![]() The starter edition will suffice for our experiments.įor the PySC2 library to work we need to download and extract some additional materials. Now go to the and get the StarCraft II installed. Now we finally are ready to install the PySC2. Or get back to it after you’ve closed your previous one, just type workon sc2 If you would like to switch to this virtual environment in another terminal tab, mkvirtualenv sc2 -p python3Īfter this command, your prompt should be prefixed with (sc2). Now we are ready to create a dedicated virtual environment which we’ll use for our further setup. Pay attention to the terminal output so there are no errors. (or whatever your shell’s preferred profile is). Source /usr/local/bin/virtualenvwrapper.shĪnd then load it for those lines to take effect source ~/.profile Pip3 install virtualenv virtualenvwrapperĪnd to make it available in your terminal console, add these lines to your shell’s profile # virtualenv and virtualenvwrapperĮxport VIRTUALENVWRAPPER_PYTHON=/usr/local/bin/python3 Let’s install and use the virtualenv mkdir ~/.virtualenvs It is useful for cases when you screw something up and would like to go “from the scratch”. The next step is not necessary but recommended. Now check that we have the right version python3 -V Then install the Python 3.6 brew install -ignore-dependencies Note: this command won’t remove the 3.7 version from your system. In case, you already have installed the Python 3.7 unlink it first brew unlink python We will use the python: update 3.6.5_1 bottle commit. We need to install the Python 3.6 for things to go smoothly later. That is being installed by default with the Homebrew. This way we update the brew package manager’s database.Īt the moment of writing this article, the TensorFlow does not support yet Python 3.7 (or whatever your preferred shell’s profile) for the added line to take effect, and then brew update Then press Ctrl + o, then enter, and finally Ctrl + x. Insert that line export PATH=/usr/local/bin:$PATH into the nano, If you have no preferred text editor, you can use nano nano ~/.profile If you are unsure you can take the “compromise” way and use the. Let’s make the installed brew command of the Homebrew project available in your terminal console.Īdd the following line export PATH=/usr/local/bin:$PATH Now we are ready to install the Homebrew /bin/bash -c "$(curl -fsSL )" Then accept the XCode license sudo xcodebuild -license accept In case it is not, here are the necessary steps.įirst, install the Command Line Tools for Xcode xcode-select -install The Homebrew is the preferred way to install packages for macOS, so make sure that it is installed. The below are described the steps necessary to set up the working environment for it, so let’s make it happen. You can learn more about the Reinforcement Learning by reading A Beginner’s Guide to Deep Reinforcement Learning article. The library does it by exposing the StarCraft II Machine Learning API as a Python RL Environment. “Scientifically” speaking, the PySC2 allows for Reinforcement Learning agents Which makes it possible to “rule” the StarCraft II with your own AI. This a bit of pretentious title is about the setup of the great PySC2 library
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |