dataset-codes

import pandas as pd

file_name = input("Enter the name (or path) of the Excel file: ")

# Read the Excel file
try:
    df = pd.read_excel(file_name)
except FileNotFoundError:
    print("The specified file could not be found.")
    exit()

filter = df['sınıf'].isin(['deprem', 'yangın', 'sel', 'alakasız'])
filtered_df = df[filter]

used_tweets = []

# Ask the user how many data sets they want to create
try:
    num_datasets = int(input("How many data sets do you want to create? Enter a number: "))
except ValueError:
    print("Invalid input. Please enter a valid number.")
    exit()

# Create the specified number of data sets
for i in range(num_datasets):
    dataset_tweets = []
    for class_name in ['deprem', 'yangın', 'sel', 'alakasız']:
        # Filter rows based on the class
        class_filter = filtered_df['sınıf'] == class_name
        class_df = filtered_df[class_filter]

        # If there are rows for the class and the tweet message has not been used before, add it to the dataset_tweets list
        if not class_df.empty:
            tweet_message = class_df.sample()['tweet'].values[0]
            if tweet_message not in used_tweets:
                dataset_tweets.append((class_name.capitalize(), tweet_message))
                used_tweets.append(tweet_message)

    # Write the dataset to a text file
    file_name = f"dataset_{i+1}.txt"
    with open(file_name, 'w') as file:
        for class_name, tweet_message in dataset_tweets:
            file.write(f"{class_name} tweet message: {tweet_message}\n")

    # Print a message to indicate that the dataset has been created
    print(f"Data Set {i + 1} has been created and saved as {file_name}")

# Print a message to indicate that all data sets have been created
print("All data sets have been created.")