was a good reason, it's just I hadn't seen it.
But how about my suggestion of allowing a list for strip_idp_domain to cover my use-case?
comment created time in a day
issue openedjupyterhub/oauthenticator
Stripping domains only for case of single allowed domain?
This is just asking for the reasoning behind the logic at https://github.com/jupyterhub/oauthenticator/blob/master/oauthenticator/cilogon.py#L206.
In the CILogon authenticator, you can specify allowed identity provider suffixes in allowed_idps' - in my case these will be['bham.ac.uk', 'student.bham.ac.uk']`.
You can also specify strip_idp_domain, which will strip the username from (e.g. in my case) [email protected] to just m.brett. This is what I want.
But the logic specifies that I can only strip_idp_domain if I have exactly one entry in allowed_idps. Why is that a requirement? It's a problem in my case, because I must allow two suffixes to allow for staff and students, but I want the bare username e.g. "m.brett" rather than "[email protected]".
Maybe strip_idp_domain could also also allow a list of domains that should be stripped?
created time in a day
push eventmatthew-brett/uob-jupyterhub
commit sha 831ba29e7a73aeaa67df47909e5406ca96a1e209
Cleaned config
push time in a day
pull request commentmatthew-brett/multibuild
Update openblas download from anaconda
Thanks Matti. Restarted.
comment created time in a day
pull request commentmatplotlib/matplotlib
Add a code of conduct link to github
Only to say that, for the Scipy code of conduct (later adopted by Numpy) we struggled with the problem that codes of conduct are often very badly written, in an unpleasant imitation of legal documents, and disheartening, as well as being light on specifics of procedure for reporting adjudication and enforcement. So we did our collective best to be specific, and to make the text as readable and positive as possible.
comment created time in 2 days
push eventmatthew-brett/uob-jupyterhub
commit sha 544ec6870e52c8a2c8a3fcc0bce30f624601a851
Teardown zonal cluster
push time in 2 days
pull request commentmatplotlib/matplotlib
Add a code of conduct link to github
@story645 - do you think the Numpy one is OK, then, for Matplotlib (once names have been changed)? I do agree that the code of conduct should provide mechanisms for reporting, and stated procedures for when things go wrong.
comment created time in 2 days
push eventmatthew-brett/uob-jupyterhub
commit sha cc7dcc096318bab46e81de721cf27bc6a871d0c1
Zonal cluster
push time in 2 days
push eventmatthew-brett/uob-jupyterhub
commit sha 0c28ba6c08043bf73bac9e9897191c7f9e9b7575
Correct authentication command
commit sha 7a1de35731405db48986d09ef584163147224d10
Revert "Clarify assumptions for disks in test rig" This reverts commit a966ffff03d5cb3f4084aa87eeb35687b2f98dd7.
push time in 2 days
push eventmatthew-brett/my-exercise
commit sha 40db662a4325ca10035effd28bfead70f2464254
Make changes
push time in 3 days
issue commentmatthew-brett/multibuild
It's a bit dangerous to build wheels against the RCs, because the binary interface could still change, though it usually doesn't, I think.
Needed is Manylinux, and a macOS binary installer ...
comment created time in 3 days
PR opened data-8/materials-x18
Link seems to have changed from earlier editions.
pr created time in 4 days
fork matthew-brett/materials-x18
Publicly available materials that are used in the Data 8X Foundations of Data Science course on EdX in 2018. Generated by data-8/assignments (private).
fork in 4 days
Pull request review commentmatthew-brett/multibuild
Friendly error when curl HTTP errors downloading macOS python
function install_mac_cpython { local py_inst=$(pyinst_fname_for_version $py_version $py_osx_ver) local inst_path=$DOWNLOADS_SDIR/$py_inst mkdir -p $DOWNLOADS_SDIR- curl $MACPYTHON_URL/$py_stripped/${py_inst} > $inst_path+ # catch 404 and other curl errors+ set +e
Sorry to return to this when my mind is cluttered with other stuff, but can't you do something on the lines of:
curl -L https://no.site/index.html > bar || (retval=$? && if [ "$retval" == 6 ]; then echo "Could not resolve host"; fi)
?
comment created time in 4 days
push eventmatthew-brett/uob-jupyterhub
commit sha a966ffff03d5cb3f4084aa87eeb35687b2f98dd7
Clarify assumptions for disks in test rig
push time in 5 days
push eventmatthew-brett/uob-jupyterhub
commit sha 26384ba7a18ebcb2f4bf66bb21c6a06326b3671a
Show gcloud utility
push time in 5 days
push eventmatthew-brett/uob-jupyterhub
commit sha 66bb347d6da336a78624f62a61c337516efc2396
Generalize cluster name
push time in 5 days
push eventmatthew-brett/uob-jupyterhub
commit sha ebfb3a117e6954945a3cd1f1fabf8e6129a9f023
Remove stray copy of vars.sh
commit sha 45a849360443a45916079a0f56a048f1a7c94e23
Make explicit that deployment is template Remarks to help use of testing YaML files for storage.
push time in 5 days
push eventmatthew-brett/uob-jupyterhub
commit sha 96de3d2a504cfc08c0932b145bd487340b629cc8
Clean commented lines from YaML
commit sha 396f74dfa8ccdb0534cdf8b95a28ccb084349245
NFS server needs to be variable-completed in order to get disk name.
push time in 5 days
push eventmatthew-brett/uob-jupyterhub
commit sha 3219ce7fc72beeaeb00446b1893f8b2c30ec1c80
Set up kubernetes environment.
push time in 5 days
push eventmatthew-brett/uob-jupyterhub
commit sha 69428f70e2ff2b745d0b386fad5b45833c183cc5
Update notes on scaling
push time in 6 days
pull request commentmatthew-brett/multibuild
update openblas download from anaconda
Getting a build failure - ?for PyPy - where it's trying to access https://anaconda.org/multibuild-wheels-staging/openblas-libs/v0.3.10/download/openblas-v0.3.10-manylinux2014_x86_64.tar.gz - that doesn't exist. Actually, I can't see that file at https://anaconda.org/multibuild-wheels-staging/openblas-libs/files - did I miss it? If not, should it exist?
comment created time in 6 days
push eventmatthew-brett/uob-jupyterhub
commit sha 2c91b2af1fc5ebaa3305e943c86f460fca96517a
Refactoring disk name into vars.sh
push time in 6 days
push eventmatthew-brett/multibuild
commit sha 0995725ce65cff471d6a7c25dcc312630b71ce7b
bump python 2.7 latest to 2.7.18
commit sha fd774f23c175f845bd4d5d7786bf825ec4fc2f1a
fix build matrix
commit sha 369d7414efa78c3a65c30a20b4cbef45fbfc5286
Update .travis.yml Co-authored-by: Andrew Murray <[email protected]>
commit sha 8a94823757e4caed8a0017a97be40c0493e4a8b4
Merge pull request #351 from robbuckley/py2718 MRG: Update python 2.7 to 2.7.18 Very minor changes over 2.7.17, but a couple are worth having for anyone still using 2.7 see https://github.com/python/cpython/blob/2.7/Misc/NEWS.d/2.7.18rc1.rst
push time in 6 days
PR merged matthew-brett/multibuild
very minor changes over 2.7.17, but a couple are worth having for anyone still using 2.7 see https://github.com/python/cpython/blob/2.7/Misc/NEWS.d/2.7.18rc1.rst
pr closed time in 6 days
pull request commentmatthew-brett/multibuild
Yes - thanks for the reminder.
comment created time in 6 days
push eventmatthew-brett/uob-jupyterhub
commit sha 80c1b6c47969dd67a22bc38c8a3fb7ed9fb31821
Comments on variable substitution.
push time in 7 days
push eventmatthew-brett/uob-jupyterhub
commit sha 5d62439831d4e332b7cd778478385a401a517878
Comment on variable filling
push time in 7 days
push eventmatthew-brett/uob-jupyterhub
commit sha 655e73a1cc0b89a641ea056eef560973823fffed
Comment on Docker image source
push time in 7 days
issue commentmatthew-brett/multibuild
LAPACK library is not found for linux builds
@mattip - sorry to ping you at once - but I'm a bit out of date on Lapack now - any thoughts?
comment created time in 8 days
issue commentmatthew-brett/multibuild
Does multibuild work with versioneer?
Yes - a new issue would be good. I'm a little out of the loop on LAPACK these days - when you submit the issue, I'll @ mention Mattip - who has done the most recent work on that.
comment created time in 8 days
push eventmatthew-brett/trusty
commit sha 20dd33d66a30f012f09473e667bb39a1fc9a6b87
Updated comment [ci skip]
commit sha dfc8fdd2de49189511f7e8c1cd898974a7f557eb
Updated Python version [ci skip]
commit sha b0c5330540347f9eaf4c1ef4f519850078a535fe
Merge pull request #9 from radarhere/patch-1 MRG: Updated comment As of c6a009e#diff-f36d243f9dd7dad91d339286d74fd515, it is no longer 3.8rc1, just 3.8
push time in 8 days
PR merged matthew-brett/trusty
As of https://github.com/matthew-brett/trusty/commit/c6a009eedb557f4157e5428b4093133a4e99363d#diff-f36d243f9dd7dad91d339286d74fd515, it is no longer 3.8rc1, just 3.8
pr closed time in 8 days
issue commentmatthew-brett/multibuild
Does multibuild work with versioneer?
Sorry for the comment-and-run - but just in case it helps debugging - I seem to remember that there is a problem with versioneer and submodules. The standard way to use multibuild is with your project as a submodule of your wheel-building repo. I wonder if you get any goodness from removing your project as a submodule, and then git cloning it in your travis.yml instead?
comment created time in 9 days
pull request commentmatthew-brett/trusty
Well spotted. Any chance of fixing the README while you're there?
comment created time in 9 days
push eventmatthew-brett/nb2plots
commit sha 3a3a429e5b1be1eb136a959e151aeb377b7ca545
Drop 3.4 testing Not compatibile with requirements: ``` pyrsistent requires Python '>=3.5' ```
push time in 11 days
push eventmatthew-brett/myconfig
commit sha 70cd603f98cd74e61f27ff8d01701c51a71cd1c0
Default CRAN repo; comment disable workspace save.
commit sha 3a0f22df9f1446738d061f36446fdcd53857515b
Make Python 3 default virtualenv.
commit sha 1dc88f915039672d7f52ce1dd0c1132c5a647bc0
Disable Cmd-F1 screen mirror shortcut
commit sha dd757a0a768043bdd1a91994c0af8276387e325b
op Python snippet
commit sha 598254073b0ad55cd3cb485664fe521a629d210f
Jupyter in spell
push time in 13 days
push eventresampling-stats/resampling-with
Deployment Bot (from Travis CI)
commit sha 0a06c451c4bb5ef3940897851c5302a26b7b8d7d
Deploy resampling-stats/resampling-with to github.com/resampling-stats/resampling-with.git:gh-pages
push time in 13 days
push eventresampling-stats/resampling-with
commit sha 1a42b89fd2af54065b0a9bab85b6b7bd27dfdc7e
Fix rebuild_chapter script
push time in 13 days
issue commentnumpy/numpy
Isn't the URL fixed in latest multibuild 'devel'?
comment created time in 13 days
issue commentnumpy/numpy
aarch64 manylinux wheel and wrong pagesize
You can always re-upload the wheels with a new build tag : https://github.com/MacPython/wiki/wiki/Build-Tags
comment created time in 14 days
push eventmatthew-brett/multibuild
commit sha 2261816905c2838167a9693fc0f4ff9560bbe55a
Test get_cmake
commit sha 2d0726690f88124c82fe52682f037fb085917374
Fixed running test_library_builders for build_multilinux
commit sha e72d0a5885b15528a47188aeb99b01ed904596a2
Install cmake instead if cmake28 cannot be found in yum
commit sha a905f4126c324f15242ee88995876ef5a0bcf808
Skip building xz on manylinux
commit sha 589a67010e2605e535871359a01e69f9035efb6c
Deprecated get_cmake, renaming it to get_modern_cmake
commit sha 0faec5eb8089e0734a9b79279bf1b721b36c882d
Added ensure_xz instead of always skipping on manylinux
commit sha ce7e7367fd18023f30a6d82f38f1d3421d9145a1
Added comment [ci skip]
commit sha c395b10e8395d43b51209760dc0853303ecfcd9e
Merge pull request #342 from radarhere/cmake MRG: Install cmake instead if cmake28 cannot be found in yum When attempting to build on aarch64, I find that cmake28 is not found. However, cmake does exist. So this PR adds `get_modern_cmake` that searches for cmake28, and if it is not found, then use cmake instead.
push time in 18 days
PR merged matthew-brett/multibuild
When attempting to build on aarch64, I find that cmake28 is not found. However, cmake does exist.
So this PR changes get_cmake to search for cmake28, and if it is not found, then use cmake instead.
pr closed time in 18 days
pull request commentmatthew-brett/multibuild
Install cmake instead if cmake28 cannot be found in yum
OK - thanks - let's do it.
comment created time in 18 days
pull request commentmatthew-brett/multibuild
Install cmake instead if cmake28 cannot be found in yum
Testing now, I agree, cmake28 exists on Manylinux 1 and 2010, but not 2014. On 2010 yum install cmake28 is identical to bare cmake in effect - installing cmake 2.8.12. So I think it is true that get_modern_cmake installs CMake >= 2.8 for all Manylinux versions (and macOS).
OK to put a comment in the function? Then I'm all OK with this. Any other thoughts?
comment created time in 18 days
pull request commentnumpy/numpy-tutorials
Fully automate dev setup with Gitpod
Just as a meta point - over at JupyterLab, we were worrying that it might be unhealthy for the ecosystem if Microsoft Visual Studio Code starts to dominate in the online notebook market. Is that a reason to prefer Gitpod, regardless of convenience? I believe that Gitpod is based on Theia, with a wider governance governance community than VSCode.
comment created time in 20 days
push eventmatthew-brett/transforms3d
commit sha 83ef928a73acb501b65ebcbb6e2c3094505d1e3e
Documentation error.
commit sha 52321496a0d98f1f698fd3ed81f680d740202553
Merge pull request #35 from FSund/euler2mat_doc_fix MRG: Documentation error in transforms3d.euler.euler2mat Fixes #34
push time in 20 days
issue closedmatthew-brett/transforms3d
euler2mat documentation says it can return 4x4
According to the transforms3d.euler.euler2mat documentation it returns
mat : array-like shape (3, 3) or (4, 4)
But as far as I can see it will only return 3x3.
M = np.eye(3)
if repetition:
M[i, i] = cj
M[i, j] = sj*si
M[i, k] = sj*ci
M[j, i] = sj*sk
M[j, j] = -cj*ss+cc
M[j, k] = -cj*cs-sc
M[k, i] = -sj*ck
M[k, j] = cj*sc+cs
M[k, k] = cj*cc-ss
else:
M[i, i] = cj*ck
M[i, j] = sj*sc-cs
M[i, k] = sj*cc+ss
M[j, i] = cj*sk
M[j, j] = sj*ss+cc
M[j, k] = sj*cs-sc
M[k, i] = -sj
M[k, j] = cj*si
M[k, k] = cj*ci
return M
Is this an error in the docs, or the code, or is there something I'm not seeing?
closed time in 20 days
FSund
pull request commentmatthew-brett/transforms3d
Documentation error in transforms3d.euler.euler2mat
Great - thank you.
comment created time in 20 days
issue commentmatthew-brett/transforms3d
euler2mat documentation says it can return 4x4
Oops - yes - looks like a documentation error. Do you have any interest in a little pull request to fix that?
comment created time in 20 days
push eventresampling-stats/resampling-with
Deployment Bot (from Travis CI)
commit sha f42549f70d601b7c882da2cc90643cef5aa48698
Deploy resampling-stats/resampling-with to github.com/resampling-stats/resampling-with.git:gh-pages
push time in 21 days
push eventresampling-stats/resampling-with
commit sha 64b6f29581f37e519a1e7c14cb2859d304baa89d
Fix Hacking ref.
push time in 21 days
push eventresampling-stats/resampling-with
Deployment Bot (from Travis CI)
commit sha 14fad1a1aeb4e82e34d96b61bddc06979c5f2524
Deploy resampling-stats/resampling-with to github.com/resampling-stats/resampling-with.git:gh-pages
push time in 21 days
push eventresampling-stats/resampling-with
commit sha a0763b22fdf230ff5c83e8395e11002f83792077
Rename bibcheck target
commit sha c1b764409180c0b5a1582e8ab991731e5a02d766
References for MC, Hacking
push time in 21 days
Pull request review commentmatthew-brett/multibuild
matrix: # OSX builds - os: osx env:- - MB_PYTHON_VERSION=2.7+ - MB_PYTHON_VERSION=2.7.17 # 2.7.18 doesnt have a 10.6 build - MB_PYTHON_OSX_VER=10.6 - TEST_BUILDS=1 - OPENSSL_ROOT=openssl-1.0.2u - OPENSSL_HASH=ecd0c6ffb493dd06707d38b14bb4d8c2288bb7033735606569d8f90f89669d16- - os: osx- env:- - MB_PYTHON_VERSION=2.7- - MB_PYTHON_OSX_VER=10.9- - TEST_BUILDS=1
I agree - I think there's no need to remove this job. But otherwise, I think it's good to go. Any other thoughts?
comment created time in 21 days
push eventrobbuckley/multibuild
commit sha 369d7414efa78c3a65c30a20b4cbef45fbfc5286
Update .travis.yml Co-authored-by: Andrew Murray <[email protected]>
push time in 21 days
push eventresampling-stats/resampling-with
Deployment Bot (from Travis CI)
commit sha d9c2c6e9a9d43e79cc331c87a22e5659ebfb2afe
Deploy resampling-stats/resampling-with to github.com/resampling-stats/resampling-with.git:gh-pages
push time in 23 days
push eventresampling-stats/resampling-with
commit sha 790c647143c44e744450f860c5c8504f350b0527
Remove make check-bib Biber not installed, and would pull in TeXlive.
push time in 23 days
push eventresampling-stats/resampling-with
commit sha 691da50698d91b30fcdcd6331c3ff989ab02366c
Fix bibliography, run make check before main build
push time in 23 days
push eventresampling-stats/resampling-with
commit sha 5d3171144b74434071292827464e6ef8d0c2ce28
BF: work round setuptools 50.0 bug See https://github.com/pypa/setuptools/issues/2352 and https://github.com/pypa/setuptools/blob/17cb9d6bf249cefe653d3bdb712582409035a7db/CHANGES.rst#v5000
push time in 23 days
push eventresampling-stats/resampling-with
commit sha 29f5b95cc03fcdd72dffc0f08b27786a9ab61b3f
Kroese MC article
push time in 23 days
issue closednumpy/numpy
Can not run Python script in powerbi desktop
<!-- Please describe the issue in detail here, and fill in the fields below --> I have create csv file named "dataset" and write a python script to import it to Powerbi desktop. When I get data in powerbi desktop and paste above script, PBI show the error as following: "Details: "ADO.NET: Python script error. C:\Users\Admin\anaconda3\lib\site-packages\numpy_init.py:140: UserWarning: mkl-service package failed to import, therefore Intel(R) MKL initialization ensuring its correct out-of-the box operation under condition when Gnu OpenMP had already been loaded by Python process is not assured. Please install mkl-service package, see http://github.com/IntelPython/mkl-service from . import distributor_init Traceback (most recent call last): File "PythonScriptWrapper.PY", line 2, in <module> import os, pandas, matplotlib File "C:\Users\Admin\anaconda3\lib\site-packages\pandas_init.py", line 17, in <module> "Unable to import required dependencies:\n" + "\n".join(missing_dependencies) ImportError: Unable to import required dependencies"_
Reproducing code example:
<!-- A short code example that reproduces the problem/missing feature. It should be self-contained, i.e., possible to run as-is via 'python myproblem.py' --> My python script is: "import pandas as pd import numpy as np import matplotlib from matplotlib import pyplot as plt df = pd.read_csv(r'C:\Users\Admin\Desktop\Python\Python_to_powerbi\Dataset.csv') print(df)"
And I also attach my csv file. https://github.com/thuantv-vp/Python-and-powerbi/blob/master/Dataset.csv For more information I use Anaconda with python version is 3.7.4 and Numpy version is 1.18.1 Matplotlib is also installed
Please help me to fix this integration error. Many [thanks]
<< import pandas as pd
import numpy as np
import matplotlib
from matplotlib import pyplot as plt
df = pd.read_csv(r'C:\Users\Admin\Desktop\Python\Python_to_powerbi\Dataset.csv')
print(df)>>
<!-- Remove these sections for a feature request -->
Error message:
<!-- If you are reporting a segfault please include a GDB traceback, which you can generate by following https://github.com/numpy/numpy/blob/master/doc/source/dev/development_environment.rst#debugging -->
<!-- Full error message, if any (starting from line Traceback: ...) -->
Numpy/Python version information:
<!-- Output from 'import sys, numpy; print(numpy.version, sys.version)' -->
closed time in 23 days
thuantv-vp
issue commentnumpy/numpy
Can not run Python script in powerbi desktop
Closing for now, please do reopen if you believe this is due to Numpy.
comment created time in 23 days
issue commentnumpy/numpy
Can not run Python script in powerbi desktop
Are you sure this is a Numpy problem? What do you get if you start Python again, and do just import numpy?
comment created time in 24 days
pull request commentmatthew-brett/multibuild
MRG: document switch from Rackspace
OK to merge this one?
comment created time in a month
push eventmatthew-brett/multibuild
commit sha e44ccfeac5dcfaaef8abb519b7ddf9d1906de84f
Update Cython code links [skip ci]
push time in a month
Pull request review commentmatthew-brett/multibuild
MRG: document switch from Rackspace
Utilities for building on Travis CI and AppVeyor ################################################ +*****************************+Update: Uploads and Rackspace+*****************************++The original Multibuild default was to upload wheels to a Rackspace container,+where Rackspace kindly donated the hosting to the Scikit-learn team. We had+a URL pointing to the Rackspace container: `http://wheels.scipy.org`.++Rackspace finally stopped subsidizing this container, and the Rackspace of+`http://wheels.scipy.org` is no more. Projects using Multibuild have moved to+using https://anaconda.org/scipy-wheels-nightly/ for weekly uploads and+https://anaconda.org/multibuild-wheels-staging for staging wheels to PyPI.++Another option is to use GitHub for staging --- as do Cython for Travis CI and for+AppVeyor. Here is the NumPy code (for running on Travis CI) to upload to
Sorry - I did that too fast, and didn't notice that you'd done that above, with basically the same changes.
comment created time in a month
Pull request review commentmatthew-brett/multibuild
MRG: document switch from Rackspace
Utilities for building on Travis CI and AppVeyor ################################################ +*****************************+Update: Uploads and Rackspace+*****************************++The original Multibuild default was to upload wheels to a Rackspace container,+where Rackspace kindly donated the hosting to the Scikit-learn team. We had+a URL pointing to the Rackspace container: `http://wheels.scipy.org`.++Rackspace finally stopped subsidizing this container, and the Rackspace of+`http://wheels.scipy.org` is no more. Projects using Multibuild have moved to+using https://anaconda.org/scipy-wheels-nightly/ for weekly uploads and+https://anaconda.org/multibuild-wheels-staging for staging wheels to PyPI.++Another option is to use GitHub for staging --- as do Cython for Travis CI and for+AppVeyor. Here is the NumPy code (for running on Travis CI) to upload to
No - that was an accident of copy-paste from the Github interface. I've put them back now.
comment created time in a month
push eventmatthew-brett/multibuild
commit sha 855871563dbc1e282e0329244b665e14b8c4108f
Update Cython code links
push time in a month
push eventmatthew-brett/multibuild
commit sha 6b3db57e7f1f5255740493a19b3aea32bb9d17e7
Apply suggestions from review Thanks @hugovk Co-authored-by: Hugo van Kemenade <[email protected]>
push time in a month
issue commentMacPython/numexpr-wheels
Documented via https://github.com/matthew-brett/multibuild/pull/350 - please have a look?
comment created time in a month
push eventmatthew-brett/multibuild
commit sha 26305eb32be5d33ce97f12a43966146ab4af1f34
Document switch from Rackspace Thanks to @mattip
push time in a month
issue commentscipy/scipy
Use manylinux2014 to get aarch64/ppc64le support
As I remember it, when I uploaded Intell-architecture Numpy wheels for previous releases, it found someone whose pipeline depended on the fact that Numpy was installing from source for an older release. I suspect this is very unusual.
comment created time in a month
issue commentMacPython/numexpr-wheels
Thanks Matti - yes - I will use this link from now on - and will put it into the README for Multibuild.
comment created time in a month
push eventresampling-stats/resampling-with
Deployment Bot (from Travis CI)
commit sha 142cd4e3f137f14df1144b2a5e32d63190e64538
Deploy resampling-stats/resampling-with to github.com/resampling-stats/resampling-with.git:gh-pages
push time in a month
push eventresampling-stats/resampling-with
commit sha 7638bc46f3b75308e79a39ac3f43be9c04487c12
Efron not Bayesian
push time in a month
issue commentmatthew-brett/multibuild
wheels.scipy.org server not found
Ah - no - sorry - we haven't got round to documenting this yet, but the Rackspace account is now closed (they stopped subsidizing it).
Most of us have now moved to hosting wheels on Anaconda.org space.
A useful Numpy thread is here: https://mail.python.org/pipermail/numpy-discussion/2020-August/080915.html
Summarizing: Numpy has some example machinery for uploading to the Anaconda.org space:
https://github.com/MacPython/numpy-wheels/blob/master/azure/posix.yml#L87
@mattip - sorry - I bet I'm just being lazy here - but can you comment on permissions for the Anaconda upload?
comment created time in a month
pull request commentmatthew-brett/multibuild
Install cmake instead if cmake28 cannot be found in yum
Ah - that's interesting - I thought this was a fix for the very old cmake on manylinux1 - but not so?
I also thought that it was somewhat equivalent in practice to asking for a cmake >= 2.8 - also not so?
comment created time in a month
push eventresampling-stats/resampling-with
Deployment Bot (from Travis CI)
commit sha ccd6a589a7e8ea68a1645ae44e3e13e871ccdbb9
Deploy resampling-stats/resampling-with to github.com/resampling-stats/resampling-with.git:gh-pages
push time in a month