pipeline.py

# Ploting packages
import matplotlib.pyplot as plt
from pandas.plotting import register_matplotlib_converters
register_matplotlib_converters()

# Date wrangling
from datetime import datetime, timedelta

# Data wrangling
import pandas as pd 

# The deep learning class
from deep_model import DeepModelTS

# Reading the configuration file
import yaml

# Directory managment 
import os

# Reading the hyper parameters for the pipeline
with open(f'{os.getcwd()}\\conf.yml') as file:
    conf = yaml.load(file, Loader=yaml.FullLoader)

# Reading the data 
d = pd.read_csv('input/DAYTON_hourly.csv')
d['Datetime'] = [datetime.strptime(x, '%Y-%m-%d %H:%M:%S') for x in d['Datetime']]

# Making sure there are no duplicated data
# If there are some duplicates we average the data during those duplicated days
d = d.groupby('Datetime', as_index=False)['DAYTON_MW'].mean()

# Sorting the values
d.sort_values('Datetime', inplace=True)

# Initiating the class 
deep_learner = DeepModelTS(
    data=d, 
    Y_var='DAYTON_MW',
    lag=conf.get('lag'),
    LSTM_layer_depth=conf.get('LSTM_layer_depth'),
    epochs=conf.get('epochs'),
    train_test_split=conf.get('train_test_split') # The share of data that will be used for validation
)

# Fitting the model 
model = deep_learner.LSTModel()

# Making the prediction on the validation set
# Only applicable if train_test_split in the conf.yml > 0
yhat = deep_learner.predict()

if len(yhat) > 0:

    # Constructing the forecast dataframe
    fc = d.tail(len(yhat)).copy()
    fc.reset_index(inplace=True)
    fc['forecast'] = yhat

    # Ploting the forecasts
    plt.figure(figsize=(12, 8))
    for dtype in ['DAYTON_MW', 'forecast']:
        plt.plot(
            'Datetime',
            dtype,
            data=fc,
            label=dtype,
            alpha=0.8
        )
    plt.legend()
    plt.grid()
    plt.show()   
    
# Forecasting n steps ahead   

# Creating the model using full data and forecasting n steps ahead
deep_learner = DeepModelTS(
    data=d, 
    Y_var='DAYTON_MW',
    lag=24,
    LSTM_layer_depth=64,
    epochs=10,
    train_test_split=0 
)

# Fitting the model 
deep_learner.LSTModel()

# Forecasting n steps ahead
n_ahead = 168
yhat = deep_learner.predict_n_ahead(n_ahead)
yhat = [y[0][0] for y in yhat]

# Constructing the forecast dataframe
fc = d.tail(400).copy() 
fc['type'] = 'original'

last_date = max(fc['Datetime'])
hat_frame = pd.DataFrame({
    'Datetime': [last_date + timedelta(hours=x + 1) for x in range(n_ahead)], 
    'DAYTON_MW': yhat,
    'type': 'forecast'
})

fc = fc.append(hat_frame)
fc.reset_index(inplace=True, drop=True)

# Ploting the forecasts 
plt.figure(figsize=(12, 8))
for col_type in ['original', 'forecast']:
    plt.plot(
        'Datetime', 
        'DAYTON_MW', 
        data=fc[fc['type']==col_type],
        label=col_type
        )

plt.legend()
plt.grid()
plt.show()