Open source scripts, reports, and preprints for in vitro biology, genetics, bioinformatics, crispr, and other biotech applications.
Open source scripts, reports, and preprints for in vitro biology, genetics, bioinformatics, crispr, and other biotech applications.
Search for:
Categories Categories Categories
Update/ install Python scikit learn latest development version in Windows
I am trying to replace/install my Python scikit-learn 0.17.0 with 0.18.dev0 latest development version in Windows, so that I can try out sklearn.neural_network.MLPClassifier . After reading and trying the advanced user manual here (scikit-learn.org/stable/developers/advanced_installation.html#install-bleeding-edge ), I am kinda stuck.
pip install -U scikit-learn => Still getting 0.17.0 but not 0.18.dev0 Could any guru provide step by step guides to install developmental Python 2.70 scikit-learn in windows? Thanks!
Answer #1:
Answer #1:
The only way to get the development version is to clone it through git, as follows:
pip install git+https://github.com/scikit-learn/scikit-learn.git For Windows, I highly recommend installing Anaconda (www.continuum.io/downloads ) because it installs all the scientific computing packages that you need for Windows.
Answer #2:
I have same problem with you. After trying several solutions, I finally get sklearn 0.18 worked by simply type:
pip install git+https://github.com/scikit-learn/scikit-learn.git Answer #2:
I have same problem with you. After trying several solutions, I finally get sklearn 0.18 worked by simply type:
pip install git+https://github.com/scikit-learn/scikit-learn.git
Answer #3:
conda install scikit-learn
seems to do the job of updating to the latest version.
you can find the package you are looking for there:
from sklearn.neural_network import MLPClassifier Answer #3:
conda install scikit-learn
seems to do the job of updating to the latest version.
you can find the package you are looking for there:
from sklearn.neural_network import MLPClassifier This is valid if you are using Anaconda, ofc.
#J-18808-Ljbffr