1NeuronPerceptron-Pypi-mdnazmulislam0087 0.0.4

its an implimentation of Perceptron

-1NeuronPerceptron_Pypi

1Neuron|Perceptron|_Pypi

""" author: Nazmul email: [e-mail] """

How to use this

First install the library using below command by using latest version-

pip install 1NeuronPerceptron-Pypi-mdnazmulislam0087==0.0.4

Run the below code to see the training and plot file for or Gate, similarly you can use AND, NAND and XOR GATE to see the difference-

from oneNeuronPerceptron.perceptron import Perceptron
from oneNeuronPerceptron.all_utils import prepare_data, save_model, save_plot


import pandas as pd
import numpy as np
import logging
import os 

logging_str = "[%(asctime)s: %(levelname)s: %(module)s] %(message)s"
logging.basicConfig(level=logging.INFO, format=logging_str)



def main(data, eta, epochs, modelfilename,plotfilename):
    df = pd.DataFrame(data)
    logging.info(f"The dataframe is : {df}")
    X,y = prepare_data(df)
    model = Perceptron(eta=eta, epochs=epochs)
    model.fit(X, y)

    _ = model.total_loss()

    save_model(model, filename=modelfilename)
    save_plot(df, file_name=plotfilename, model=model)

if __name__=="__main__": # << entry point <<
    OR = {
        "x1": [0,0,1,1],
        "x2": [0,1,0,1],
        "y": [0,1,1,1],
    }
    
    ETA = 0.3 # 0 and 1
    EPOCHS = 10
    
    
    
    
    try:
        logging.info(">>>>> starting training >>>>>")
        main(data=OR, eta=ETA, epochs=EPOCHS, modelfilename="or.model", plotfilename="or.png")
        logging.info("<<<<< training done successfully<<<<<\n")
    except Exception as e:
        logging.exception(e)
        raise e  

Packages required-

1. matplotlib
2. numpy
3. pandas
4. joblib
5. tqdm

Limitation

Using one Neuron Perceptron, We cant make decision boundary for XOR GATe, In summary XOR Gate classification is not possible using one Neuron Perceptron

Reference -

official python docs

github docs for github actions

Read me editor

more details can be found

1Neuron Perceptron