Perform Efficient ML/DL Modelling
Project description
Cerbo
Cerbo is "brain" in Esperanto.
It is a high-level API wrapping Scikit-Learn, Tensorflow and Keras. Allowing, you to efficiently perform ML modelling and preprocessing.
Install
Installing Cerbo:
pip install cerbo
or
python -m pip install cerbo
Writing your first program!
Currently, Cerbo performs efficient ML/DL modelling in a couple lines with limited preprocessing capabilites, we are adding new ones daily. Currently, to train a model from a CSV file all you have to do is call
from cerbo.preprocessing import *
data, col_names = load_custom_data("path_to_csv", "column_you_want_to_predict", num_features=4, id=False)
data is a dictionary containing X and y values, for training.
col_names is a list of features
Note: set id to true when there is an Id column in the CSV File, and set Num_Features to any value(as long it is within the # of colunns in the file"
After running this you will get 2 .png files labelled correlation, and features respectively.
- Correlation.png
- Will show a correlation matrix of all of the features in the CSV file
- feature.png
- Will show a Pandas Scatter Matrix of with a N x N grid with N being num_features.
To train a model on this data just do
gb, preds = Boosting("r", data, algo="gb", seed=42)
Which quickly trains a Gradient Boosting Regressor on this data.
You can also do
dt, preds = DecisionTree("c", data, seed=42)
To train a quick DT Classifier.
Authors
- Karthik Bhargav
- Siddharth Sharma
- Sauman Das
- Andy Phung
- Felix Liu
- Anaiy Somalwar
- Nathan Z.
- Aurko Routh
- Keshav Shah
- Navein Suresh
- Ayush Karupakula
- Ishan Jain
- Shrey Gupta
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distributions
Hashes for cerbo-0.1.2-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | e859e22ddc15de00b4c18e1ca0deb35c5557c0f6827ab97df33b2fb352fc05a8 |
|
MD5 | 0fb201746a9b812ec7f3991e2399b359 |
|
BLAKE2b-256 | 11b781cf277ad4430e82ec499b282525c51eda20ef8b51402f00f35cfb5f3480 |