LSTM预测风速

最新推荐文章于 2023-12-14 07:51:43 发布

zxzxwdwd

最新推荐文章于 2023-12-14 07:51:43 发布

阅读量1.1k

点赞数

分类专栏：深度学习文章标签： python lstm

本文链接：https://blog.csdn.net/zxzxwdwd/article/details/123995700

版权

深度学习专栏收录该内容

1 篇文章 0 订阅

订阅专栏

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.preprocessing import MinMaxScaler
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import LSTM, Dense, Activation
#创建LSTM模型
def create_model():
    model = Sequential()
    #输入数据的shape为(n_samples, timestamps, features)
    #隐藏层设置为256, input_shape元组第二个参数1意指features为1
    #下面还有个lstm，故return_sequences设置为True
    model.add(LSTM(units=256,input_shape=(None,1),return_sequences=True))
    model.add(LSTM(units=256))
    #后接全连接层，直接输出单个值，故units为1
    model.add(Dense(units=1))
    model.add(Activation('linear'))
    model.compile(loss='mse',optimizer='adam')
    return model

导入数据

df=pd.read_excel(r'C:\Users\张逊\Documents\WeChat Files\wxid_a8oh18cik1m211\FileStorage\File\2021-05\合肥历史天气(1).xlsx')

---------------------------------------------------------------------------

FileNotFoundError                         Traceback (most recent call last)

<ipython-input-2-6b583db91cf3> in <module>
----> 1 df=pd.read_excel(r'C:\Users\张逊\Documents\WeChat Files\wxid_a8oh18cik1m211\FileStorage\File\2021-05\合肥历史天气(1).xlsx')


C:\Anaconda\lib\site-packages\pandas\util\_decorators.py in wrapper(*args, **kwargs)
    297                 )
    298                 warnings.warn(msg, FutureWarning, stacklevel=stacklevel)
--> 299             return func(*args, **kwargs)
    300 
    301         return wrapper


C:\Anaconda\lib\site-packages\pandas\io\excel\_base.py in read_excel(io, sheet_name, header, names, index_col, usecols, squeeze, dtype, engine, converters, true_values, false_values, skiprows, nrows, na_values, keep_default_na, na_filter, verbose, parse_dates, date_parser, thousands, comment, skipfooter, convert_float, mangle_dupe_cols, storage_options)
    334     if not isinstance(io, ExcelFile):
    335         should_close = True
--> 336         io = ExcelFile(io, storage_options=storage_options, engine=engine)
    337     elif engine and engine != io.engine:
    338         raise ValueError(


C:\Anaconda\lib\site-packages\pandas\io\excel\_base.py in __init__(self, path_or_buffer, engine, storage_options)
   1079             else:
   1080                 ext = inspect_excel_format(
-> 1081                     content=path_or_buffer, storage_options=storage_options
   1082                 )
   1083 


C:\Anaconda\lib\site-packages\pandas\io\excel\_base.py in inspect_excel_format(path, content, storage_options)
    957 
    958     with get_handle(
--> 959         content_or_path, "rb", storage_options=storage_options, is_text=False
    960     ) as handle:
    961         stream = handle.handle


C:\Anaconda\lib\site-packages\pandas\io\common.py in get_handle(path_or_buf, mode, encoding, compression, memory_map, is_text, errors, storage_options)
    654         else:
    655             # Binary mode
--> 656             handle = open(handle, ioargs.mode)
    657         handles.append(handle)
    658 


FileNotFoundError: [Errno 2] No such file or directory: 'C:\\Users\\张逊\\Documents\\WeChat Files\\wxid_a8oh18cik1m211\\FileStorage\\File\\2021-05\\合肥历史天气(1).xlsx'

#取前1000行
df=df[['wind/m/sec']]
df=df.iloc[0:1000,:]
df

	wind/m/sec
0	2.67
1	2.73
2	2.26
3	1.44
4	2.17
...	...
995	4.12
996	5.41
997	4.43
998	3.43
999	3.30

1000 rows × 1 columns

data

---------------------------------------------------------------------------

NameError                                 Traceback (most recent call last)

<ipython-input-4-c5d84736ba45> in <module>
----> 1 data


NameError: name 'data' is not defined

数据归一化，构造训练数据结构

scaler_minmax = MinMaxScaler()
data = scaler_minmax.fit_transform(df)
infer_seq_length = 10#用于推断的历史序列长度

d = []
for i in range(data.shape[0]-infer_seq_length):
    d.append(data[i:i+infer_seq_length+1].tolist())
d = np.array(d)
d

array([[[0.35871404],
        [0.36886633],
        [0.2893401 ],
        ...,
        [0.44500846],
        [0.34179357],
        [0.19966159]],

       [[0.36886633],
        [0.2893401 ],
        [0.15059222],
        ...,
        [0.34179357],
        [0.19966159],
        [0.22165821]],

       [[0.2893401 ],
        [0.15059222],
        [0.27411168],
        ...,
        [0.19966159],
        [0.22165821],
        [0.37563452]],

       ...,

       [[0.08121827],
        [0.28257191],
        [0.35025381],
        ...,
        [0.60406091],
        [0.82233503],
        [0.65651438]],

       [[0.28257191],
        [0.35025381],
        [0.36209814],
        ...,
        [0.82233503],
        [0.65651438],
        [0.48730964]],

       [[0.35025381],
        [0.36209814],
        [0.321489  ],
        ...,
        [0.65651438],
        [0.48730964],
        [0.46531303]]])

划分训练集

split_rate = 0.9
X_train, y_train = d[:int(d.shape[0]*split_rate),:-1], d[:int(d.shape[0]*split_rate),-1]
y_train

array([[0.19966159],
       [0.22165821],
       [0.37563452],
       [0.50761421],
       [0.49746193],
       [0.22673435],
       [0.27241963],
       [0.30964467],
       [0.38240271],
       [0.21827411],
       [0.13705584],
       [0.34686971],
       [0.23350254],
       [0.54145516],
       [0.55837563],
       [0.26903553],
       [0.46869712],
       [0.38747885],
       [0.6142132 ],
       [0.84263959],
       [0.49069374],
       [0.12521151],
       [0.21658206],
       [0.26565144],
       [0.14890017],
       [0.22335025],
       [0.25380711],
       [0.27411168],
       [0.14213198],
       [0.31133672],
       [0.47038917],
       [0.43654822],
       [0.32656514],
       [0.54314721],
       [0.2927242 ],
       [0.50423012],
       [0.69204738],
       [0.43824027],
       [0.642978  ],
       [0.1928934 ],
       [0.6142132 ],
       [0.38747885],
       [0.61759729],
       [0.75972927],
       [0.52622673],
       [0.20812183],
       [0.23857868],
       [0.43485618],
       [0.34010152],
       [0.31133672],
       [0.17089679],
       [0.13536379],
       [0.1962775 ],
       [0.39593909],
       [0.24534687],
       [0.1962775 ],
       [0.29441624],
       [0.20304569],
       [0.48730964],
       [0.58714044],
       [0.45516074],
       [0.40778342],
       [0.31302876],
       [0.36379019],
       [0.39255499],
       [0.23011844],
       [0.48392555],
       [0.34686971],
       [0.23350254],
       [0.37732657],
       [0.16582064],
       [0.14551607],
       [0.68020305],
       [0.64467005],
       [0.60406091],
       [0.3248731 ],
       [0.41116751],
       [0.28426396],
       [0.01015228],
       [0.10998308],
       [0.14213198],
       [0.46192893],
       [0.74788494],
       [0.3891709 ],
       [0.22165821],
       [0.43485618],
       [0.59052453],
       [0.32825719],
       [0.27241963],
       [0.38071066],
       [0.60067682],
       [0.7394247 ],
       [0.46531303],
       [0.33333333],
       [0.0676819 ],
       [0.26734349],
       [0.43993232],
       [0.40101523],
       [0.25211506],
       [0.42808799],
       [0.70558376],
       [0.3536379 ],
       [0.45516074],
       [0.13536379],
       [0.38240271],
       [0.19796954],
       [0.36040609],
       [0.2605753 ],
       [0.84602369],
       [0.39593909],
       [0.46023689],
       [0.39255499],
       [0.43316413],
       [0.41116751],
       [0.68866328],
       [0.50423012],
       [0.19796954],
       [0.31810491],
       [0.35532995],
       [0.1641286 ],
       [0.24365482],
       [0.20981387],
       [0.29441624],
       [0.43147208],
       [0.59052453],
       [0.80541455],
       [0.63113367],
       [0.49746193],
       [0.34517766],
       [0.33333333],
       [0.14043993],
       [0.40778342],
       [0.20304569],
       [0.14382403],
       [0.33333333],
       [0.34517766],
       [0.2893401 ],
       [0.47546531],
       [0.39086294],
       [0.47715736],
       [0.16582064],
       [0.07952623],
       [0.1607445 ],
       [0.27411168],
       [0.25380711],
       [0.3536379 ],
       [0.34686971],
       [0.5177665 ],
       [0.73265651],
       [0.46700508],
       [0.46362098],
       [0.11844332],
       [0.09306261],
       [0.25380711],
       [0.24196277],
       [0.31133672],
       [0.49576988],
       [0.36209814],
       [0.37901861],
       [0.29610829],
       [0.40101523],
       [0.06598985],
       [0.25211506],
       [0.52284264],
       [0.59052453],
       [0.37732657],
       [0.24703892],
       [0.10490694],
       [0.07106599],
       [0.35025381],
       [0.43147208],
       [0.48392555],
       [0.33333333],
       [0.09813875],
       [0.36717428],
       [0.39424704],
       [0.54483926],
       [0.68020305],
       [0.37394247],
       [0.30795262],
       [0.09475465],
       [0.1928934 ],
       [0.321489  ],
       [0.46192893],
       [0.54822335],
       [0.52622673],
       [0.54145516],
       [0.63282572],
       [0.4213198 ],
       [0.32994924],
       [0.08629442],
       [0.2284264 ],
       [0.2571912 ],
       [0.39593909],
       [0.357022  ],
       [0.11844332],
       [0.15905245],
       [0.02368866],
       [0.08121827],
       [0.39763113],
       [0.48392555],
       [0.49238579],
       [0.43654822],
       [0.19796954],
       [0.08967851],
       [0.37394247],
       [0.47038917],
       [0.27749577],
       [0.14382403],
       [0.19458545],
       [0.19458545],
       [0.09137056],
       [0.27411168],
       [0.34179357],
       [0.31810491],
       [0.1928934 ],
       [0.12351946],
       [0.21996616],
       [0.44670051],
       [0.37732657],
       [0.43485618],
       [0.70219966],
       [0.89170897],
       [0.30626058],
       [0.11675127],
       [0.31810491],
       [0.2605753 ],
       [0.24196277],
       [0.40101523],
       [0.43993232],
       [0.16920474],
       [0.34348562],
       [0.83756345],
       [0.97292724],
       [0.40947547],
       [0.24196277],
       [0.30626058],
       [0.36379019],
       [0.36040609],
       [0.34179357],
       [0.65651438],
       [0.75126904],
       [0.37055838],
       [0.49915398],
       [0.33333333],
       [0.43993232],
       [0.30964467],
       [0.31472081],
       [0.18781726],
       [0.4213198 ],
       [0.23857868],
       [0.17935702],
       [0.27411168],
       [0.23519459],
       [0.25211506],
       [0.43147208],
       [0.50592217],
       [0.34179357],
       [0.29103215],
       [0.17935702],
       [0.21827411],
       [0.24534687],
       [0.21319797],
       [0.29103215],
       [0.29610829],
       [0.11167513],
       [0.10998308],
       [0.12690355],
       [0.39763113],
       [0.31641286],
       [0.3248731 ],
       [0.59052453],
       [0.32656514],
       [0.12690355],
       [0.07783418],
       [0.15566836],
       [0.11505922],
       [0.0676819 ],
       [0.05414552],
       [0.11167513],
       [0.35871404],
       [0.29103215],
       [0.24534687],
       [0.35194585],
       [0.26226734],
       [0.27918782],
       [0.56175973],
       [0.41285956],
       [0.10490694],
       [0.12521151],
       [0.07614213],
       [0.14213198],
       [0.25549915],
       [0.16582064],
       [0.10152284],
       [0.32825719],
       [0.07275804],
       [0.18950931],
       [0.07783418],
       [0.08121827],
       [0.49576988],
       [0.63959391],
       [0.91032149],
       [0.09306261],
       [0.21319797],
       [0.15228426],
       [0.14043993],
       [0.10998308],
       [0.2250423 ],
       [0.32994924],
       [0.11505922],
       [0.19458545],
       [0.321489  ],
       [0.82064298],
       [0.24027073],
       [0.06598985],
       [0.27918782],
       [0.11675127],
       [0.18950931],
       [0.14890017],
       [0.14043993],
       [0.27580372],
       [0.40270728],
       [1.        ],
       [0.59898477],
       [0.34010152],
       [0.08121827],
       [0.18274112],
       [0.24365482],
       [0.34856176],
       [0.1928934 ],
       [0.13367174],
       [0.18612521],
       [0.50930626],
       [0.59729272],
       [0.29780034],
       [0.12013536],
       [0.25549915],
       [0.31302876],
       [0.47038917],
       [0.40270728],
       [0.15566836],
       [0.17089679],
       [0.28087986],
       [0.35532995],
       [0.35025381],
       [0.17258883],
       [0.20642978],
       [0.51945854],
       [0.34179357],
       [0.39424704],
       [0.10998308],
       [0.46362098],
       [0.14382403],
       [0.53807107],
       [0.49069374],
       [0.23181049],
       [0.13536379],
       [0.1641286 ],
       [0.14382403],
       [0.20135364],
       [0.357022  ],
       [0.32318105],
       [0.33164129],
       [0.23350254],
       [0.24365482],
       [0.38071066],
       [0.30287648],
       [0.15736041],
       [0.23519459],
       [0.21827411],
       [0.48392555],
       [0.69543147],
       [0.27580372],
       [0.11336717],
       [0.30456853],
       [0.47546531],
       [0.43147208],
       [0.20981387],
       [0.14551607],
       [0.3536379 ],
       [0.4500846 ],
       [0.52961083],
       [0.42808799],
       [0.47208122],
       [0.28087986],
       [0.19458545],
       [0.19796954],
       [0.11505922],
       [0.17258883],
       [0.26226734],
       [0.10829103],
       [0.09306261],
       [0.44331641],
       [0.32825719],
       [0.18443316],
       [0.30118443],
       [0.84771574],
       [0.59390863],
       [0.3857868 ],
       [0.30118443],
       [0.25888325],
       [0.36886633],
       [0.24534687],
       [0.48054146],
       [0.54145516],
       [0.32656514],
       [0.50084602],
       [0.36209814],
       [0.60575296],
       [0.59729272],
       [0.33333333],
       [0.24534687],
       [0.85448393],
       [0.20812183],
       [0.03553299],
       [0.33502538],
       [0.19966159],
       [0.12690355],
       [0.48730964],
       [0.59898477],
       [0.63790186],
       [0.33671743],
       [0.62605753],
       [0.3857868 ],
       [0.31472081],
       [0.33671743],
       [0.68697124],
       [0.33333333],
       [0.10659898],
       [0.14720812],
       [0.33502538],
       [0.4534687 ],
       [0.5786802 ],
       [0.84094755],
       [0.33502538],
       [0.23181049],
       [0.40609137],
       [0.4500846 ],
       [0.65989848],
       [0.47715736],
       [0.33502538],
       [0.27241963],
       [0.37563452],
       [0.28257191],
       [0.26395939],
       [0.25211506],
       [0.24365482],
       [0.20642978],
       [0.17935702],
       [0.34348562],
       [0.27580372],
       [0.30287648],
       [0.29610829],
       [0.30964467],
       [0.4822335 ],
       [0.48730964],
       [0.2927242 ],
       [0.17597293],
       [0.1641286 ],
       [0.52453469],
       [0.678511  ],
       [0.56175973],
       [0.22335025],
       [0.14720812],
       [0.14043993],
       [0.11336717],
       [0.22673435],
       [0.24703892],
       [0.39932318],
       [0.39932318],
       [0.13705584],
       [0.29441624],
       [0.26395939],
       [0.29441624],
       [0.39255499],
       [0.08967851],
       [0.27918782],
       [0.55837563],
       [0.54483926],
       [0.53299492],
       [0.25380711],
       [0.30118443],
       [0.23519459],
       [0.20642978],
       [0.12013536],
       [0.11844332],
       [0.27749577],
       [0.15397631],
       [0.48900169],
       [0.30964467],
       [0.20642978],
       [0.38071066],
       [0.2284264 ],
       [0.26226734],
       [0.30456853],
       [0.26395939],
       [0.25549915],
       [0.37055838],
       [0.642978  ],
       [0.48730964],
       [0.28764805],
       [0.21658206],
       [0.48392555],
       [0.25211506],
       [0.29441624],
       [0.40101523],
       [0.34686971],
       [0.11336717],
       [0.11336717],
       [0.17258883],
       [0.11675127],
       [0.24703892],
       [0.40439932],
       [0.25888325],
       [0.60406091],
       [0.70219966],
       [0.36040609],
       [0.36040609],
       [0.36548223],
       [0.30795262],
       [0.17428088],
       [0.15736041],
       [0.17935702],
       [0.07106599],
       [0.19458545],
       [0.21658206],
       [0.13536379],
       [0.09475465],
       [0.12013536],
       [0.08967851],
       [0.18443316],
       [0.38409475],
       [0.53130288],
       [0.4856176 ],
       [0.37901861],
       [0.31810491],
       [0.15059222],
       [0.36886633],
       [0.21996616],
       [0.30964467],
       [0.65651438],
       [0.27580372],
       [0.28595601],
       [0.31641286],
       [0.25380711],
       [0.2284264 ],
       [0.2893401 ],
       [0.39932318],
       [0.25211506],
       [0.31641286],
       [0.18612521],
       [0.16243655],
       [0.14551607],
       [0.28257191],
       [0.36717428],
       [0.4179357 ],
       [0.51269036],
       [0.29780034],
       [0.29780034],
       [0.28087986],
       [0.10152284],
       [0.13705584],
       [0.17428088],
       [0.23350254],
       [0.1607445 ],
       [0.15566836],
       [0.26226734],
       [0.08798646],
       [0.10829103],
       [0.12013536],
       [0.2927242 ],
       [0.43824027],
       [0.52453469],
       [0.41116751],
       [0.13536379],
       [0.04568528],
       [0.14551607],
       [0.17089679],
       [0.18781726],
       [0.07614213],
       [0.19796954],
       [0.38240271],
       [0.26395939],
       [0.26565144],
       [0.25549915],
       [0.23011844],
       [0.20981387],
       [0.26226734],
       [0.2250423 ],
       [0.17597293],
       [0.0998308 ],
       [0.2893401 ],
       [0.31979695],
       [0.25549915],
       [0.        ],
       [0.14213198],
       [0.16751269],
       [0.23181049],
       [0.09306261],
       [0.18950931],
       [0.34856176],
       [0.26903553],
       [0.36548223],
       [0.36886633],
       [0.29441624],
       [0.21150592],
       [0.14213198],
       [0.3891709 ],
       [0.33502538],
       [0.27918782],
       [0.17089679],
       [0.31133672],
       [0.39763113],
       [0.36717428],
       [0.14551607],
       [0.41455161],
       [0.27241963],
       [0.22335025],
       [0.21319797],
       [0.35194585],
       [0.27241963],
       [0.25211506],
       [0.39424704],
       [0.40270728],
       [0.22673435],
       [0.40270728],
       [0.2927242 ],
       [0.29780034],
       [0.18781726],
       [0.61928934],
       [0.34010152],
       [0.11505922],
       [0.15566836],
       [0.12521151],
       [0.30456853],
       [0.24196277],
       [0.16582064],
       [0.07952623],
       [0.10152284],
       [0.12690355],
       [0.46531303],
       [0.38240271],
       [0.24534687],
       [0.15566836],
       [0.2927242 ],
       [0.34856176],
       [0.14551607],
       [0.16243655],
       [0.21319797],
       [0.17766497],
       [0.33333333],
       [0.35194585],
       [0.62436548],
       [0.46023689],
       [0.14043993],
       [0.10152284],
       [0.1607445 ],
       [0.21319797],
       [0.20642978],
       [0.18104907],
       [0.3857868 ],
       [0.46362098],
       [0.52284264],
       [0.30118443],
       [0.54822335],
       [0.15736041],
       [0.23688663],
       [0.30795262],
       [0.1607445 ],
       [0.14720812],
       [0.23181049],
       [0.22165821],
       [0.18443316],
       [0.04737733],
       [0.08629442],
       [0.24196277],
       [0.49576988],
       [0.2284264 ],
       [0.11675127],
       [0.12013536],
       [0.32825719],
       [0.69543147],
       [0.41285956],
       [0.18274112],
       [0.15397631],
       [0.10659898],
       [0.12521151],
       [0.15736041],
       [0.35194585],
       [0.20135364],
       [0.14043993],
       [0.30964467],
       [0.23857868],
       [0.13028765],
       [0.46362098],
       [0.3248731 ],
       [0.43824027],
       [0.357022  ],
       [0.34517766],
       [0.23350254],
       [0.20135364],
       [0.2284264 ],
       [0.17089679],
       [0.12690355],
       [0.53130288],
       [0.25042301],
       [0.37394247],
       [0.1607445 ],
       [0.26226734],
       [0.2284264 ],
       [0.50761421],
       [0.78172589],
       [0.58714044],
       [0.4534687 ],
       [0.60067682],
       [0.4500846 ],
       [0.16243655],
       [0.14551607],
       [0.09306261],
       [0.20473773],
       [0.10829103],
       [0.12013536],
       [0.30964467],
       [0.23350254],
       [0.47715736],
       [0.19458545],
       [0.17766497],
       [0.23181049],
       [0.24365482],
       [0.15228426],
       [0.10998308],
       [0.60406091],
       [0.25042301],
       [0.11505922],
       [0.30626058],
       [0.16243655],
       [0.2284264 ],
       [0.29780034],
       [0.0642978 ],
       [0.357022  ],
       [0.09137056],
       [0.0998308 ],
       [0.11167513],
       [0.19966159],
       [0.1319797 ],
       [0.17428088],
       [0.32994924],
       [0.38409475],
       [0.65989848],
       [0.54314721],
       [0.30287648],
       [0.18104907],
       [0.29441624],
       [0.1928934 ],
       [0.15059222],
       [0.29949239],
       [0.39086294],
       [0.34348562],
       [0.22335025],
       [0.19966159],
       [0.18781726],
       [0.33164129],
       [0.2250423 ],
       [0.24703892],
       [0.36379019],
       [0.47038917],
       [0.23857868],
       [0.30118443],
       [0.44670051],
       [0.16243655],
       [0.11167513],
       [0.21658206],
       [0.63451777],
       [0.17428088],
       [0.23688663],
       [0.26565144],
       [0.46023689],
       [0.52791878],
       [0.30626058],
       [0.73604061],
       [0.7072758 ],
       [0.61252115],
       [0.53637902],
       [0.46869712],
       [0.21996616],
       [0.19966159],
       [0.0998308 ],
       [0.0676819 ],
       [0.24873096],
       [0.13874788],
       [0.31979695],
       [0.30456853],
       [0.29780034],
       [0.69543147],
       [0.79357022],
       [0.30964467],
       [0.09475465],
       [0.41624365],
       [0.4534687 ],
       [0.45516074],
       [0.49238579],
       [0.24873096],
       [0.11167513],
       [0.45854484],
       [0.33502538],
       [0.17428088],
       [0.1319797 ],
       [0.12351946],
       [0.39932318],
       [0.15397631],
       [0.10659898],
       [0.26903553],
       [0.27411168],
       [0.10998308],
       [0.34010152],
       [0.62944162],
       [0.58375635],
       [0.39086294],
       [0.52453469],
       [0.38409475],
       [0.47038917],
       [0.2250423 ],
       [0.14213198],
       [0.29103215],
       [0.46700508],
       [0.43824027],
       [0.77664975],
       [0.40947547],
       [0.27749577],
       [0.45516074],
       [0.5465313 ],
       [0.15397631],
       [0.19796954],
       [0.357022  ],
       [0.36548223],
       [0.14720812],
       [0.24534687],
       [0.24703892],
       [0.47884941],
       [0.11844332],
       [0.59052453],
       [0.76649746],
       [0.5465313 ],
       [0.54822335],
       [0.66666667],
       [0.52284264],
       [0.46700508],
       [0.357022  ],
       [0.57360406],
       [0.43993232],
       [0.53468697],
       [0.20981387],
       [0.20642978],
       [0.3857868 ],
       [0.08629442],
       [0.39424704],
       [0.5143824 ],
       [0.0998308 ],
       [0.1285956 ],
       [0.30795262],
       [0.27411168],
       [0.21658206],
       [0.33840948],
       [0.08629442],
       [0.16582064],
       [0.34348562],
       [0.08629442],
       [0.26734349],
       [0.30626058],
       [0.30626058],
       [0.2927242 ],
       [0.50930626],
       [0.43147208],
       [0.32656514],
       [0.19966159],
       [0.37901861],
       [0.54145516],
       [0.19120135],
       [0.5786802 ],
       [0.63620981],
       [0.2250423 ],
       [0.50592217],
       [0.321489  ],
       [0.28087986],
       [0.64974619],
       [0.52791878],
       [0.47377327],
       [0.2893401 ],
       [0.3248731 ],
       [0.43147208]])

model =create_model()
model.fit(X_train, y_train, batch_size=20,epochs=100,validation_split=0.1)

Epoch 1/100
41/41 [==============================] - 1s 28ms/step - loss: 0.0356 - val_loss: 0.0273
Epoch 2/100
41/41 [==============================] - 1s 17ms/step - loss: 0.0285 - val_loss: 0.0287
Epoch 3/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0286 - val_loss: 0.0268
Epoch 4/100
41/41 [==============================] - 1s 17ms/step - loss: 0.0266 - val_loss: 0.0267
Epoch 5/100
41/41 [==============================] - 1s 17ms/step - loss: 0.0252 - val_loss: 0.0266
Epoch 6/100
41/41 [==============================] - 1s 17ms/step - loss: 0.0239 - val_loss: 0.0276
Epoch 7/100
41/41 [==============================] - 1s 17ms/step - loss: 0.0235 - val_loss: 0.0299
Epoch 8/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0251 - val_loss: 0.0291
Epoch 9/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0235 - val_loss: 0.0278
Epoch 10/100
41/41 [==============================] - 1s 17ms/step - loss: 0.0233 - val_loss: 0.0273
Epoch 11/100
41/41 [==============================] - 1s 17ms/step - loss: 0.0234 - val_loss: 0.0279
Epoch 12/100
41/41 [==============================] - 1s 17ms/step - loss: 0.0236 - val_loss: 0.0274
Epoch 13/100
41/41 [==============================] - 1s 17ms/step - loss: 0.0234 - val_loss: 0.0277
Epoch 14/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0240 - val_loss: 0.0273
Epoch 15/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0251 - val_loss: 0.0312
Epoch 16/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0242 - val_loss: 0.0282
Epoch 17/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0235 - val_loss: 0.0321
Epoch 18/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0238 - val_loss: 0.0281
Epoch 19/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0234 - val_loss: 0.0286
Epoch 20/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0233 - val_loss: 0.0273
Epoch 21/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0233 - val_loss: 0.0302
Epoch 22/100
41/41 [==============================] - 1s 17ms/step - loss: 0.0232 - val_loss: 0.0284
Epoch 23/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0235 - val_loss: 0.0288
Epoch 24/100
41/41 [==============================] - 1s 17ms/step - loss: 0.0235 - val_loss: 0.0280
Epoch 25/100
41/41 [==============================] - 1s 17ms/step - loss: 0.0236 - val_loss: 0.0273
Epoch 26/100
41/41 [==============================] - 1s 17ms/step - loss: 0.0232 - val_loss: 0.0297
Epoch 27/100
41/41 [==============================] - 1s 17ms/step - loss: 0.0238 - val_loss: 0.0276
Epoch 28/100
41/41 [==============================] - 1s 17ms/step - loss: 0.0238 - val_loss: 0.0286
Epoch 29/100
41/41 [==============================] - 1s 17ms/step - loss: 0.0234 - val_loss: 0.0287
Epoch 30/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0232 - val_loss: 0.0272
Epoch 31/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0240 - val_loss: 0.0309
Epoch 32/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0233 - val_loss: 0.0285
Epoch 33/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0231 - val_loss: 0.0290
Epoch 34/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0235 - val_loss: 0.0284
Epoch 35/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0232 - val_loss: 0.0286
Epoch 36/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0233 - val_loss: 0.0289
Epoch 37/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0233 - val_loss: 0.0279
Epoch 38/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0234 - val_loss: 0.0276
Epoch 39/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0235 - val_loss: 0.0286
Epoch 40/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0231 - val_loss: 0.0277
Epoch 41/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0231 - val_loss: 0.0274
Epoch 42/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0230 - val_loss: 0.0281
Epoch 43/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0232 - val_loss: 0.0285
Epoch 44/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0234 - val_loss: 0.0303
Epoch 45/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0234 - val_loss: 0.0308
Epoch 46/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0236 - val_loss: 0.0283
Epoch 47/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0234 - val_loss: 0.0278
Epoch 48/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0229 - val_loss: 0.0280
Epoch 49/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0232 - val_loss: 0.0294
Epoch 50/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0230 - val_loss: 0.0328
Epoch 51/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0239 - val_loss: 0.0282
Epoch 52/100
41/41 [==============================] - 1s 18ms/step - loss: 0.0230 - val_loss: 0.0287
Epoch 53/100
41/41 [==============================] - 1s 20ms/step - loss: 0.0234 - val_loss: 0.0274
Epoch 54/100
41/41 [==============================] - 1s 22ms/step - loss: 0.0231 - val_loss: 0.0287
Epoch 55/100
41/41 [==============================] - 1s 22ms/step - loss: 0.0228 - val_loss: 0.0285
Epoch 56/100
41/41 [==============================] - 1s 22ms/step - loss: 0.0230 - val_loss: 0.0286
Epoch 57/100
41/41 [==============================] - 1s 22ms/step - loss: 0.0227 - val_loss: 0.0296
Epoch 58/100
41/41 [==============================] - 1s 21ms/step - loss: 0.0230 - val_loss: 0.0284
Epoch 59/100
41/41 [==============================] - 1s 22ms/step - loss: 0.0229 - val_loss: 0.0284
Epoch 60/100
41/41 [==============================] - 1s 21ms/step - loss: 0.0234 - val_loss: 0.0314
Epoch 61/100
41/41 [==============================] - 1s 21ms/step - loss: 0.0232 - val_loss: 0.0292
Epoch 62/100
41/41 [==============================] - 1s 21ms/step - loss: 0.0229 - val_loss: 0.0294
Epoch 63/100
41/41 [==============================] - 1s 21ms/step - loss: 0.0229 - val_loss: 0.0301
Epoch 64/100
41/41 [==============================] - 1s 22ms/step - loss: 0.0230 - val_loss: 0.0299
Epoch 65/100
41/41 [==============================] - 1s 21ms/step - loss: 0.0229 - val_loss: 0.0299
Epoch 66/100
41/41 [==============================] - 1s 21ms/step - loss: 0.0233 - val_loss: 0.0285
Epoch 67/100
41/41 [==============================] - 1s 21ms/step - loss: 0.0228 - val_loss: 0.0291
Epoch 68/100
41/41 [==============================] - 1s 20ms/step - loss: 0.0230 - val_loss: 0.0302
Epoch 69/100
41/41 [==============================] - 1s 20ms/step - loss: 0.0228 - val_loss: 0.0314
Epoch 70/100
41/41 [==============================] - 1s 20ms/step - loss: 0.0232 - val_loss: 0.0296
Epoch 71/100
41/41 [==============================] - 1s 21ms/step - loss: 0.0229 - val_loss: 0.0313
Epoch 72/100
41/41 [==============================] - 1s 21ms/step - loss: 0.0228 - val_loss: 0.0292
Epoch 73/100
41/41 [==============================] - 1s 20ms/step - loss: 0.0229 - val_loss: 0.0300
Epoch 74/100
41/41 [==============================] - 1s 21ms/step - loss: 0.0230 - val_loss: 0.0293
Epoch 75/100
41/41 [==============================] - 1s 20ms/step - loss: 0.0228 - val_loss: 0.0302
Epoch 76/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0231 - val_loss: 0.0293
Epoch 77/100
41/41 [==============================] - 1s 20ms/step - loss: 0.0230 - val_loss: 0.0316
Epoch 78/100
41/41 [==============================] - 1s 20ms/step - loss: 0.0229 - val_loss: 0.0303
Epoch 79/100
41/41 [==============================] - 1s 20ms/step - loss: 0.0233 - val_loss: 0.0295
Epoch 80/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0229 - val_loss: 0.0297
Epoch 81/100
41/41 [==============================] - 1s 20ms/step - loss: 0.0231 - val_loss: 0.0299
Epoch 82/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0233 - val_loss: 0.0300
Epoch 83/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0229 - val_loss: 0.0317
Epoch 84/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0229 - val_loss: 0.0304
Epoch 85/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0230 - val_loss: 0.0299
Epoch 86/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0230 - val_loss: 0.0293
Epoch 87/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0239 - val_loss: 0.0300
Epoch 88/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0234 - val_loss: 0.0295
Epoch 89/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0227 - val_loss: 0.0298
Epoch 90/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0227 - val_loss: 0.0308
Epoch 91/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0228 - val_loss: 0.0306
Epoch 92/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0227 - val_loss: 0.0302
Epoch 93/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0226 - val_loss: 0.0296
Epoch 94/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0233 - val_loss: 0.0290
Epoch 95/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0228 - val_loss: 0.0310
Epoch 96/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0227 - val_loss: 0.0302
Epoch 97/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0226 - val_loss: 0.0291
Epoch 98/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0228 - val_loss: 0.0327
Epoch 99/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0228 - val_loss: 0.0308
Epoch 100/100
41/41 [==============================] - 1s 19ms/step - loss: 0.0226 - val_loss: 0.0304





<tensorflow.python.keras.callbacks.History at 0x25ed1caa508>

训练集效果

#inverse_transform获得归一化前的原始数据
plt.figure(figsize=(16,10))
plt.plot(scaler_minmax.inverse_transform(d[:,-1]),label='true data')
plt.plot(scaler_minmax.inverse_transform(model.predict(d[:,:-1])),'r:',label='predict')
plt.legend()
pd.DataFrame(scaler_minmax.inverse_transform(model.predict(d[:,:-1])),)

	0
0	2.519001
1	2.000833
2	2.232971
3	2.671180
4	2.993458
...	...
985	2.999729
986	2.851660
987	3.358534
988	2.944040
989	2.537982

990 rows × 1 columns

[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-HYpYOtUn-1649237934550)(output_11_1.png)]

测试集


plt.figure(figsize=(16,10))
plt.plot(scaler_minmax.inverse_transform(d[int(len(d)*split_rate):,-1]),label='true data')
plt.plot(scaler_minmax.inverse_transform(model.predict(d[int(len(d)*split_rate):,:-1])),'r:',label='predict')
plt.legend()

请添加图片描述

测试

plt.rcParams['font.sans-serif'] = ['SimHei']   # 指定默认字体
plt.rcParams['axes.unicode_minus'] = False   # 解决保存图像时负号'-'显示为方块的问题

predict=scaler_minmax.inverse_transform(model.predict(d[int(len(d)*split_rate):,:-1]))
true_data=scaler_minmax.inverse_transform(d[int(len(d)*split_rate):,-1])

plt.figure(figsize=(16,10))
plt.plot(abs(true_data-predict)/true_data,label='相对误差')
plt.plot(abs(true_data-predict),label='绝对误差')
plt.legend()