python,kerasで学習予測のデータ表示

Question

前提・実現したいこと

日経株価の学習予測をしています。エラーが出て解決策が分からない状況です。

発生している問題・エラーメッセージ

Traceback (most recent call last):
  File "C:\Users\gakusyuuyosoku1.py", line 114, in <module>
    model.fit(X_train, Y_train,
  File "C:\Users\AppData\Local\Programs\Python\Python39\lib\site-packages\keras\engine\training.py", line 1158, in fit
    tmp_logs = self.train_function(iterator)
  File "C:\Users\AppData\Local\Programs\Python\Python39\lib\site-packages\tensorflow\python\eager\def_function.py", line 889, in __call__
    result = self._call(*args, **kwds)
  File "C:\Users\AppData\Local\Programs\Python\Python39\lib\site-packages\tensorflow\python\eager\def_function.py", line 950, in _call
    return self._stateless_fn(*args, **kwds)
  File "C:\Users\AppData\Local\Programs\Python\Python39\lib\site-packages\tensorflow\python\eager\function.py", line 3023, in __call__
    return graph_function._call_flat(
  File "C:\Users\AppData\Local\Programs\Python\Python39\lib\site-packages\tensorflow\python\eager\function.py", line 1960, in _call_flat
    return self._build_call_outputs(self._inference_function.call(
  File "C:\Users\AppData\Local\Programs\Python\Python39\lib\site-packages\tensorflow\python\eager\function.py", line 591, in call
    outputs = execute.execute(
  File "C:\Users\AppData\Local\Programs\Python\Python39\lib\site-packages\tensorflow\python\eager\execute.py", line 59, in quick_execute
    tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,
tensorflow.python.framework.errors_impl.UnimplementedError:  Cast string to float is not supported
         [[node mean_squared_error/Cast (defined at \AppData\Local\Programs\Python\Python39\lib\site-packages\keras\losses.py:1201) ]] [Op:__inference_train_function_3684]

Errors may have originated from an input operation.
Input Source operations connected to node mean_squared_error/Cast:
 IteratorGetNext (defined at \AppData\Local\Programs\Python\Python39\lib\site-packages\keras\engine\training.py:819)

Function call stack:
train_function

ソースコード

import numpy as np
import pandas as pd
from keras.models import Sequential
from keras.layers.core import Dense, Activation
from keras.layers.recurrent import LSTM
from keras.optimizers import Adam
from sklearn.model_selection import train_test_split

owarine = []
data = []
data2 = []
data3 = []
target = []
kairi = []
jyousyou = []
maxlen = 60
day1 = []
day2 = []
day3 = []
owarine = []

def normalization(a):
    b = []
    for i in a:
        b.append((i-min(a))/(max(a)-min(a)))
    return b

file=pd.read_csv('^nkx_d.csv',delimiter=',')
for i in range(len(file)):
   day1.append(file['Date'][i]) 
   owarine.append(file['Close'][i])

with open("kairiritu.txt","r",encoding="utf-8") as f:
    next(f)
    for line in f:
        data2 = line.split(' ')
        day2.append(str(data2[0]))
        kairi.append(float(data2[1]))

with open("jousyoudo.txt","r",encoding="utf-8") as f:
    next(f)
    for line in f:
        data3 = line.split(' ')
        day3.append(str(data3[0]))
        jyousyou.append(str(data3[1]))
'''
日付合わせ
'''
for i in range(len(owarine)-1,0,-1):
    if(day1[i] == day3[-1]):
        del day1[i+1:len(day1)]
        del owarine[i+1:len(owarine)]
        break
for i in range(len(kairi)-1,0,-1):
    if(day2[i] == day3[-1]):
        del day2[i+1:len(day2)]
        del kairi[i+1:len(kairi)]
        break
if(len(day2) > len(day3)):
    for i in range(0,len(owarine)):
        if(day1[i] == day3[0]):
            del day1[0:i]
            del owarine[0:i]
            break
    for i in range(0,len(kairi)-1):
        if(day2[i] == day3[0]):
            del day2[0:i]
            del kairi[0:i]
            break
else:
    for i in range(0,len(owarine)):
        if(day1[i] == day2[0]):
            del day1[0:i]
            del owarine[0:i]
            break
    for i in range(0,len(jyousyou)):
        if(day3[i] == day2[0]):
            del day3[0:i]
            del jyousyou[0:i]
            break
for i in range(0, len(kairi)-maxlen):
    data.append(normalization(kairi[i: i + maxlen]))
    target.append(jyousyou[i + maxlen])
X = np.array(data).reshape(len(data), maxlen, 1)
Y = np.array(target).reshape(len(data), 1) 
N_train = int(len(data) * 0.5)
N_validation = len(data) - N_train
X_train, X_validation, Y_train, Y_validation = train_test_split(X, Y, test_size=N_validation,shuffle=False)

'''
モデル設定
'''
n_in = len(X[0][0])  
n_hidden = 60
n_out = len(Y[0])  
def weight_variable(shape, name=None):
    return np.random.normal(scale=.01, size=shape)
model = Sequential()
model.add(LSTM(n_hidden,
                    kernel_initializer="random_uniform",
                    input_shape=(maxlen, n_in)))
model.add(Dense(n_hidden, kernel_initializer="random_uniform"))
model.add(Activation('sigmoid'))
model.add(Dense(n_out, kernel_initializer="random_uniform"))
model.add(Activation('sigmoid'))
optimizer = Adam(lr=0.001, beta_1=0.9, beta_2=0.999)
model.compile(loss='mean_squared_error',
              optimizer=optimizer)
'''
モデル学習
'''
epochs = 100
batch_size = 1000
model.fit(X_train, Y_train,
          batch_size=batch_size,
          epochs=epochs,
          validation_split = 0.25)          
y_ = model.predict(X_validation)
yosoku = []
for i in range (0,len(y_)):
    if (y_[i]>=0.5):
        yosoku.append('1')
    else:
        yosoku.append('0')        
#学習の予測データ表示
b = open("yosoku.txt","w",encoding="utf-8") 
b.write("日付 上昇度" + "\n")
print(len(yosoku))
for s in range(len(yosoku)):
    b.write(str(day3[(len(jyousyou))-(len(yosoku))+s])+" "+str(yosoku[s])+"\n")

補足情報

Python 3.9.6
Keras 2.4.3
numpy 1.19.5
pandas 1.3.0
tensorflow 2.5.0

Answer 1

目的変数Yの要素が文字列になってしまっており、文字列からの変換はサポートしていない、という意味のエラーが出ています。目的変数Yを数値にすることで、エラーは解消するでしょう。

<詳細>

以下で文字列リストjyousyouを、そのままtargetに組み込み、目的変数Yとしてしまっています。

for i in range(0, len(kairi)-maxlen):
    data.append(normalization(kairi[i: i + maxlen]))
    target.append(jyousyou[i + maxlen])
X = np.array(data).reshape(len(data), maxlen, 1)
Y = np.array(target).reshape(len(data), 1) 

エラーメッセージのポイントは以下です。（stringからfloatへのcastはサポートしていません）

tensorflow.python.framework.errors_impl.UnimplementedError:  Cast string to float is not supported

なお、jyousyouの中身が質問文からは判別不能ですので、具体的に数値にする方法まではわかりません。