training.py

from flask import Flask, session, jsonify, request
import pandas as pd
import numpy as np
import pickle
import os
from sklearn import metrics
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
import json

###################Load config.json and get path variables
with open('config.json','r') as f:
    config = json.load(f) 

dataset_csv_path = os.path.join(config['output_folder_path']) 
model_path = os.path.join(config['output_model_path']) 


#################Function for training the model
def train_model():
    data = pd.DataFrame(
        columns=[
            'corporation',
            'lastmonth_activity',
            'lastyear_activity',
            'number_of_employees',
            'exited']
        )
    #use this logistic regression for training
    model = LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,
                    intercept_scaling=1, l1_ratio=None, max_iter=100,
                    multi_class='ovr', n_jobs=None, penalty='l2',
                    random_state=0, solver='liblinear', tol=0.0001, verbose=0,
                    warm_start=False)
    
    for file in os.listdir(os.getcwd()+'/'+dataset_csv_path+'/'):
        if file.endswith('.csv'):
            read_data = pd.read_csv(os.getcwd()+'/'+dataset_csv_path+'/'+file)
            data = data.append(read_data)
        else:
            continue
    y = data.pop('exited').astype(int)
    X = data[['lastmonth_activity','lastyear_activity','number_of_employees']]

    #fit the logistic regression to your data
    model.fit(X, y)
    #write the trained model to your workspace in a file called trainedmodel.pkl
    if os.path.isdir(os.getcwd()+'/'+model_path):
        pickle.dump(model, open(model_path+'/trainedmodel.pkl', 'wb'))
    else:
        os.mkdir(os.getcwd()+'/'+model_path)
        pickle.dump(model, open(model_path+'/trainedmodel.pkl', 'wb'))

if __name__ == '__main__':
    train_model()