In [83]:
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline
In [84]:
data = pd.read_csv("data/EURUSD_daily.csv", index_col='Date')
In [85]:
data.index = pd.to_datetime(data.index)
In [86]:
data.plot(figsize=(16,9))
Out[86]:
<matplotlib.axes._subplots.AxesSubplot at 0x7fc500b5e208>
In [87]:
data['shifted'] = data.shift(-1)
In [88]:
data.columns = ['t', 't+1']
In [89]:
data.dropna(inplace=True)
In [90]:
data.head()
Out[90]:
t
t+1
Date
2000-01-03
1.0276
1.0299
2000-01-04
1.0299
1.0317
2000-01-05
1.0317
1.0299
2000-01-06
1.0299
1.0283
2000-01-07
1.0283
1.0256
In [91]:
# direction: 1 for rising 0 for falling
def isRising(row):
if row['t+1'] >= row['t']:
val = 1
else:
val = 0
return val
In [92]:
data['direction'] = data.apply(isRising, axis=1)
In [93]:
split_date = pd.Timestamp('10-10-2015')
train = data.loc[:split_date]
test = data.loc[split_date:]
In [94]:
from sklearn.preprocessing import MinMaxScaler
sc = MinMaxScaler()
train_scaled = sc.fit_transform(train)
test_scaled = sc.transform(test)
In [95]:
train_scaled_df = pd.DataFrame(train_scaled, columns=['t','t+1','dir'], index = train.index)
test_scaled_df = pd.DataFrame(test_scaled, columns=['t','t+1','dir'], index = test.index)
In [96]:
for s in range(1,60):
train_scaled_df['t-{}'.format(s)] = train_scaled_df['t'].shift(s)
test_scaled_df['t-{}'.format(s)] = test_scaled_df['t'].shift(s)
In [97]:
train_scaled_df.head()
Out[97]:
t
t+1
dir
t-1
t-2
t-3
t-4
t-5
t-6
t-7
...
t-50
t-51
t-52
t-53
t-54
t-55
t-56
t-57
t-58
t-59
Date
2000-01-03
0.259034
0.262013
1.0
NaN
NaN
NaN
NaN
NaN
NaN
NaN
...
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN
2000-01-04
0.262013
0.264344
1.0
0.259034
NaN
NaN
NaN
NaN
NaN
NaN
...
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN
2000-01-05
0.264344
0.262013
0.0
0.262013
0.259034
NaN
NaN
NaN
NaN
NaN
...
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN
2000-01-06
0.262013
0.259940
0.0
0.264344
0.262013
0.259034
NaN
NaN
NaN
NaN
...
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN
2000-01-07
0.259940
0.256443
0.0
0.262013
0.264344
0.262013
0.259034
NaN
NaN
NaN
...
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN
5 rows × 62 columns
In [98]:
X_train = train_scaled_df.dropna().drop('t+1', axis=1).drop('dir', axis=1)
y_train = train_scaled_df.dropna()[['dir']]
X_test = test_scaled_df.dropna().drop('t+1', axis=1).drop('dir', axis=1)
y_test = test_scaled_df.dropna()[['dir']]
In [99]:
#converts to numpy array
X_train = X_train.values
X_test= X_test.values
y_train = y_train.values
y_test = y_test.values
In [100]:
X_train.shape
Out[100]:
(4056, 60)
In [101]:
from keras.models import Sequential
from keras.layers import Dense
from keras.optimizers import Adam
from keras.utils.np_utils import to_categorical
import keras.backend as K
In [125]:
K.clear_session()
model = Sequential()
model.add(Dense(512, input_dim=60, activation='tanh'))
model.add(Dense(1024, activation='tanh'))
model.add(Dense(2048, activation='relu'))
model.add(Dense(1, activation='sigmoid'))
model.compile(loss='binary_crossentropy', optimizer='Adam', metrics=['accuracy'])
In [ ]:
In [ ]:
In [126]:
model.fit(X_train, y_train, batch_size=32, epochs=100, validation_split=0.1)
Train on 3650 samples, validate on 406 samples
Epoch 1/100
3650/3650 [==============================] - 0s - loss: 0.7143 - acc: 0.4992 - val_loss: 0.6966 - val_acc: 0.4631
Epoch 2/100
3650/3650 [==============================] - 0s - loss: 0.6982 - acc: 0.5175 - val_loss: 0.6951 - val_acc: 0.4631
Epoch 3/100
3650/3650 [==============================] - 0s - loss: 0.6931 - acc: 0.5395 - val_loss: 0.7396 - val_acc: 0.4631
Epoch 4/100
3650/3650 [==============================] - 0s - loss: 0.6952 - acc: 0.5068 - val_loss: 0.6948 - val_acc: 0.4631
Epoch 5/100
3650/3650 [==============================] - 0s - loss: 0.6980 - acc: 0.4975 - val_loss: 0.6993 - val_acc: 0.4631
Epoch 6/100
3650/3650 [==============================] - 0s - loss: 0.6934 - acc: 0.5184 - val_loss: 0.6942 - val_acc: 0.4631
Epoch 7/100
3650/3650 [==============================] - 0s - loss: 0.6927 - acc: 0.5186 - val_loss: 0.6950 - val_acc: 0.4631
Epoch 8/100
3650/3650 [==============================] - 0s - loss: 0.6926 - acc: 0.5186 - val_loss: 0.6955 - val_acc: 0.4631
Epoch 9/100
3650/3650 [==============================] - 0s - loss: 0.6928 - acc: 0.5186 - val_loss: 0.6955 - val_acc: 0.4631
Epoch 10/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6959 - val_acc: 0.4631
Epoch 11/100
3650/3650 [==============================] - 0s - loss: 0.6932 - acc: 0.5140 - val_loss: 0.6995 - val_acc: 0.4631
Epoch 12/100
3650/3650 [==============================] - 0s - loss: 0.6942 - acc: 0.5156 - val_loss: 0.6946 - val_acc: 0.4631
Epoch 13/100
3650/3650 [==============================] - 0s - loss: 0.6924 - acc: 0.5153 - val_loss: 0.6958 - val_acc: 0.4631
Epoch 14/100
3650/3650 [==============================] - 0s - loss: 0.6926 - acc: 0.5178 - val_loss: 0.6959 - val_acc: 0.4631
Epoch 15/100
3650/3650 [==============================] - 0s - loss: 0.6927 - acc: 0.5156 - val_loss: 0.6960 - val_acc: 0.4631
Epoch 16/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6961 - val_acc: 0.4631
Epoch 17/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6963 - val_acc: 0.4631
Epoch 18/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6964 - val_acc: 0.4631
Epoch 19/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6962 - val_acc: 0.4631
Epoch 20/100
3650/3650 [==============================] - 0s - loss: 0.6924 - acc: 0.5186 - val_loss: 0.6964 - val_acc: 0.4631
Epoch 21/100
3650/3650 [==============================] - 0s - loss: 0.6933 - acc: 0.5134 - val_loss: 0.6960 - val_acc: 0.4631
Epoch 22/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6962 - val_acc: 0.4631
Epoch 23/100
3650/3650 [==============================] - 0s - loss: 0.6924 - acc: 0.5186 - val_loss: 0.6962 - val_acc: 0.4631
Epoch 24/100
3650/3650 [==============================] - 0s - loss: 0.6934 - acc: 0.5038 - val_loss: 0.6975 - val_acc: 0.4631
Epoch 25/100
3650/3650 [==============================] - 0s - loss: 0.6924 - acc: 0.5186 - val_loss: 0.6964 - val_acc: 0.4631
Epoch 26/100
3650/3650 [==============================] - 0s - loss: 0.6931 - acc: 0.5134 - val_loss: 0.6922 - val_acc: 0.5369
Epoch 27/100
3650/3650 [==============================] - 0s - loss: 0.6930 - acc: 0.5104 - val_loss: 0.6969 - val_acc: 0.4631
Epoch 28/100
3650/3650 [==============================] - 0s - loss: 0.6955 - acc: 0.5093 - val_loss: 0.6968 - val_acc: 0.4631
Epoch 29/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6968 - val_acc: 0.4631
Epoch 30/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6970 - val_acc: 0.4631
Epoch 31/100
3650/3650 [==============================] - 0s - loss: 0.6927 - acc: 0.5186 - val_loss: 0.7011 - val_acc: 0.4631
Epoch 32/100
3650/3650 [==============================] - 0s - loss: 0.6949 - acc: 0.5189 - val_loss: 0.7023 - val_acc: 0.4631
Epoch 33/100
3650/3650 [==============================] - 0s - loss: 0.6944 - acc: 0.5184 - val_loss: 0.6987 - val_acc: 0.4631
Epoch 34/100
3650/3650 [==============================] - 0s - loss: 0.6922 - acc: 0.5137 - val_loss: 0.6961 - val_acc: 0.4631
Epoch 35/100
3650/3650 [==============================] - 0s - loss: 0.6928 - acc: 0.5142 - val_loss: 0.6965 - val_acc: 0.4631
Epoch 36/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6963 - val_acc: 0.4631
Epoch 37/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6964 - val_acc: 0.4631
Epoch 38/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6964 - val_acc: 0.4631
Epoch 39/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6968 - val_acc: 0.4631
Epoch 40/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6968 - val_acc: 0.4631
Epoch 41/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6968 - val_acc: 0.4631
Epoch 42/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6965 - val_acc: 0.4631
Epoch 43/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6965 - val_acc: 0.4631
Epoch 44/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6966 - val_acc: 0.4631
Epoch 45/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6966 - val_acc: 0.4631
Epoch 46/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6965 - val_acc: 0.4631
Epoch 47/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6962 - val_acc: 0.4631
Epoch 48/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6966 - val_acc: 0.4631
Epoch 49/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6964 - val_acc: 0.4631
Epoch 50/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6966 - val_acc: 0.4631
Epoch 51/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6963 - val_acc: 0.4631
Epoch 52/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6962 - val_acc: 0.4631
Epoch 53/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6962 - val_acc: 0.4631
Epoch 54/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6963 - val_acc: 0.4631
Epoch 55/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6963 - val_acc: 0.4631
Epoch 56/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6965 - val_acc: 0.4631
Epoch 57/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6967 - val_acc: 0.4631
Epoch 58/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6968 - val_acc: 0.4631
Epoch 59/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6966 - val_acc: 0.4631
Epoch 60/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6968 - val_acc: 0.4631
Epoch 61/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6967 - val_acc: 0.4631
Epoch 62/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6969 - val_acc: 0.4631
Epoch 63/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6971 - val_acc: 0.4631
Epoch 64/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6970 - val_acc: 0.4631
Epoch 65/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6966 - val_acc: 0.4631
Epoch 66/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6964 - val_acc: 0.4631
Epoch 67/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6962 - val_acc: 0.4631
Epoch 68/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6961 - val_acc: 0.4631
Epoch 69/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6962 - val_acc: 0.4631
Epoch 70/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6964 - val_acc: 0.4631
Epoch 71/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6966 - val_acc: 0.4631
Epoch 72/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6967 - val_acc: 0.4631
Epoch 73/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6966 - val_acc: 0.4631
Epoch 74/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6967 - val_acc: 0.4631
Epoch 75/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6970 - val_acc: 0.4631
Epoch 76/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6966 - val_acc: 0.4631
Epoch 77/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6966 - val_acc: 0.4631
Epoch 78/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6968 - val_acc: 0.4631
Epoch 79/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6964 - val_acc: 0.4631
Epoch 80/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6968 - val_acc: 0.4631
Epoch 81/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6966 - val_acc: 0.4631
Epoch 82/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6964 - val_acc: 0.4631
Epoch 83/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6967 - val_acc: 0.4631
Epoch 84/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6967 - val_acc: 0.4631
Epoch 85/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6967 - val_acc: 0.4631
Epoch 86/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6967 - val_acc: 0.4631
Epoch 87/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6964 - val_acc: 0.4631
Epoch 88/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6967 - val_acc: 0.4631
Epoch 89/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6963 - val_acc: 0.4631
Epoch 90/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6964 - val_acc: 0.4631
Epoch 91/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6966 - val_acc: 0.4631
Epoch 92/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6965 - val_acc: 0.4631
Epoch 93/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6966 - val_acc: 0.4631
Epoch 94/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6967 - val_acc: 0.4631
Epoch 95/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6966 - val_acc: 0.4631
Epoch 96/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6967 - val_acc: 0.4631
Epoch 97/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6969 - val_acc: 0.4631
Epoch 98/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6970 - val_acc: 0.4631
Epoch 99/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6966 - val_acc: 0.4631
Epoch 100/100
3650/3650 [==============================] - 0s - loss: 0.6925 - acc: 0.5186 - val_loss: 0.6966 - val_acc: 0.4631
Out[126]:
<keras.callbacks.History at 0x7fc4d78615c0>
In [127]:
y_pred = model.predict(X_test)
X_test.shape
Out[127]:
(461, 60)
In [128]:
y_pred
Out[128]:
array([[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901],
[ 0.51844901]], dtype=float32)
In [129]:
y_pred[:,0] = y_pred[:,0]>0.5
In [130]:
y_pred
Out[130]:
array([[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.],
[ 1.]], dtype=float32)
In [131]:
eval_df = pd.DataFrame(y_pred, columns=['predicted'])
In [132]:
eval_df['target'] = y_test
In [133]:
eval_df['result'] = eval_df['predicted'] == eval_df['target']
In [134]:
eval_df[eval_df['target']==False].count()
Out[134]:
predicted 220
target 220
result 220
dtype: int64
In [ ]:
In [ ]:
Content source: Robak23/forex_deep
Similar notebooks: