Skip to content

BenoitConnan/pipely

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

To install the library:

pip install pipely

1. About

The library pipely can execute any class or any sequance of classes in any order.

2. How it works

2.1. Declarative way

To build a pipeline with classes to execute, you have to create a config .yml file in a root directory in the following form:

steps:
    [step_name_1]:
        exec: [relative path to a file]:[class to execute]
    [step_name_2]:
        exec: [relative path to a file]:[class to execute]
    [step_name_3]:
        exec: [relative path to a file]:[class to execute]

Let's create a config file with the name collect.yml:

steps:
    preProcessing:
        exec: src/calculation.py:Calculate
    kMeans:
        exec: src/models.py:Kmeans
    hyperTuning:
        exec: src/tuning.py:GridSearch
        depends_on: 
            - preProcessing
            - kMeans

Pipely will be able to automatically detect which steps are independent and could be done in parallel. In this case, it will execute steps preProcessing and kMeans in parallel, and right after they are finished, start executing hyperTuning.

In order for the pipely to work,

  • the names of your [steps] in collect.yml should be unique;
  • the executable classes should have a __call__ method. For instance, if we open src/calculation.py and look at Calculate class that we trigger, we see call method in the end.
#src/calculation.py

class Calculate(object):
	def sum(self):
		a=2
		b=4
		self.c=a+b

	def divide(self):
		f=4
		self.d = self.c/f

    def show_result(self):
            print(self.d)

	def __call__(self):
		self.sum()
		self.divide()
        self.show_result()

After creating a configuration .yml file in your root directory, use the following command to trigger the pipeline in terminal:

python -m pipely.cli --config_file collect.yml

2.2. Imperative way

Pipely can also trigger a specific class from a specific .py file.

python -m pipely.cli --trigger_class path/to/file.py:TestClass

Below is an example of command that triggers a Calculate class from src/calculation.py file.

python -m pipely.cli --trigger_class src/calculation.py:Calculate

Again, Calculate class should have a __call__ method that executes desired class functions.

About

light pipeline library

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 100.0%