train.py

import pandas as pd
import numpy as np
from nnet import YapaySinirAgi


def category_encode(etiketler, etiketCesit = 10):

    categorized_etiketler = []
    for i in etiketler:
        vector = np.zeros(etiketCesit)
        vector[i] = 1
        categorized_etiketler.append(vector)
    return np.array(categorized_etiketler)

def optimizasyon(X):
    X = X / 255
    X_matrix = []
    size = len(X)
    for i in range(size):
        X_matrix.append(np.array(X.iloc[i]).reshape(( 28, 28, 1)))
    return np.array(X_matrix)

dataset = pd.read_csv("dataset.csv")

y = dataset["label"]
y = category_encode(y)

X = dataset.drop("label", axis = 1)
X = optimizasyon(X)


def egitimVerisi():   pass

egitimVerisi.X = X
egitimVerisi.y = y


yapayzeka = YapaySinirAgi()
yapayzeka.egit(egitimVerisi,50)
yapayzeka.kaydet("ytu_egitilmis_model_50")