수식의 변경

import FinanceDataReader as fdr

df = fdr.DataReader('005930', '2010')

df['Close'].plot()

위  수식의 그래프가

아래의 수식에서 작동 할 수 있도록 

변경이 가능 한지요?

도와 주시면 감사 하겠습니다.

                                  # https://github.com/WegraLee/deep-learning-from-scratch-3/blob/master/steps/step60.py    #   에서 참조 하였습니다.

if '__file__' in globals():

    import os, sys

    sys.path.append(os.path.join(os.path.dirname(__file__), '..'))

import numpy as np

import matplotlib.pyplot as plt

import dezero

from dezero import Model

from dezero import SeqDataLoader

import dezero.functions as F

import dezero.layers as L

max_epoch = 100

batch_size = 10

hidden_size = 200

bptt_length = 30

train_set = dezero.datasets.SinCurve(train=True)     

dataloader = SeqDataLoader(train_set, batch_size=batch_size)   

seqlen = len(train_set)

class BetterRNN(Model):

    def __init__(self, hidden_size, out_size):

        super().__init__()

        self.rnn = L.LSTM(hidden_size)                           

        self.fc = L.Linear(out_size)

    def reset_state(self):

        self.rnn.reset_state()

    def __call__(self, x):

        y = self.rnn(x)

        y = self.fc(y)

        return y

model = BetterRNN(hidden_size, 1)

optimizer = dezero.optimizers.Adam().setup(model)

for epoch in range(max_epoch):

    model.reset_state()

    loss, count = 0, 0

    for x, t in dataloader:

        y = model(x)

        loss += F.mean_squared_error(y, t)

        count += 1

        if count % bptt_length == 0 or count == seqlen:                                                          

            model.cleargrads()

            loss.backward()

            loss.unchain_backward()                      

            optimizer.update()

    avg_loss = float(loss.data) / count

    print('| epoch %d | loss %f' % (epoch + 1, avg_loss))

# Plot

xs = np.cos(np.linspace(0, 4 * np.pi, 1000))

model.reset_state()                                      

pred_list = []

with dezero.no_grad():

    for x in xs:

        x = np.array(x).reshape(1, 1)

        y = model(x)

        pred_list.append(float(y.data))

plt.plot(np.arange(len(xs)), xs, label='y=cos(x)')

plt.plot(np.arange(len(xs)), pred_list, label='predict')

plt.xlabel('x')

plt.ylabel('y')

plt.legend()

plt.show()