import numpy as np import pandas as pd import torch from torch import nn from multiprocessing import Pool import matplotlib.pyplot as plt import os 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-5): dataset_x.append(data[i:(i + days_for_train)]) dataset_y.append(data[i + days_for_train:i + days_for_train+5]) # 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[(ndlow)] def data_preprocessing(data): data.columns = data.columns.map(lambda x: x.strip()) data.index = pd.to_datetime(data.index) 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[data.values != 0] data = data.astype(float) for col in data.columns: data[col] = normal(data[col]) return data # 拼接数据集 file_dir = r'C:\Users\鸽子\Desktop\浙江各地市分电压日电量数据' excel = os.listdir(file_dir)[0] data = pd.read_excel(os.path.join(file_dir, excel), sheet_name=0, index_col=' stat_date ') data = 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 = 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)) print(dataset_x,dataset_y,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) # 划分训练集和测试集 train_size = 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, 5) # 转为pytorch的tensor对象 train_x = torch.from_numpy(train_x).to(device) train_y = torch.from_numpy(train_y).to(device) 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) for i in range(1500): out = model(train_x) loss = loss_function(out, train_y) loss.backward() optimizer.step() optimizer.zero_grad() train_loss.append(loss.item()) # print(loss) # 保存模型 torch.save(model.state_dict(),'dy5.pth') # for test model = model.eval() # 转换成测试模式 # model.load_state_dict(torch.load(os.path.join(model_save_dir,model_file))) # 读取参数 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) pred_test = model(dataset_x) # 全量训练集 # 模型输出 (seq_size, batch_size, output_size) pred_test = pred_test.view(-1) pred_test = np.concatenate((np.zeros(DAYS_FOR_TRAIN), pred_test.cpu().detach().numpy())) plt.plot(pred_test, 'r', label='prediction') plt.plot(df, 'b', label='real') plt.plot((train_size, train_size), (0, 1), 'g--') # 分割线 左边是训练数据 右边是测试数据的输出 plt.legend(loc='best') plt.show() # 创建测试集 # result_list = [] # 以x为基础实际数据,滚动预测未来3天 # x = torch.from_numpy(df[-14:-4]).to(device) # pred = model(x.reshape(-1,1,DAYS_FOR_TRAIN)).view(-1).detach().numpy() # 反归一化 # pred = pred * (max_value - min_value) + min_value # df = df * (max_value - min_value) + min_value # print(pred) # # 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)