getting started with conda on aws

gistfile1.txt

# ssh to host
# download latest miniconda
wget bit.ly/get-miniconda
bash get-miniconda -b -p ~/miniconda
~/miniconda/bin/conda init bash

# exit shell
# ssh to host again

# install jupyter
conda create -n jupyter jupyter jupyterlab pandas numpy matplotlib holoviews

# Activate your new conda environment
conda activate jupyter

# create and edit your jupyter config so that you have a secure jupyter 
# server with a password that you can connect to
jupyter notebook --generate-config
# should output the following:
# Writing default config to: /home/ubuntu/.jupyter/jupyter_notebook_config.py

# Create your notebook password in IPython:
from notebook.auth import passwd; passwd()
# it will prompt you to enter your password. Copy the output string into your config path

# Edit your config.py file and set the following parameters:
c.NotebookApp.port = 8888
c.NotebookApp.password = <output from ipython passwd() command>
c.NotebookApp.ip = '*'

# Launch your notebook server:
jupyter notebook
# or `jupyter lab` if you prefer

# Go to your web browser and enter the public FQDN of your server 
# (should be the hostname you ssh'd to) and add :<port at the end>, 
# so it should be something like this:
ec2-111-222-333-444.compute-1.amazonaws.com:8888

# Type your password into the jupyter password prompt -- this is the same
# password you typed into the IPython session