-
Create a droplet, use ssh key, 4GB ram
-
Connect to your Droplet via ssh https://docs.digitalocean.com/products/droplets/how-to/connect-with-ssh/openssh/
-
Do the initial server setup: create new user, access that user on your local via ss etc https://www.digitalocean.com/community/tutorials/initial-server-setup-with-ubuntu-20-04
-
On your user, setup python 3, pip3, python3-venv, and other essential dev tools https://www.digitalocean.com/community/tutorials/how-to-install-python-3-and-set-up-a-programming-environment-on-an-ubuntu-20-04-server
-
Follow as in the README (https://github.com/datature/portal), which will result to errors
git clone https://github.com/datature/portal
cd portal
./setup-virtualenv.sh
pip3 install -r requirements.txt
./setup.sh
python3 portal.py
- Resolve errors one by one:
-
https://stackoverflow.com/questions/55313610/importerror-libgl-so-1-cannot-open-shared-object-file-no-such-file-or-directo
sudo apt install libgl1-mesa-glx -
https://stackoverflow.com/questions/72441758/typeerror-descriptors-cannot-not-be-created-directly/72493690#72493690
pip install protobuf==3.20.* -
cs01/gdbgui#425
pip install werkzeug==2.0.0
- By this time you should be able to successfully run
python3 portal.py
* Running on http://127.0.0.1:9449/ (Press CTRL+C to quit)
- Do local port forwarding to open the GUI on your browser
ssh -L [LOCAL_IP:]LOCAL_PORT:DESTINATION:DESTINATION_PORT [USER@]SSH_SERVER
# in my case
ssh -L 8080:127.0.0.1:9449 [email protected] -i PATH/TO/SSH_KEY
Note: I'm not currently utilizing the GPU but I could try it in the future https://stackoverflow.com/questions/59823283/could-not-load-dynamic-library-cudart64-101-dll-on-tensorflow-cpu-only-install