forecastfix.py

# -*- coding: utf-8 -*-
"""ForecastFix.ipynb

Automatically generated by Colaboratory.

Original file is located at
    https://colab.research.google.com/drive/14PCmGiF7VVNLDdlvnjlXs2KI5v4apF5V
"""

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import streamlit as st

csv_dataset = pd.read_excel("datawaldi2.xlsx")
csv_dataset.plot()
plt.show()

optimal_alpha = None
best_mse = None
db = csv_dataset.iloc[:, :].values.astype('float32')
mean_results_for_all_possible_alpha_values = np.zeros(9)
for alpha in range(0, 9):
    pt = np.mean(db[:, 0][0:5])
    mean_for_alpha = np.zeros(len(db))
    mean_for_alpha[0] = np.power(db[0][0] - pt, 2)
    for i in range(1, len(db)):
        pt = pt + ((alpha + 1) * 0.1) * (db[i - 1][0] - pt)
        mean_for_alpha[i] = np.power(db[i][0] - pt, 2)
    mean_results_for_all_possible_alpha_values[alpha] = np.mean(mean_for_alpha)
optimal_alpha = (np.argmin(mean_results_for_all_possible_alpha_values) + 1) * 0.1
best_mse = np.min(mean_results_for_all_possible_alpha_values)
print("Best MSE = %s" % best_mse)
print("Optimal alpha = %s" % optimal_alpha)

pt = np.mean(db[:, 0][0:5])
for i in range(1, len(db) + 1):
    pt = pt + optimal_alpha * (db[i - 1][0] - pt)
print("Next observation = %s" % pt)

forecast = np.zeros(len(db) + 1)
pt = np.mean(db[:, 0][0:5])
forecast[0] = pt
for i in range(1, len(db) + 1):
    pt = pt + optimal_alpha * (db[i - 1][0] - pt)
    forecast[i] = pt
plt.plot(db[:, 0],label = 'Data Asli')
plt.plot(forecast, label = 'Data Peramalan')
plt.legend()
plt.show()

st.header('UAS Forecasting')

fig, ax = plt.subplots(figsize=(16, 8))
ax.plot(
    csv_dataset["Penjualan Y"],
    marker="o",
    linewidth=2,
    color="#85CAF9"
)

st.subheader('Data Asli')
st.pyplot(fig)

fig2, ay = plt.subplots(figsize=(16, 8))
ay.plot(
    forecast[i],
    marker="o",
    linewidth=2,
    color="#FF8C00"
)

st.subheader('Data Forecast')
st.pyplot(fig2)