-
-
Notifications
You must be signed in to change notification settings - Fork 25.8k
pip install scikit-learn fails #8242
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Comments
pip install scikit-learn fails
Indeed, and the error message tells you why.
|
Shouldn't numpy, and scipy be listed in the requirements since they are required? |
See #7867 and discussion within. |
This still happens on WIndows 10 if numpy and scipy are installed. |
@jamesBrosnahan commenting on a closed PR will not attract contributors and therefore it will not help you to solve your problem. PS: I am also unsure that you will get the message error below with numpy installed, even in Windows 10.
|
Some weirdness around installing scikit-learn. See: * scikit-learn/scikit-learn#8242 * scikit-learn/scikit-learn#7867 Solution is to force numpy and scipy installation before requirements.txt is installed.
Some weirdness around installing scikit-learn. See: * scikit-learn/scikit-learn#8242 * scikit-learn/scikit-learn#7867 Solution is to force numpy and scipy installation before requirements.txt is installed.
* Add reformat.R to convert smartsurvey format. * Add feature generators. * Extract data from db not flat files. * Update README. * Add db_to_flat_file.py to create merged db dump. * Add urllookup as module. * Add urllookup tests on travis. * Add PII removal using scrubadub. * Dump to pickle objects. Note: DataFrameSelector outputs a numpy array, but it is easier to calculate date features using pandas prior to conversion to a numpy array. For this reason, create an argument as to whether the DataFrameSelector is dealing with date features or not, and then apply the date transforms if so, and output a numpy array. Some weirdness around installing scikit-learn. See: * scikit-learn/scikit-learn#8242 * scikit-learn/scikit-learn#7867 Solution is to force numpy and scipy installation before requirements.txt is installed.
* Add reformat.R to convert smartsurvey format. * Add feature generators. * Extract data from db not flat files. * Update README. * Add db_to_flat_file.py to create merged db dump. * Add urllookup as module. * Add urllookup tests on travis. * Add PII removal using scrubadub. * Dump to pickle objects. Note: DataFrameSelector outputs a numpy array, but it is easier to calculate date features using pandas prior to conversion to a numpy array. For this reason, create an argument as to whether the DataFrameSelector is dealing with date features or not, and then apply the date transforms if so, and output a numpy array. Some weirdness around installing scikit-learn. See: * scikit-learn/scikit-learn#8242 * scikit-learn/scikit-learn#7867 Solution is to force numpy and scipy installation before requirements.txt is installed.
Try without specifying version of scikit-learn as : |
What should a person do if the original computer an issue occurred on is mia? |
If you can't reproduce the error, we probably can't either.
|
Install numpy making sure it's compatible with the required scikit-learn. Then try reinstalling scikit-learn specifying the version you want - for example with 0.18.1 VERSION=0.18.1
pip install --force-reinstall --no-cache-dir scikit-learn==$VERSION |
Description
pip install scikit-learn
failsSteps/Code to Reproduce
python3.6 -m venv anenv
. ./anenv/bin/activate
pip install scikit-learn
Expected Results
scikit-learn installs
Actual Results
Versions
The text was updated successfully, but these errors were encountered: