|
|
|
|
@ -2,10 +2,12 @@ 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
|
|
|
|
|
|
|
|
|
|
DAYS_FOR_TRAIN = 9
|
|
|
|
|
|
|
|
|
|
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):
|
|
|
|
|
@ -31,22 +33,30 @@ def create_dataset(data, days_for_train=5) -> (np.array, np.array):
|
|
|
|
|
dataset_y.append(data[i + days_for_train])
|
|
|
|
|
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)]
|
|
|
|
|
|
|
|
|
|
if __name__ == '__main__':
|
|
|
|
|
data = pd.read_excel(r'C:\Users\鸽子\Desktop\浙江21年-23年行业以及分电压等级日电量.xlsx', sheet_name=1)
|
|
|
|
|
def run(file_dir,excel):
|
|
|
|
|
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
|
|
|
|
|
|
|
|
|
|
data = pd.read_excel(os.path.join(file_dir,excel), sheet_name=0,index_col=' stat_date ')
|
|
|
|
|
|
|
|
|
|
data.columns = data.columns.map(lambda x: x.strip())
|
|
|
|
|
data.sort_values(by='pt_date', ascending=True)
|
|
|
|
|
data['pt_date'] = pd.to_datetime(data['pt_date'], format='%Y%m%d', errors='coerce').astype('string')
|
|
|
|
|
data.set_index('pt_date', inplace=True)
|
|
|
|
|
data = data[(data.index.str[:6] != '202309') & (data.index.str[:6] != '202310')]
|
|
|
|
|
|
|
|
|
|
data.index = pd.to_datetime(data.index)
|
|
|
|
|
print(data.head())
|
|
|
|
|
data = data.loc['2021-01':'2023-09']
|
|
|
|
|
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
|
|
|
|
|
print('len(data):', len(data))
|
|
|
|
|
for level in data.columns:
|
|
|
|
|
df = data[level]
|
|
|
|
|
df = df[df.values != 0] # 去除0值行
|
|
|
|
|
df = normal(df)
|
|
|
|
|
df = df.astype('float32').values # 转换数据类型
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
# 标准化到0~1
|
|
|
|
|
max_value = np.max(df)
|
|
|
|
|
min_value = np.min(df)
|
|
|
|
|
@ -56,7 +66,7 @@ if __name__ == '__main__':
|
|
|
|
|
print('len(dataset_x:)', len(dataset_x))
|
|
|
|
|
|
|
|
|
|
# 划分训练集和测试集
|
|
|
|
|
train_size = len(dataset_x) - 31
|
|
|
|
|
train_size = len(dataset_x) - 3
|
|
|
|
|
train_x = dataset_x[:train_size]
|
|
|
|
|
train_y = dataset_y[:train_size]
|
|
|
|
|
|
|
|
|
|
@ -65,23 +75,23 @@ if __name__ == '__main__':
|
|
|
|
|
train_y = train_y.reshape(-1, 1, 1)
|
|
|
|
|
|
|
|
|
|
# 转为pytorch的tensor对象
|
|
|
|
|
train_x = torch.from_numpy(train_x)
|
|
|
|
|
train_y = torch.from_numpy(train_y)
|
|
|
|
|
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=1, num_layers=2) # 导入模型并设置模型的参数输入输出层、隐藏层等
|
|
|
|
|
model = LSTM_Regression(DAYS_FOR_TRAIN, 32, output_size=1, 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=0.001, betas=(0.9, 0.999), eps=1e-08, weight_decay=0)
|
|
|
|
|
for i in range(1200):
|
|
|
|
|
optimizer = torch.optim.Adam(model.parameters(), lr=0.005, betas=(0.9, 0.999), eps=1e-08, weight_decay=0)
|
|
|
|
|
for i in range(2000):
|
|
|
|
|
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(),save_filename)
|
|
|
|
|
# torch.save(model.state_dict(),os.path.join(model_save_dir,model_file))
|
|
|
|
|
@ -89,29 +99,55 @@ if __name__ == '__main__':
|
|
|
|
|
# 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[-13:-4])
|
|
|
|
|
x = torch.from_numpy(df[-14:-4]).to(device)
|
|
|
|
|
for i in range(3):
|
|
|
|
|
next_1_8 = x[1:]
|
|
|
|
|
next_9 = model(x.reshape(-1,1,DAYS_FOR_TRAIN))
|
|
|
|
|
# print(next_9,next_1_8)
|
|
|
|
|
x = torch.concatenate((next_1_8, next_9.view(-1)))
|
|
|
|
|
result_list.append(next_9.view(-1).item())
|
|
|
|
|
# print(result_list,len(result_list))
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
# 反归一化
|
|
|
|
|
pred = np.array(result_list) * (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_test': pred, 'real': df[-3:]})
|
|
|
|
|
target = (result_eight['pred_test'].sum() - result_eight['real'].sum()) / df[-31:].sum()
|
|
|
|
|
# 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'] = target
|
|
|
|
|
print(target)
|
|
|
|
|
print(result_eight)
|
|
|
|
|
# with open(fr'C:\Users\鸽子\Desktop\各地市8月行业电量预测\{excel[:2]}.txt', 'a', encoding='utf-8') as f:
|
|
|
|
|
# f.write(f'{excel[:2]}市{industry}行业月底偏差率:{round(target, 5)}\n')
|
|
|
|
|
|
|
|
|
|
# result_eight.to_csv(f'./电压等级电量预测结果/9月{excel}.txt', sep='\t', mode='a')
|
|
|
|
|
# with open(fr'./偏差率/9月偏差率.txt', 'a', encoding='utf-8') as f:
|
|
|
|
|
# f.write(f'{excel}电压等级{level}月底偏差率:{round(target, 5)}\n')
|
|
|
|
|
|
|
|
|
|
if __name__ == '__main__':
|
|
|
|
|
file_dir = r'C:\Users\user\Desktop\浙江各地市分电压日电量数据'
|
|
|
|
|
p = Pool(4)
|
|
|
|
|
for excel in os.listdir(file_dir):
|
|
|
|
|
p.apply_async(func=run,args=(file_dir,excel))
|
|
|
|
|
p.close()
|
|
|
|
|
p.join()
|
