21 lines
631 B
Python
21 lines
631 B
Python
from aitextgen.TokenDataset import TokenDataset
|
|
from aitextgen.tokenizers import train_tokenizer
|
|
from aitextgen.utils import GPT2ConfigCPU
|
|
from aitextgen import aitextgen
|
|
import json
|
|
|
|
with open('config.json', 'r') as file:
|
|
json_object = json.load(file)
|
|
|
|
file_name = json_object['file']
|
|
|
|
train_tokenizer(file_name)
|
|
tokenizer_file = "aitextgen.tokenizer.json"
|
|
config = GPT2ConfigCPU()
|
|
|
|
ai = aitextgen(tokenizer_file=tokenizer_file, config=config)
|
|
data = TokenDataset(
|
|
file_name, tokenizer_file=tokenizer_file, block_size=64)
|
|
ai.train(data, batch_size=16, num_steps=1000000,
|
|
generate_every=10000, save_every=5000)
|