pytorch/浙江行业电量/行业电量_输出为3_步长为10.py

import numpy as np
import pandas as pd
import torch
from torch import nn
import os
import time
import matplotlib.pyplot as plt
t1 = time.time()

os.environ["KMP_DUPLICATE_LIB_OK"]="TRUE"
DAYS_FOR_TRAIN = 10
torch.manual_seed(42)
class LSTM_Regression(nn.Module):

    def __init__(self, input_size, hidden_size, output_size=1, num_layers=2):
        super().__init__()

        self.lstm = nn.LSTM(input_size, hidden_size, num_layers)
        self.fc = nn.Linear(hidden_size, output_size)

    def forward(self, _x):
        x, _ = self.lstm(_x)  # _x is input, size (seq_len, batch, input_size)
        s, b, h = x.shape  # x is output, size (seq_len, batch, hidden_size)
        x = x.view(s * b, h)
        x = self.fc(x)
        x = x.view(s, b, -1)  # 把形状改回来
        return x


def create_dataset(data, days_for_train=5) -> (np.array, np.array):
    dataset_x, dataset_y = [], []
    for i in range(len(data) - days_for_train-3):
        dataset_x.append(data[i:(i + days_for_train)])
        dataset_y.append(data[i + days_for_train:i + days_for_train+3])
        # print(dataset_x,dataset_y)
    return (np.array(dataset_x), np.array(dataset_y))

def normal(nd):
    high = nd.describe()['75%'] + 1.5*(nd.describe()['75%']-nd.describe()['25%'])
    low = nd.describe()['25%'] - 1.5*(nd.describe()['75%']-nd.describe()['25%'])
    return nd[(nd<high)&(nd>low)]

def data_preprocessing(data):
    data.columns = data.columns.map(lambda x: x.strip())
    data.index = data.index.map(lambda x:str(x).strip()[:10])
    data.index = pd.to_datetime(data.index,format='%Y-%m-%d')
    data.sort_index(inplace=True)
    data = data.loc['2021-01':'2023-08']
    data.drop(columns=[i for i in data.columns if (data[i] == 0).sum() / len(data) >= 0.5], inplace=True)  # 去除0值列

    data = data.astype(float)
    for col in data.columns:
        data[col] = normal(data[col])
    return data

# 拼接数据集
file_dir = './浙江各地市行业电量数据'
excel = os.listdir(file_dir)[0]
data = pd.read_excel(os.path.join(file_dir, excel), sheet_name=0, index_col='stat_date')
data.drop(columns='地市',inplace=True)
data = data_preprocessing(data)

df = data[data.columns[0]]
df.dropna(inplace = True)
dataset_x, dataset_y = create_dataset(df, DAYS_FOR_TRAIN)

for level in data.columns[1:]:
    df = data[level]
    df.dropna(inplace=True)
    x, y = create_dataset(df, DAYS_FOR_TRAIN)
    dataset_x = np.concatenate((dataset_x, x))
    dataset_y = np.concatenate((dataset_y, y))


for excel in os.listdir(file_dir)[1:]:

    data = pd.read_excel(os.path.join(file_dir,excel), sheet_name=0,index_col='stat_date')
    data.drop(columns='地市', inplace=True)
    data = data_preprocessing(data)

    for level in data.columns:
        df = data[level]
        df.dropna(inplace=True)
        x,y = create_dataset(df,DAYS_FOR_TRAIN)
        dataset_x = np.concatenate((dataset_x,x))
        dataset_y = np.concatenate((dataset_y,y))


df_x_10 = pd.DataFrame(dataset_x)
df_y_10 = pd.DataFrame(dataset_y)
df_x_10.to_csv('df_x_10.csv',index=False)
df_y_10.to_csv('df_y_10.csv',index=False)
dataset_x = pd.read_csv('df_x_10.csv').values
dataset_y = pd.read_csv('df_y_10.csv').values
print(dataset_x.shape,dataset_y.shape)
# # 训练
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')

# 标准化到0~1
max_value = np.max(dataset_x)
min_value = np.min(dataset_x)
dataset_x = (dataset_x - min_value) / (max_value - min_value)
dataset_y = (dataset_y - min_value) / (max_value - min_value)
print('max_value:',max_value,'min_value:',min_value)
# 划分训练集和测试集
train_size = int(len(dataset_x)*0.7)
train_x = dataset_x[:train_size]
train_y = dataset_y[:train_size]

# # 将数据改变形状，RNN 读入的数据维度是 (seq_size, batch_size, feature_size)
train_x = train_x.reshape(-1, 1, DAYS_FOR_TRAIN)
train_y = train_y.reshape(-1, 1, 3)

# # 转为pytorch的tensor对象
train_x = torch.from_numpy(train_x).to(device).type(torch.float32)
train_y = torch.from_numpy(train_y).to(device).type(torch.float32)

model = LSTM_Regression(DAYS_FOR_TRAIN, 32, output_size=3, num_layers=2).to(device)  # 导入模型并设置模型的参数输入输出层、隐藏层等

train_loss = []
loss_function = nn.MSELoss()
optimizer = torch.optim.Adam(model.parameters(), lr=0.005, betas=(0.9, 0.999), eps=1e-08, weight_decay=0)
min_loss = 1
for i in range(500):
    train_x,train_y = train_x.to(device),train_y.to(device)
    out = model(train_x)
    loss = loss_function(out, train_y)
    loss.backward()
    optimizer.step()
    optimizer.zero_grad()
    train_loss.append(loss.item())

    if loss <= min_loss:
        min_loss = loss
        best_para = model.state_dict()
    if i % 100 == 0:
        print(f'epoch {i+1}: loss:{loss}')


# 保存/读取模型
torch.save(best_para,'hy3.pth')
model = LSTM_Regression(DAYS_FOR_TRAIN, 32, output_size=3, num_layers=2).to(device)
model.load_state_dict(torch.load('hy3.pth'))
# 测试
model = model.eval()

dataset_x = dataset_x.reshape(-1, 1, DAYS_FOR_TRAIN)  # (seq_size, batch_size, feature_size)
dataset_x = torch.from_numpy(dataset_x).to(device).type(torch.float32)

pred_test = model(dataset_x)  # 全量训练集
pred_test = pred_test.view(-1)
pred_test = np.concatenate((np.zeros(DAYS_FOR_TRAIN), pred_test.cpu().detach().numpy()))

plt.plot(pred_test.reshape(-1), 'r', label='prediction')
plt.plot(dataset_y.reshape(-1), 'b', label='real')
plt.plot((train_size*3, train_size*3), (0, 1), 'g--')
plt.legend(loc='best')
plt.show()

# model.load_state_dict(torch.load('hy3.pth',map_location=torch.device('cpu')))
# max_value = 354024930.8
# min_value = 0.0

# 创建测试集
file_dir = './浙江各地市行业电量数据'
for excel in os.listdir(file_dir):
    df_city = pd.read_excel(os.path.join(file_dir,excel),index_col='stat_date')
    df_city.index = df_city.index.map(lambda x:str(x).strip()[:10])

    df_city.index = pd.to_datetime(df_city.index)
    df_city = df_city.loc['2023-9'][:-3]
    city = df_city['地市'].iloc[0]
    result_dict = {}
    for industry in df_city.columns[2:]:
        df_city[industry] = df_city[industry].astype('float')
        x, y = create_dataset(df_city[industry], 10)
        x = (x - min_value) / (max_value - min_value)
        x = x.reshape(-1, 1, 10)
        x = torch.from_numpy(x).type(torch.float32).to(device)
        pred = model(x).view(-1)
        pred = pred * (max_value - min_value) + min_value
        result = pred.cpu().detach().numpy()[-5:-2]
        result_dict[industry] = list(result)
    df = pd.DataFrame(result_dict,index=['2023-09-28','2023-09-29','2023-09-30'])
    df.to_excel(fr'C:\Users\user\Desktop\9月行业电量预测28-30\{city} .xlsx')
print(time.time()-t1)
print(result_dict)

# 反归一化
pred = pred * (max_value - min_value) + min_value
df = df * (max_value - min_value) + min_value


# 打印指标
print(abs(pred - df[-3:]).mean() / df[-3:].mean())
result_eight = pd.DataFrame({'pred': np.round(pred,1),'real': df[-3:]})
target = (result_eight['pred'].sum() - result_eight['real'].sum()) / df[-31:].sum()
result_eight['loss_rate'] = round(target, 5)
result_eight['level'] = level
list_app.append(result_eight)
print(target)
print(result_eight)
final_df = pd.concat(list_app,ignore_index=True)
final_df.to_csv('市行业电量.csv',encoding='gbk')
print(final_df)