Deep learning in data science
https://www.overleaf.com/9578595952hnrxvvymqytp
Open a terminal at folder and run
conda env create -f environment.yml
conda activate DD2424 && jupyter notebook
First pull updates from remote
git pull
You can always see which files are staged, updated and not tracked with
git status
Add the file to index
git add <file>
or add all files with
git add .
Then commit all staged files with an mandatory message
git commit -m "The script now handles the X case"
At last, upload to the remote
git push