Getting Started

Warning: Doodler is still in active development and beta version. Check back later or listen to announcements on twitter for the first official release.

Doodler is a python program that is designed to run within a conda environment, accessed from the command line (terminal). It is designed to work on all modern Windows, Mac OS X, and Linux distributions, with python 3.6 or greater. It therefore requires some familiarity with process of navigating to a directory, and running a python script from a command line interface such as a Anaconda prompt terminal window, Powershell terminal window, git bash shell, or other terminal/command line interface.

The following instructions assume you are using the popular (and free) Anaconda or Miniconda python distribution.

Clone the github repo

git clone --depth 1 https://github.com/dbuscombe-usgs/doodle_labeller.git

There are a lot of folders and files when you clone the repository and you only need to know about some of them. You can ignore

• everything inside __pycache__
• everything inside website and docs (unless you plan to contribute and submit a pull request)
• Dockerfile and docker-compose.yml (they are for building the website)

The following files are python 'scripts' that you run from the command line

• doodler.py: the main program you call to annotate on imagery
• merge.py: the program you call to merge multiple sets of label images

The following files are for information and for internal use by the other programs:

• README.md (this file)
• funcs.py

The doodler.yml is used to install a conda environment on your computer - see below for more details. What remains are three folders:

• config: contains config files, written by you to use with the program (see here)
• data: contains input data (images) and output data (label images)
• examples: some example outputs

Create a conda environment

If you are a regular conda user, now would be a good time to

conda clean --all
conda update conda
conda update anaconda

Issue the following command from your Anaconda shell, power shell, or terminal window:

conda env create -f doodler.yml

Activate the conda environment

Use this command to activate the environment, in order to use it

conda activate doodler

If you are a Windows user (only) who wishes to use unix style commands, install m2-base

conda install m2-base

A video is available that covers the installation process:

Updating the software.

Occasionally, the software is updated. If you watch the repository on github, you can receive alerts about when this happens.

It is best practice to move your images from data/images and outputs from data/label_images. For example, you could create a separate folder of (e.g. “images_done”) inside data or another location.

For most users who have not modified the contents from the last git pull, to update should be a simple matter of carrying out another

git pull

However, when you have changes on your working copy, from command line use git stash:

git stash

If you need to see what is in your stash, use:

git stash list

This will stash your changes and clear your status report. Next do:

git pull

This will apply stashed changes back to working copy and remove the changes from stash unless you have conflicts. In the case of conflict, they will stay in stash so you can start over if needed:

git stash pop

A one-liner with no stash checking is:

git stash && git pull && git stash pop