[ML] LSTM - Univariate Bidirectional LSTM Models

GITHUB

https://github.com/yiwonjae/Project_Lotto/blob/master/Book_001/p130.py

 

0. 목표

'''
data
[10, 20, 30, 40, 50, 60, 70, 80, 90]

X(input),       y(output)
10, 20, 30,     40
20, 30, 40,     50
30, 40, 50,     60
'''

 

1. DATA

raw_seq = np.asarray([10, 20, 30, 40, 50, 60, 70, 80, 90])

 

2. DATA 정제

X

import numpy as np
from numpy import ndarray


def split_sequence(sequence:ndarray, n_steps:int)->(ndarray, ndarray):

    x = []
    y = []

    for i in range(len(sequence)):

        if(i+n_steps>=len(sequence)):
            break

        x.append(sequence[i:i+n_steps])
        y.append(sequence[i+n_steps])


    return (np.asarray(x), np.asarray(y))


raw_seq = np.asarray([10, 20, 30, 40, 50, 60, 70, 80, 90])

n_steps = 3

(x, y) = split_sequence(raw_seq, n_steps)

print(x.shape)
print(y.shape)

n_features = 1

x = x.reshape(x.shape[0], n_steps, n_features)

 

3. 학습

from keras import Sequential
from keras.layers import Dense, LSTM, Bidirectional

model = Sequential()
model.add(Bidirectional(LSTM(50, activation='relu', input_shape=(n_steps, n_features))))
model.add(Dense(1))
model.compile(optimizer='adam', loss='mse')

model.fit(x, y, epochs=200, verbose=1)

4. 표시

x_input = np.asarray([70, 80, 90])      # expect 100
x_input = x_input.reshape(1, n_steps, n_features)
yhat = model.predict(x_input, verbose=1)

print(yhat) # [[100.722206]]