Last active
March 17, 2023 16:50
-
-
Save carterprince/aa43e4e15b758a0bf97a924ddae76d66 to your computer and use it in GitHub Desktop.
python script that takes a given file and splits it into N new files with N GPT-3 tokens each
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import sys | |
from transformers import GPT2Tokenizer | |
def split_file(file_path, output_prefix, tokens_per_file): | |
tokenizer = GPT2Tokenizer.from_pretrained("gpt2") | |
with open(file_path, 'r') as file: | |
text = file.read() | |
tokens = tokenizer.encode(text) | |
file_count = 1 | |
idx = 0 | |
while idx < len(tokens): | |
output_file_name = f"{output_prefix}_{file_count}.txt" | |
current_tokens = [] | |
while len(current_tokens) < tokens_per_file and idx < len(tokens): | |
current_tokens.append(tokens[idx]) | |
idx += 1 | |
with open(output_file_name, 'w') as output_file: | |
output_file.write(tokenizer.decode(current_tokens)) | |
file_count += 1 | |
print(f"Successfully split the file into {file_count - 1} files.") | |
if __name__ == "__main__": | |
if len(sys.argv) != 4: | |
print("Usage: python split_file.py <input_file> <output_prefix> <tokens_per_file>") | |
sys.exit(1) | |
input_file = sys.argv[1] | |
output_prefix = sys.argv[2] | |
tokens_per_file = int(sys.argv[3]) | |
if tokens_per_file < 1: | |
print("Error: tokens_per_file must be a positive integer.") | |
sys.exit(1) | |
split_file(input_file, output_prefix, tokens_per_file) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment