๐ก Please review Running DVC on Windows for important tips to improve your experience using DVC on Windows.
To use DVC as a Python library, please install with pip or with conda.
The easiest way to install from command line for most cases is to install
Chocolatey on your machine, and use the choco
command:
$ choco install dvc
Requires Miniconda or Anaconda Distribution.
You can use conda
from
Anaconda Prompt,
a POSIX-like command line terminal in Windows.
$ conda install -c conda-forge dvc
Currently, it supports Python 3.6-3.8
From Command Prompt or other recommended consoles:
$ pip install dvc
Depending on the type of the remote storage you
plan to use, you might need to install optional dependencies: [s3]
, [azure]
,
[gdrive]
, [gs]
, [oss]
, [ssh]
. Use [all]
to include them all.
An easy way is to use the self-contained, executable installer (binary), which is available from the big "Download" button on the home page. You can also get it from the release page on GitHub.
You'll need to download and run the installer again each time you want to update DVC. You may use Windows Uninstaller to remove the program.
Note that this method by default enables symlink permissions for all users, so they can use them to optimize DVC operations.