Data access and time-series statistics cyberseminar

Repo for "Data access and time-series statistics" WaterHackWeek cyberseminar. This seminar took place on February 7, 2019. The seminar series, including abstracts and links to seminar recordings, is available at https://www.cuahsi.org/education/cyberseminars/waterhackweek-cyberseminar-series/

Launch on to run and explore the Jupyter notebooks.

Conda environment and Jupyter

First make sure the miniconda or anaconda conda version is installed. See instructions below for miniconda. See below for installation instructions.

To install the conda environment used for running the Jupyter notebooks in this seminar, change to the directory where the environment.yml file is found, downloaded from this GitHub repository (or based on a git clone). Then, at the terminal, run:

conda env create -f environment.yml

An environment called whwtimeseries will be created. Note that this environment doesn't include Jupyter. It assumes you are running Jupyterlab (or jupyter notebook) using a different conda environment where Jupyterlab is installed.

Install miniconda and setup jupyter lab

Steps taken from https://geohackweek.github.io/preliminary/01-conda-tutorial/ Instructions for MacOSX and Windows are also available there.

On linux:

# Install miniconda
url=https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh
wget $url -O miniconda.sh
bash miniconda.sh -b -p $HOME/miniconda
export PATH="$HOME/miniconda/bin:$PATH"
conda update conda --yes

# Create a conda environment with jupyterlab
conda create -n jupyterlab -c conda-forge python=3.6 jupyterlab nb_conda_kernels

# Starting jupyter lab
source activate jupyterlab
# Then run jupyter lab
jupyter lab