看样子你的代码应该如下：
import pandas as pd
import matplotlib.pyplot as plt
from keras import layers
import tensorflow as tf
from keras.models import Model
from keras.layers import Dense, Dropout, LSTM, Input, Activation, concatenate
import numpy as np
from keras import optimizers
from sklearn import preprocessing
np.random.seed(4)
tf.random.set_seed(4)
history_points = 50
def csv_to_dataset(csv_path):
data = pd.read_csv(csv_path)
data = data.drop("date", axis=1)
data = data.drop(0, axis=0)
data_normaliser = preprocessing.MinMaxScaler()
data_normalised = data_normaliser.fit_transform(data)
# using the last {history_points} open high low close volume data points, predict the next open value
ohlcv_histories_normalised = np.array(
[
data_normalised[i : i + history_points].copy()
for i in range(len(data_normalised) - history_points)
]
)
next_day_open_values_normalised = np.array(
[
data_normalised[:, 0][i + history_points].copy()
for i in range(len(data_normalised) - history_points)
]
)
next_day_open_values_normalised = np.expand_dims(
next_day_open_values_normalised, -1
)
next_day_open_values = np.array(
[
data[:, 0][i + history_points].copy()
for i in range(len(data) - history_points)
]
)
next_day_open_values = np.expand_dims(next_day_open_values_normalised, -1)
y_normaliser = preprocessing.MinMaxScaler()
y_normaliser.fit(np.expand_dims(next_day_open_values))
assert (
ohlcv_histories_normalised.shape[0] == next_day_open_values_normalised.shape[0]
)
return (
ohlcv_histories_normalised,
next_day_open_values_normalised,
next_day_open_values,
y_normaliser,
)
if __name__ == "__main__":
ohlcv_histories, next_day_open_values, unscaled_y, y_normaliser = csv_to_dataset(
'daily.csv'
)
test_split = 0.9 # the percent of data to be used for testing
n = int(ohlcv_histories.shape[0] * test_split)
# splitting the dataset up into train and test sets
ohlcv_train = ohlcv_histories[:n]
y_train = next_day_open_values[:n]
ohlcv_test = ohlcv_histories[n:]
y_test = next_day_open_values[n:]
unscaled_y_test = unscaled_y[n:]
lstm_input = Input(shape=(history_points, 5), name="lstm_input")
x = LSTM(50, name="lstm_0")(lstm_input)
x = Dropout(0.2, name="lstm_dropout_0")(x)
x = Dense(64, name="dense_0")(x)
x = Activation("sigmoid", name="sigmoid_0")(x)
x = Dense(1, name="dense_1")(x)
output = Activation("linear", name="linear_output")(x)
model = Model(inputs=lstm_input, outputs=output)
adam = optimizers.Adam(lr=0.0005)
model.compile(optimizer=adam, loss="mse")
from keras.utils import plot_model
# plot_model(model, to_file='model.png')
model.fit(
x=ohlcv_train,
y=y_train,
batch_size=32,
epochs=50,
shuffle=True,
validation_split=0.1,
)
evaluation = model.evaluate(ohlcv_test, y_test)
print(evaluation)
| 
发表于 2023-9-12 17:32
|
发表于 2023-9-12 17:46
