Metadata-Version: 2.1
Name: blitzml
Version: 0.3.0
Summary: A low-code library for machine learning pipelines
Author: AI Team
Author-email: AI Team <info@realpython.com>
Project-URL: Homepage, https://github.com/AhmedMohamed25/blitzml
Keywords: ml,machine learning,classification
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python
Classifier: Programming Language :: Python :: 3
Requires-Python: >=3.8
Description-Content-Type: text/markdown
Requires-Dist: joblib (>=1.2.0)
Requires-Dist: numpy (>=1.23.4)
Requires-Dist: pandas (>=1.5.1)
Requires-Dist: scikit-learn (>=1.1.3)
Provides-Extra: dev
Requires-Dist: pip-tools ; extra == 'dev'
Requires-Dist: pytest ; extra == 'dev'


# blitzml

Automate machine learning piplines rapidly


## How to install


```bash
  pip install blitzml
```


## Usage

```python
from blitzml.csv import Pipeline

dataset_folder = "auxiliary/data/" # contains train.csv and test.csv
ground_truth_path = "auxiliary/ground_truth.csv"
output_folder_path = "auxiliary/output/"

auto = Pipeline(dataset_folder, ground_truth_path, output_folder_path, classifier = 'RF', n_estimators = 50)

auto.preprocess()
auto.train_the_model()
auto.gen_metrics_dict()

metrics_dict = auto.metrics_dict
```

## Possible Classifiers 
- Random Forest 'RF'
- LinearDiscriminantAnalysis 'LDA'
- Support Vector Classifier 'SVC'
