dataset_generator_program.py

import pandas as pd
import numpy as np

# Define the number of samples
num_samples = 1000

# Generate random data for GPU, RAM, and Processor
gpu = np.random.randint(1, 11, size=num_samples)  # Assuming GPU from 1 to 10
ram = np.random.randint(4, 33, size=num_samples)  # Assuming RAM from 4GB to 32GB
processor = np.random.randint(1, 9, size=num_samples)  # Assuming processor cores from 1 to 8

# Calculate whether it is good for a transformer to run (binary: 0 or 1)
# You would replace this logic with your actual criteria for determining suitability
def is_good_for_transformer(gpu, ram, processor):
    return ((gpu >= 6) & (ram >= 16) & (processor >= 4)).astype(int)

output = is_good_for_transformer(gpu, ram, processor)

# Create a DataFrame
data = pd.DataFrame({'GPU': gpu, 'RAM': ram, 'Processor': processor, 'Good_for_Transformer': output})

# Save the DataFrame to a CSV file
data.to_csv(r'Data_csv\transformer_dataset.csv', index=False)