import pandas as pd
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.preprocessing import OneHotEncoder
import numpy as np
import os

csv_dir = r"C:\Users\xyz\Desktop\CSVModel\preparing_dataset\Seperated\Airline\csvs"

datasets = []

column_mapping = {
    "Bar Chart_1": ["Nationality", "First Name"],
    "Bar Chart_2": ["Country Name", "Flight Status"],
    "Line Chart_1": ["Name", "Departure data"],
    "Pie Chart_1": ["Flight Status"],
    "Pie Chart_2": ["Continents", "First Name"],
    "Stacked Bar Chart_1": ["Continents", "Gender"]
}

for filename in os.listdir(csv_dir):
    if filename.endswith(".csv"):
        file_path = os.path.join(csv_dir, filename)

        
        chart_type = filename.split("_")[1]  # Assumes the pattern "Airline_ChartType_1.csv"

        
        data = pd.read_csv(file_path)

        
        column_names = data.columns.tolist()
        print(f"Processing file: {filename}, Chart Type: {chart_type}")
        print(f"Column Names: {column_names}")

        
        columns_to_use = column_mapping.get(chart_type, [])  
        features = data[columns_to_use]

        if not features.empty:
            min_rows = features.shape[0]
            dataset = pd.DataFrame({'DatasetID': range(1, min_rows + 1),
                                    'Features': [list(row) for row in features.to_numpy()],
                                    'GraphType': [chart_type] * min_rows})

            
            datasets.append(dataset)

if datasets:
    
    final_dataset = pd.concat(datasets, ignore_index=True)

    
    print(final_dataset)
else:
    print("No valid datasets were created.")

我的“columns_to_use”变量是空的，即使我已经交叉检查了我的列和CSV文件的名称
我想从我不同的csv文件中提取各种功能。每个csv文件都有不同的功能，由于某种原因，它不阅读我的列。当我单独阅读这些文件时，它的工作非常完美。请帮助我理解这个问题
我得到的输出：

Processing file: Airline_Bar Chart_1.csv, Chart Type: Bar Chart
Column Names: ['Nationality', 'First Name']
Processing file: Airline_Bar Chart_2.csv, Chart Type: Bar Chart
Column Names: ['Country Name', 'Flight Status']
Processing file: Airline_Line Chart_1.csv, Chart Type: Line Chart
Column Names: ['First Name', 'Departure Date']
Processing file: Airline_Pie Chart_1.csv, Chart Type: Pie Chart
Column Names: ['Flight Status']
Processing file: Airline_Pie Chart_2.csv, Chart Type: Pie Chart
Column Names: ['Continents', 'First Name']
Processing file: Airline_Stacked Bar Chart_1.csv, Chart Type: Stacked Bar Chart
Column Names: ['Continents', 'Gender']
No valid datasets were created.