Add pip, links to develop and docker to install. Update docker docs

Dockerfile

@@ -39,11 +39,12 @@ RUN useradd \
         -r \
         -s /sbin/nologin \
          geomag_user \
-    && mkdir -p /home/geomag_user/notebooks \
-    && chown -R geomag_user:geomag_user /home/geomag_user
+    && mkdir -p /data \
+    && chown -R geomag_user:geomag_user /data
 
 USER geomag_user
-WORKDIR /geomag-algorithms
+WORKDIR /data
 EXPOSE 8000
+
 # entrypoint needs double quotes
 ENTRYPOINT [ "/geomag-algorithms/docker-entrypoint.sh" ]

docker-entrypoint.sh

 #! /bin/bash
 
-# run jupyter in the background, and forward SIGTERM manually.
-# "exec" seems like a much simpler solution for this,
-# however, jupyter kernels die noisy deaths when using exec.
-
-# _term () {
-#  echo 'Caught SIGTERM'
-#  kill -TERM "$child"
-#}
-#trap _term SIGTERM
-
 # add geomagio to notebook path
 export PYTHONPATH=/geomag-algorithms
 
 # run jupyter notebook server
 exec jupyter notebook \
     --ip='*' \
-    --notebook-dir=/home/geomag_user/notebooks \
+    --notebook-dir=/data \
     --no-browser \
     --port=8000
-
-#child=$!
-#wait "$child"

docs/install.md

-Installation
-============
+# Installation
 
+> - [Develop](./develop.md) describes how to install locally.
+> - [Docker](./install_docker.md) describes container usage.
 
+## Requirements:
 
-1) Install Python and dependencies (obspy and pycurl)
+- Python 3.6 or newer
+- ObsPy 1.2.1 or newer
 
-    Using Anaconda is recommended ( https://conda.io/miniconda.html ).
+## Procedure:
 
-        conda env create -f geomagenv.yml
-        source activate geomagenv
+1.  Install Python and dependencies
 
-    or
+    - Option 1: Anaconda/Miniconda (https://conda.io/miniconda.html):
 
-        conda create --name=geomagenv --channel conda-forge python=3 obspy pycurl
-        source activate geomagenv
+          conda create --name=geomagenv --channel conda-forge python=3 obspy pycurl
+          source activate geomagenv
 
-2) Install **geomagio**
+    - Option 2: Pip
 
-        pip install git+https://github.com/usgs/geomag-algorithms.git
+      > `pyenv` can install specific/multiple versions of python
+      >
+      > - https://github.com/pyenv/pyenv
+      > - https://github.com/pyenv-win/pyenv-win
+
+          # create virtual environment (optional)
+          python -m venv path/to/geomagenv
+          source path/to/geomagenv/bin/activate
+
+          # install dependencies
+          pip install numpy
+          pip install obspy
 
+2.  Install **geomagio**
+
+        pip install git+https://github.com/usgs/geomag-algorithms.git
 
-3) Use **geomagio**
+3.  Use **geomagio**
 
     - Use the main script, `geomag.py -h`
     - In python scripts, `import geomagio`

docs/install_docker.md

 ## Docker
+
 https://www.docker.com/
 
 Docker containers bundle all dependencies needed to use geomag-algorithms.
 The container includes a Jupyter Notebook server for interactive development.
 
-> Docker images are built using the `Dockerfile` at the root of this project.
+### Run docker container
 
+> This page provides one example of how to run a container. See https://docs.docker.com/engine/reference/run/ for detailed `docker run` documentation.
 
-### Create a new docker container
+The following command creates and starts a container in the background.
 
-The following command creates and starts a container.
-```
-docker run -d --name geomagio -p 8000:80 usgs/geomag-algorithms
-```
+    docker run -rm -it --name geomagio -p 8000:8000 -v $(pwd)/data:/data usgs/geomag-algorithms
 
-- `-d` runs container in the background
+- `-rm` runs a temporary container
+- `-it` makes the container interactive (so you can stop it with `ctrl+c`)
 - `-name geomagio` assigns the name `geomagio`
-- `-p 8000:80` forwards system port `8000` to container port `80`
+- `-p 8000:8000` forwards system port `8000` to container port `8000`
+- `-v $(pwd)/data:/data` mounts a local data directory (./data) into the container so notebooks are saved
 - `usgs/geomag-algorithms:latest` refers to the
   latest version of the geomag-algorithms docker image
 
-  > Notebooks are stored in the container in the directory
-  > `/home/geomag_user/notebooks`
-
-
 ### Use the container
 
-- Start a stopped container:
-```
-docker start geomagio
-```
-
 - Run an interactive python prompt
-```
-docker exec -it geomagio python
-```
+
+
+      docker exec -it geomagio python
 
 - Use the Jupyter Notebook server
-```
-open http://localhost:8000/
-```
+
+  Check the output for a URL like this, that can be opened in a web browser: `http://127.0.0.1:8000/?token=...`
 
 - Use the `geomag.py` command line interface
-```
-docker exec -it geomagio geomag.py \
-    --inchannels H E Z F \
-    --input edge \
-    --interval minute \
-    --observatory BOU \
-    --output iaga2002 \
-    --output-stdout \
-    --starttime 2016-07-04T00:00:00Z \
-    --endtime 2016-07-04T23:59:00Z
-```
+
+      docker exec -it geomagio geomag.py \
+          --inchannels H E Z F \
+          --input edge \
+          --interval minute \
+          --observatory BOU \
+          --output iaga2002 \
+          --output-stdout \
+          --starttime 2016-07-04T00:00:00Z \
+          --endtime 2016-07-04T23:59:00Z
 
 - Stop a running container:
-```
-docker stop geomagio
-```
+
+  Press `Ctrl+C` and follow prompts to stop the container.
+
+### Build container
+
+Docker images are built using the `Dockerfile` at the root of this project.
+
+    docker build -t usgs/geomag-algorithms:TAG .