Image Analysis Training Resources: Guidelines for Contributing

Contributions to this project are very welcome. Changes should be submitted by merge request from a new branch to master. Merge requests should be reviewed by at least one Maintainer before merging.

See README.md for information on the structure of the repository.

Step by step guide

To contribute to this project, please follow those steps:

  1. Clone this repository: git clone https://github.com/NEUBIAS/training-resources/
  2. On your computer, make a new branch. For example, if you would like to contribute python code to the binarization.md module you may: git checkout -b pythonBinarization
  3. Now add your changes on your computer (staying in this branch) - see “Adding a new module” section, below.
  4. When you are done, please git add .; git commit -m "some message"
  5. Now you can upload your branch to the online repository by typing: git push --set-upstream origin pythonBinarization.
  6. Go to the online repository on gitlab: https://git.embl.de/grp-bio-it/image-analysis-training-resources
  7. On gitlab, there will now be button at the top of the page. Click this button to stage a “merge request” of your contribution (in your branch) to the master branch. There will also a possibility to assign a project maintainer to review your contribution and to merge it. Please select someone appropriate here.
  8. Thank you for your contribution!

Adding a new module

Each module page is built from a template (_layouts/module.html), ensuring a consistent structure and style for the whole collection. To create a new module, you will need to add a few files in a few different places in this repository.

Module file

Most important is the module file itself. This module file should be saved with a short, descriptive name (no spaces!) ending with the .md (Markdown) extension. The content of this Markdown file can be limited to a header, written in YAML, according to the specification below. However, you may wish to add additional page content, written in Markdown, beneath the closing --- of this header. Any content written there will appear in the rendered page after the concept map and/or figure, and before the Activity section.

YAML Header Specification

All fields not marked as optional are required for the page to build. You can check that your YAML is valid with this tool.

---
title:     Title of the Module
layout:    module               # don't change this
prerequisites:
  - "a list of things that learners should know"
  - "in order to understand this module"
objectives:
  - "a list of learning objectives"
  - "see note 1 below for more info"
motivation: |
  A description of *why* you would want to learn this.
  Can be written in
  (GitHub-flavoured) [Markdown](https://github.com/adam-p/markdown-here/wiki/Markdown-Cheatsheet).
  Line breaks will be respected, so you can add lists etc using Markdown syntax.
concept_map: > # see note 2
  graph TD
      A[Christmas] -->|Get money| B(Go shopping)
      B --> C{Let me think}
      C -->|One| D[Laptop]
      C -->|Two| E[iPhone]
      C -->|Three| F[fa:fa-car Car];
figure: /figures/mymodule.png # store the example image for your module in the `figures` folder and provide the absolute path from the root of the site here.
figure_legend: Some description of the figure. (optional)
activity_preface: |
  Some general description of the activity for
  that learners will do while studying the module.
  It will be followed by platform-specific instructions/example code.
  (optional)
activities: # platform-specific activity instruction/example code files (see note 3) (optional)
  - ["ImageJ GUI", "mymodule/activities/mymodule_imagejgui.md", "markdown"]
  - ["ImageJ Macro": "mymodule/activities/mymodule_imagejmacro.ijm", "java"]
  - ["Jython", "mymodule/activities/mymodule_jython.py", "python"]
exercises: # platform-specific exercises (in Markdown files) (see note 4) (optional)
  "ImageJ GUI": "mymodule/exercises/mymodule_imagejgui.md"
  "ImageJ Macro": "mymodule/exercises/mymodule_imagejmacro.md"
  "Jython": "mymodule/exercises/mymodule_jython.md"
  "MATLAB": "mymodule/exercises/mymodule_matlab.md"
assessment: |
  Language-agnostic questions to assess learner understanding of the key concept
  covered in the module.
  (optional)
learn_next: # see note 5
  - "[name_of_one](calibration)"
  - "[or_more_modules](object_splitting)"
  - "[to link to next](display)"
external_links:
  - "[link to](https://external.page.com)"
---

Notes:

  1. Learning objectives should be worded as endings to a sentence beginning “After completing this lesson, learners should be able to…”. We recommend starting each learning objective with a verb from Bloom’s Taxonomy
  2. Concept maps are drawn with Mermaid.js. The indentation of the chart description is important, so be careful!
  3. The activities field should be populated with three-entry arrays, where the first value is the name of the platform, the second value is the path to the instructions or script for the activity, and the third is a lower-case language identifier to tell GitHub Pages if and how the content of the file should be highlighted. For files containing activity instructions written in Markdown - as opposed to a script written in some programming language - the value in the third position should be "markdown".
  4. The exercises field should be populated with key-value pairs, where the key is the name of the platform (e.g. “ImageJ GUI”, “Python”, etc) and the value is the path (relative to _includes/) to the file containing the exercises for that platform.
  5. The points in “Learn Next” are Markdown links, which should be formed as [Module Title](modulefilename), where the extension has been removed from the filename.

Associated files

Below is a list of all the other files that you should provide to accompany a new module, as well as the appropriate location for each (relative to the top level of the repository). Examples are given for a /modules/mymodule.md

Adding exercises/activity instructions for a new platform

Contributions of instructions and exercises for more platforms are very welcome - please see the “Associated files” subsection above for details of where these contributed files should be added.

Customising the material for a course

This repository is designed to work as a central reference point for all modules and platforms. However, we expect that instructors will want to prepare more specific material to be used in individual workshops/courses.

To do this, fork or import* this repository and follow the steps described below.

* Each GitHub user can only have a single fork of a repository so importing may be the better choice if you expect to make multiple customised sites from this material.

Setting a defualt platform

You can specify a default plaform for your site by adjusting the default-platform field in the site _config.yml. Beware: your choices will be limited by the availability of platform options for exercises and activities throughout the material! If you want to set a default platform that does not have exercises/activities prepared for every module, you should add the paths to any module files missing exercises/activities for your given platform to the exclude list in _config.yml. This will prevent any errors being raised due to a missing parameter for those modules.

Adjusting module set and order

Use the module_order field in _config.yml to specify which modules should appear in your site and in what order. Other modules not mentioned in this list will still be built (unless you add their paths to the exclude field) but will not be included in the page navigation i.e. via dropdown menu, next/previous buttons, etc.

Questions about the module layout

If you have questions about the module layout, please contact image-analysis-support@embl.de.

Building locally OSX

To test your changes locally, install jekyll on your system. Instructions for Mac OSX are here: https://jekyllrb.com/docs/installation/macos/.

Once you have jekyll and bundler setup, clone and move into this repository, and run:

make serve

All going well, your built pages are now beng served locally. Copy the URL provided in the output (should be http://127.0.0.1:4000/image-analysis-training-resources/) and paste it into your web browser. Now you can navigate around the locally-built version of the pages and check whether you’re happy to submit your changes to be merged into master :+1:

Building locally Windows (without make)

You need to install several tools (ruby and then jekyll). Please follow the instructions [https://jekyllrb.com/docs/installation/windows/] follow step 1-4

All commands are run from a standard/gitbash terminal.

Caveat: Some users must run their gem install commands as administrator. We do not know exactly why, but it may depend on their respective Windows installation.

See your new module

To see a module edit the file _config.yml and add your module (the file name in _modules with .md) to the list and the appropriate location

module_order:
- pixels
- spatial_calibration
- lut
- binarization
- connected_components
- datatypes
- measure_shapes
- workflow_segment_2d_nuclei_measure_shape
- measure_intensities
- global_background_correction
- filter_neighbourhood
- median_filter
- local_background_correction
- filter_objects
- workflow_segment_2d_noisy_nuclei_filter_objects_measure_shape
- MY_MODULE

Note that when jekyll is serving your page, changes on existing files are automatically updated.