Coding-基于exl序号寻找文件夹中对应序号的txt

• 基于exl序号寻找文件夹中对应序号的txt

基于exl序号寻找文件夹中对应序号的txt

import pandas as pd
import numpy as np
import os
import glob
import fnmatch
# 读取CSV文件
csv_file_path = 'now_have.csv'  # 替换为你的CSV文件路径,前两列为序号对,后一列为真值，都有表头

# 读取 CSV 文件到 DataFrame
# data = pd.read_csv(csv_file_path, delimiter=' ', names=['c', 'l', 't'], header=None)
data_pd = pd.read_csv(csv_file_path, header=0) 

# 打印第一列的内容
print("First column (c):")

# 定义文件夹路径
folder_path = 'dataset\\toushe\\toushe_kongjianguangyuan_aiwantisi_hs2048'  # 空间光源+好光谱

save_name = 'gao_1121_good_light_good_spectrum_green_data.npy'

# 使用glob模块获取所有.txt文件
txt_files = glob.glob(os.path.join(folder_path, '*.txt'))

# 过滤掉文件名中包含 "gualazhi" 字符串的文件
filtered_txt_files = [file for file in txt_files if not fnmatch.fnmatch(os.path.basename(file), '*gualazhi*')]
data_y = []
# 打印所有.txt文件的路径
for file_path_ in txt_files:

    file_path = os.path.basename(file_path_)
    # 使用分隔符分割文件名
    name_parts = file_path.split('-')
    
    # 提取 "10-1" 部分
    extracted_part = (int(name_parts[0]), int(name_parts[1]))

    # 查找匹配的行
    matching_rows = data_pd[(data_pd['c'] == extracted_part[0]) & (data_pd['l'] == extracted_part[1])]
    
    # 如果找到匹配的行，打印 "OK"
    if not matching_rows.empty:
        with open(file_path_, 'r') as file:
            lines = file.readlines()

            # 提取数据部分
            data_lines = lines[8:]  # 跳过前8行元数据和表头
            data = []

            for line in data_lines:
                if line.strip():  # 忽略空行
                    parts = line.strip().split(';')

                    wave = float(parts[0])
                    sample = float(parts[1])
                    dark = float(parts[2])
                    reference = float(parts[3])
                    corrected = float(parts[4])

                    feature=corrected/(reference-dark)
                    true_label = float(matching_rows['t'])
                    data.append([feature,true_label])

            data_np = np.array(data)
        data_y.append(data_np)
        print("OK")
    else:
        # print("No match found")
        pass


# 数据处理完
data_y_np = np.array(data_y)
transposed_data = np.transpose(data_y_np, (0, 2, 1))
np.save(save_name, transposed_data)