A template to help you start a new project in the SDyPy ecosystem.
To use this template, you have multiple options. The following two will cover most use cases:
You can use GitHub's templating functionality. A new repository will be created on GitHub for your project. Use this option if your project does not yet have an online repository.
Click the "Use this template" button on the project template Github repository (see image below).
Simply select and confirm a name for your new repository, and a copy of this template will be created for you.
You can now clone your new repository onto your local machine. If your new repository is located at
https://github.com/<your_name>/<my_new_project>
, for example:$ git clone https://github.com/<your_name>/<my_new_project>
A folder named
<my_new_project>
will be created on your machine. It is already setup with a connection to your new GitHub repository, and you can begin developing your package!If you already have a repository for your project, located for example at
https://github.com/<your_name>/<my_existing_project>
, you can use our template by cloning in onto you local machine. This downloads the files into a local folder, with a connection with the online repository already set up. Do this by running :$ git clone https://github.com/sdypy/sdypy_template_project
Our template files will be downloaded into the
sdypy_template_project
folder.You can now either copy these files into you existing local project folder, or connect the cloned repository in the
sdypy_template_project
folder with your existing online repository :$ git remote rm origin $ git remote add origin https://github.com/ladisk/<my_existing_project>.git
You are now setup to begin working on your project.
To begin development, install the required packages with :
$ python -m pip install -r requirements.dev.txt
Now you can replace the core source code modules in sdypy_template_project/
with your code.
Remember to also replace the poject name (sdypy_template_project
) with your own project name in the following files:
- setup.py
- README.rst
- CONTRIBUTING.rst
- the "sdypy_template_project" directory name
Consider adding unit-tests for your project by modifying the files, found in tests/
. The provided test file structure is setup to work with pytest.
To also use the sphinx documentation, modify files in docs/source
, or remove the docs/
folder and quickstart a fresh documentation version using the sphinx-quickstart
command (see Sphinx - Getting started for more info).
The project code is structured as follows:
- setup.py
- the Python setup script, used to package the project
- requirements.txt
- a list of packages, required to use this project
- requirements.dev.txt
- a list of packages, required to develop this project
- README.rst
- the main projecdt description / documentation file
- CONTRIBUTING.rst
- a document containing information for potential contrubutors (developers) of the package
- License
- the project License
- .travis.yml
- contains the set of instructions to run wit the TravisCI continuous integration service after the file repository has been updated
- .gitignore
- defines the files in the project directory to be excluded from version control
- tests/
- contains project unit-tests
- sdypy_template_project/
- contains the core project source code, separated into meaningful sub-modules
- examples/
- scripts, notebooks with examples to showcase the project
- docs/
- the documentation source and built files
(For a more complex and custumuzable project structure, see the Cookiecutter project.)
By setting up ReadTheDocs, your project documentation can automatically be built and puclished as a publicly available website.
To test your documentation locally, run the following (starting from the main project directory) :
$ cd docs
$ make clean
$ make html
Your documentation files will be built inside the docs/build/html
folder.
You can build your project and publish it to the Python Package Index with the following basic steps:
- Build you project source code :
$ python setup.py sdist bdist_wheel
The built project can be tested locally by installing the resulting .whl
file, found in the dist/
folder in a new virtual environemtn:
$ python -m virtualenv venv
$ venv/Scripts/activate
$ python -m pip install <sdypy_template_project-#>.whl
(replace <sdypy_template_project-#>
above with the actual .whl
file name).
- Upload the distribution files from
dist/
to PyPI :
$ python -m twine upload --repository-url https://test.pypi.org/legacy/ dist/*
(--repository-url https://test.pypi.org/legacy/
uploads the package to the test PyPI for testing. To publish you package to the main PyPI repository, simply ommit this option from the above command.)
For more information on the publishng process, see this simpel Python packaging tutorial.
- After that, the sdypy_template_project will be available on PyPI and can be installed with pip.
$ pip install sdypy_template_project
After installing sdypy_template_project you can use it like any other Python module.
Here is a simple example with the current example code:
import sdypy_template_project as iep
import numpy as np
import matplotlib.pyplot as plt
video = np.load('examples/speckle.npy', mmap_mode='r')
results = iep.get_displacements(video, point=[5, 5], roi_size=[7, 7])
plt.figure()
plt.plot(results[0], label='x [px]')
plt.plot(results[1], label='y [px]')
plt.legend()
plt.show()
You can also run this basic example by running the following command in the project base direcotry:
$ python -m examples.basic_example
The Read the Docs page provides the project documentation.