model.py

# This file is showing how model process data form raw to training,
# for modelling there is saperate file
# importing Required libraries
import logging
import pandas as pd
from sklearn.preprocessing import StandardScaler
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
import pickle
import warnings
warnings.simplefilter(action='ignore', category=FutureWarning)

# log file initialization 
logging.basicConfig(filename='debug.log', level=logging.DEBUG,
                    format='%(asctime)s:%(levelname)s:%(message)s')

logging.debug(' Model.py File execution started ')

# loading database with pandas library
df = pd.read_csv("iris.csv")
logging.debug(' Database Loaded ')

# model featuring
X = df[["Sepal_Length", "Sepal_Width", "Petal_Length", "Petal_Width"]]
y = df["Class"]

# Data Spliting For model training
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3)


# training a Naive Bayes classifier
from sklearn.naive_bayes import GaussianNB
gnb = GaussianNB().fit(X_train, y_train)
gnb_predictions = gnb.predict(X_test)
accuracy = gnb.score(X_test, y_test)
print(accuracy)


# pkl export & finish log
pickle.dump(gnb, open("model.pkl", "wb"))
logging.debug(' Execution of Model.py is finished ')