This image contains a sample application that performs change detection on Planet Labs Imagery. It being a docker image means that all the need python and system libraries are pre-packaged and the algorithm is ready to use.
When this image is run via
docker run timbr/sample_app1:latest is will start a Jupyter Notebook running on port
A running docker host is required to run this image. Once an instance of docker is running the following commands can be used to start and run the notebook:
docker pull timbr/sample_app1:latest docker run -it -p 8888:8888 timbr/sample_app1:latest visit http://192.168.99.100:8888/ # NOTE: this url may be different for your docker host
Once the container is running you can access the notebooks on port
8888 on your docker host. Typically this will look like:
Changing the port
You can change the port that the notebook starts up on by passing the
-p flag to the
docker run command like so:
docker run -p 8080:8888 -it timbr/sample_app1:latest
Mapping local directories
If you want to map a local directory onto the docker container you can use the
-v flag. This is useful for a few reasons: you can save downloaded images and reuse them across docker containers, results can be saved, and snapshots can potentially be accessed from your local machine.
docker run -p 8080:8888 -v ./your/local/dir:/notebooks -it timbr/sample_app1:latest