This package will help you perform ML on ANY tabular data. The result of modelling will be a weighted network describing which features were associated with which modelling method. This will help you understand:
- Feature Bias
- Useless Features
- Co-dependency of Features
- PCA efficiency
Apart from the above valuable information, this package requires minimal requirements.txt
. The odds are you probably have these already installed.
To install the package, you can use pip
with the URL of the GitHub repository.
git clone https://github.com/your-username/auto-sklearn.git
cd auto-sklearn
You can set the environment's name as you wish by replacing auto-env
.
python -m venv auto-env
source auto-env/bin/activate # On Windows: auto-env\Scripts\activate
Note that the -e
flag is important.
pip install -e .
import auto-sklearn
To learn intricacies of ML. ML is not a statistical method when dealing with data. I see it as a means to get some result through methodical data morphology. Results and my inference is what makes a good data analysis. To ensure that I am able to see through ML well enough that I can focus in honing inferential skills.
I am developing this project to blaze through basic modelling when I want to:
- Test new datasets.
- Experiment with modelling methods.
- Compare modelling with Deep Learning etc.
Finally, this package will help me write a Data Science blog in a very short time by reducing preliminary data testing time. This inturn helps me focus on novelty and creative aspects of Data Science. I also get to learn new areas of DS in a weekly basis.
You can find my blogs here: bhargavkantheti.com.
TBA