It is intended for junior high school students to university students who are new to AI.
I am developing a learning environment for reinforcement learning with chokozainerRL, but the flow of changing it with local git, sending it remotely, and checking it with colaboratory is quite troublesome.
When checking for simple module changes, if you have a local python environment, it's faster to check there. At that time, you can change it with git, send it to the remote, and install pip again.
Directly anaconda3/envs/chainer/lib/python3.7/site-packages/chokozainerrl If you change, it will be copied when you upload it with git, so ...
So I put a symbolic link.
ln -s /home/{user name}/chainerrl/chainerrl anaconda3/envs/chainer/lib/python3.7/site-packages/chainerrl
ln -s /home/{user name}/chokozainerrl/chokozainerrl anaconda3/envs/chainer/lib/python3.7/site-packages/chokozainerrl
Please change each {user name}. Now you can do a simple check. It's easier to do a simple check with your local jupyter notebook and then upload it to Colaboratory to check it. It seems that it takes time to upload chainerRL to PyPI after it is updated with git, so it is convenient to do this when trying a new one.
Recommended Posts