DataNalysis/DataBase/db-light/splitdata.py

import pandas as pd
import utils.savedata


def a():
    # 读取大文件
    # large_file = "power_15min.csv"
    # large_df = utils.savedata.readData(large_file)
    #
    # # 读取每个小文件并将其存储在一个字典中
    # small_files = []
    # for i in range(6):
    #     small_files.append("/Dataset_training/NWP/NWP_{}.csv".format(i))
    # small_dfs = {}
    #
    # for file in small_files:
    #     small_dfs[file] = utils.savedata.readData(file)
    #
    # # 根据每个小文件的时间范围拆分大文件
    # i = 0
    # for file, df in small_dfs.items():
    #     min_time = df["C_TIME"].min()
    #     max_time = df["C_TIME"].max()
    #     splitted_df = large_df[(large_df["C_TIME"] >= min_time) & (large_df["C_TIME"] <= max_time)]
    #     utils.savedata.saveData("/Dataset_training/power/power_{}.csv".format(i), splitted_df)
    #     i = i + 1

    filenames = ["Dataset_training/NWP/NWP_0.csv","Dataset_training/power/power_0.csv"]
    dataframes = []
    for name in filenames:
        dataframes.append(utils.savedata.readData(name))

    # 查找最大起始时间和最小结束时间
    max_start_time = max(df['C_TIME'].min() for df in dataframes)
    min_end_time = min(df['C_TIME'].max() for df in dataframes)

    print(max_start_time)
    print(min_end_time)

    # 重新调整每个 DataFrame 的时间范围，只保留在 [max_start_time, min_end_time] 区间内的数据
    for i, df in enumerate(dataframes):
        df['C_TIME'] = pd.to_datetime(df['C_TIME'])  # 确保时间列是 datetime 类型
        df_filtered = df[(df['C_TIME'] >= max_start_time) & (df['C_TIME'] <= min_end_time)]

        # 将结果保存到新文件，文件名为原文件名加上 "_filtered" 后缀
        utils.savedata.saveData(filenames[i],df_filtered)

def split_test():
    # 读取 NWP_5.csv 和 power_5.csv 文件
    nwp_df = utils.savedata.readData("Dataset_training/NWP/NWP_0.csv")
    power_df = utils.savedata.readData("Dataset_training/power/power_0.csv")

    small_files = []
    for i in range(6):
        small_files.append("/Dataset_training/NWP/NWP_{}.csv".format(i))
    for i in range(6):
        small_files.append("/Dataset_training/power/power_{}.csv".format(i))
    dataframes = []
    for name in small_files:
        dataframes.append(utils.savedata.readData(name))
    small_dfs = {}
    l = 0
    for df in dataframes:
        l = l + len(df)
    test_size = int(l* 0.1)

    nwp_test = nwp_df.iloc[-test_size:]
    power_test = power_df.iloc[-test_size:]

    nwp_train = nwp_df[~nwp_df["C_TIME"].isin(nwp_test["C_TIME"])]
    power_train = power_df[~power_df["C_TIME"].isin(power_test["C_TIME"])]

    utils.savedata.saveData("/Dataset_test/NWP/NWP_test.csv", nwp_test)
    utils.savedata.saveData("/Dataset_test/power/power_test.csv", power_test)

    utils.savedata.saveData("/Dataset_training/NWP/NWP_5.csv", nwp_train)
    utils.savedata.saveData("/Dataset_training/power/power_5.csv", power_train)
if __name__ == '__main__':
    a()
    #split_test()