ssh [uni username]@mlis2.nottingham.ac.ukbash /shared/Anaconda3-2024.10-1-Linux-x86_64.sh% Why not miniconda?
- Press ENTER to confirm default install location (
/home/[uni username]/anaconda3) - Type
yeswhen asked to initialize conda - Then run:
source ~/.bashrcconda create --name picar2 python=3.9 -y
conda activate picar2pip install numpy pandas matplotlib scikit-learn opencv-python pillow tensorflow kaggleGet your API token from kaggle.com → Settings → API → Generate New Token, then:
export KAGGLE_API_TOKEN=KGAT_xxxxxxxxxxxxxxxxxxxxTo make it persist across sessions:
echo 'export KAGGLE_API_TOKEN=KGAT_xxxxxxxxxxxxxxxxxxxx' >> ~/.bashrcmkdir -p ~/picar-kaggle/data
kaggle competitions download -c machine-learning-in-science-ii-2026
unzip machine-learning-in-science-ii-2026.zip -d ~/picar-kaggle/data/% fix this to save data into project, where scripts expect it
ssh-keygen -t ed25519 -C "your_email@example.com"Press Enter for all prompts (default location, no passphrase). Then copy the public key:
cat ~/.ssh/id_ed25519.pubCopy the entire output line, go to github.com → Settings → SSH and GPG keys → New SSH key, paste it in, and save.
Test the connection:
ssh -T git@github.comType yes when prompted about host authenticity. You should see: Hi <username>! You've successfully authenticated...
cd ~
git clone git@github.com:lldvdll/PiCar.git
cd PiCar
cp -r ~/picar-kaggle/data/ ./data/Watch for nested folders — cp -r can create data/data/. Fix with:
mv data/data/* data/
rmdir data/dataAlso check for double-nested image folders (training_data/training_data/).
git config --global user.email "your_email@example.com"
git config --global user.name "Ningqian"Set remote to SSH (if cloned via HTTPS):
git remote set-url origin git@github.com:lldvdll/PiCar.gitEvery time you reconnect to the VM:
source ~/.bashrc
conda activate picar2
cd ~/PiCar