Metadata-Version: 2.1
Name: blob_creator
Version: 3.1.3
Summary: Package to create dummy datasets for analysis tasks
Home-page: https://github.com/mckoh/blob_creator
Author: Michael Kohlegger
Author-email: michael@datenberge.org
License: UNKNOWN
Project-URL: Bug Tracker, https://github.com/mckoh/blob_creator/issues
Description: # Blob Dataset Creator
        
        [![Publish to PyPI](https://github.com/mckoh/blob_creator/actions/workflows/pypi.yml/badge.svg)](https://github.com/mckoh/blob_creator/actions/workflows/pypi.yml) [![Lint with PyLint](https://github.com/mckoh/blob_creator/actions/workflows/pylint.yml/badge.svg)](https://github.com/mckoh/blob_creator/actions/workflows/pylint.yml) [![Test with Pytest](https://github.com/mckoh/blob_creator/actions/workflows/pytest.yml/badge.svg)](https://github.com/mckoh/blob_creator/actions/workflows/pytest.yml)
        
        Blobs are little monsters that you can use to demonstrate data analysis procedures. They have a size, a weight, a color and a cuteness level. As you can see, a blob dataset contains all types of variables that you can find in the wild (nominal, ordinal and metric).
        
        ## Generator Output
        
        This generator creates a population of blobs. It delivers (a) an Excel data frame with every blob's properties, (b) a population plot that visualizes all blobs for presentation purposes and (c) a set of visualizations of the population data.
        
        ![Dataframe](https://github.com/mckoh/blob_creator/raw/main/static/dataframe.png)
        
        The **population plot** displays all blobs with color and size. In addition to that, it shows blob names. There are three different kinds of base images that can be used to visualize blobs.
        
        ![Blob population](https://github.com/mckoh/blob_creator/raw/main/static/population_alien.png)
        ![Blob population](https://github.com/mckoh/blob_creator/raw/main/static/population_monster.png)
        ![Blob population](https://github.com/mckoh/blob_creator/raw/main/static/populatio_boy.png)
        
        The **visualization plot** inklude a histogram of blob size and blob weights, as well as a bar chart with cuteness level counts and a scatter plot of size and weight.
        
        ![Blob population analysis](https://github.com/mckoh/blob_creator/raw/main/static/histograms.png)
        
        ## Use the Generator
        
        Blob populations can either have large (12) or small (1) **variability**, determined by the population's scatter index. Moreover, they can consist of a large **number** of monsters or can also be small, determined by the population's n.
        
        You can **generate a new population** by calling:
        
        ```python
        from blob_creator.core import BlobFactory
        
        # This instantiates a new blob factory
        blob_factory = BlobFactory(n=20, scatter=12, kind="monster")
        
        # This will create the blobs
        blob_factory.create_blobs()
        
        # This will export your dataset and your population plot to disk
        blob_factory.export_data()
        
        # Finally, the output directory can be cleaned-up
        blob_factory.delete_individual_pngs()
        ```
        
        ## Package Installation
        
        This package can be installed using Python's package index. Use the following command to do this:
        
        ```shell
        pip install blob-creator
        ```
        
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
