text-generation-webui/convert-to-safetensors.py

41 lines
1.7 KiB
Python
Raw Normal View History

'''
Converts a transformers model to safetensors format and shards it.
This makes it faster to load (because of safetensors) and lowers its RAM usage
while loading (because of sharding).
Based on the original script by 81300:
https://gist.github.com/81300/fe5b08bff1cba45296a829b9d6b0f303
'''
2023-02-20 19:06:27 +01:00
import argparse
from pathlib import Path
from sys import argv
import torch
from transformers import AutoModelForCausalLM
from transformers import AutoTokenizer
parser = argparse.ArgumentParser(formatter_class=lambda prog: argparse.HelpFormatter(prog,max_help_position=54))
parser.add_argument('MODEL', type=str, default=None, nargs='?', help="Path to the input model.")
parser.add_argument('--output', type=str, default=None, help='Path to the output folder (default: models/{model_name}_safetensors).')
parser.add_argument("--max-shard-size", type=str, default="2GB", help="Maximum size of a shard in GB or MB (default: %(default)s).")
parser.add_argument('--bf16', action='store_true', help='Load the model with bfloat16 precision. Requires NVIDIA Ampere GPU.')
args = parser.parse_args()
if __name__ == '__main__':
path = Path(args.MODEL)
model_name = path.name
print(f"Loading {model_name}...")
model = AutoModelForCausalLM.from_pretrained(path, low_cpu_mem_usage=True, torch_dtype=torch.bfloat16 if args.bf16 else torch.float16)
tokenizer = AutoTokenizer.from_pretrained(path)
out_folder = args.output or Path(f"models/{model_name}_safetensors")
print(f"Saving the converted model to {out_folder} with a maximum shard size of {args.max_shard_size}...")
model.save_pretrained(out_folder, max_shard_size=args.max_shard_size, safe_serialization=True)
tokenizer.save_pretrained(out_folder)