|
"""Module for testing prompt tokenizers.""" |
|
import json |
|
import logging |
|
import unittest |
|
from pathlib import Path |
|
|
|
from transformers import AutoTokenizer |
|
|
|
from axolotl.prompt_tokenizers import ShareGPTPromptTokenizingStrategy |
|
from axolotl.prompters import ShareGPTPrompter |
|
|
|
logging.basicConfig(level="INFO") |
|
|
|
|
|
class TestPromptTokenizationStrategies(unittest.TestCase): |
|
""" |
|
Test class for prompt tokenization strategies. |
|
""" |
|
|
|
def setUp(self) -> None: |
|
|
|
self.tokenizer = AutoTokenizer.from_pretrained("huggyllama/llama-7b") |
|
self.tokenizer.add_special_tokens( |
|
{ |
|
"bos_token": "<s>", |
|
"eos_token": "</s>", |
|
"unk_token": "<unk>", |
|
} |
|
) |
|
|
|
def test_sharegpt_integration(self): |
|
print(Path(__file__).parent) |
|
with open( |
|
Path(__file__).parent / "fixtures/conversation.json", encoding="utf-8" |
|
) as fin: |
|
data = fin.read() |
|
conversation = json.loads(data) |
|
with open( |
|
Path(__file__).parent / "fixtures/conversation.tokenized.json", |
|
encoding="utf-8", |
|
) as fin: |
|
data = fin.read() |
|
tokenized_conversation = json.loads(data) |
|
prompter = ShareGPTPrompter("chat") |
|
strat = ShareGPTPromptTokenizingStrategy( |
|
prompter, |
|
self.tokenizer, |
|
False, |
|
2048, |
|
) |
|
example = strat.tokenize_prompt(conversation) |
|
for fields in ["input_ids", "attention_mask", "labels"]: |
|
self.assertEqual(len(example[fields]), len(tokenized_conversation[fields])) |
|
self.assertEqual(example[fields], tokenized_conversation[fields]) |
|
|
|
|
|
if __name__ == "__main__": |
|
unittest.main() |
|
|