Unit Tests

The Archai project welcomes contributions through the implementation of unit tests using Pytest. If you are interested in contributing to the project in this way, please follow these steps:

1. Ensure that Pytest is installed. You can install it using pip install archai[tests].

2. Check out the Archai codebase and create a new branch for your changes. This will allow for easy submission of your code as a pull request upon completion.

3. Create a .py test file in the tests directory. For example, if writing a unit test for the archai/trainers/nlp/hf_training_args.py file, the corresponding path would be tests/trainers/nlp/test_hf_training_args.py.

4. Write test functions inside the created file. These functions should be named using the pattern: test_<name of function being tested>.

5. Test functions should utilize the assert statement to verify that the output of the tested function is correct. For example:

   from archai.trainers.nlp.hf_training_args import DistillerTrainingArguments
   
   def test_distiller_training_arguments():
       args = DistillerTrainingArguments("tmp")
       assert args.alpha == 0.5
       assert args.temperature == 1.0
   
       args = DistillerTrainingArguments("tmp", alpha=0.75, temperature=1.25)
       assert args.alpha == 0.75
       assert args.temperature == 1.25
   

6. Run your tests using the pytest command. This will automatically discover and execute all test functions in the tests directory.

7. To run a specific test file, use the pytest command followed by the path to the desired file. For example: pytest tests/trainers/nlp/test_hf_training_args.py.

After writing and running your tests, you may submit your code as a pull request to the main project repository. Please include a description of the changes made and any relevant issue references.