cf0c79d52e
* fix: update chat_template * fix: handle gemma3 showing a lot of no content for turn 0 * fix: remove unknown config from examples * fix: test * fix: temporary disable gemma2 test * fix: stop overwriting config.text_config unnecessarily * fix: handling of set cache to the text_config section * feat: add liger gemma support and bump liger to 0.5.5 * fix: add double use_cache setting * fix: add support for final_logit_softcap in CCE for gemma2/3 * fix: set use_cache before model load * feat: add missing layernorm override * fix: handle gemma3 rmsnorm * fix: use wrapper to pass dim as hidden_size * fix: change dim to positional * fix: patch with wrong mlp * chore: refactor use_cache handling * fix import issues * fix tests.e2e.utils import --------- Co-authored-by: Wing Lian <wing@axolotl.ai>
73 lines
2.3 KiB
Python
73 lines
2.3 KiB
Python
"""Module for testing dataset sequence packing"""
|
|
|
|
import unittest
|
|
from pathlib import Path
|
|
|
|
from datasets import Dataset, load_dataset
|
|
from transformers import AutoTokenizer
|
|
|
|
from axolotl.datasets import ConstantLengthDataset, TokenizedPromptDataset
|
|
from axolotl.prompt_tokenizers import AlpacaPromptTokenizingStrategy
|
|
from axolotl.prompters import AlpacaPrompter
|
|
|
|
from tests.hf_offline_utils import enable_hf_offline
|
|
|
|
|
|
class TestPacking(unittest.TestCase):
|
|
"""
|
|
Test class for packing dataset sequences
|
|
"""
|
|
|
|
@enable_hf_offline
|
|
def setUp(self) -> None:
|
|
# pylint: disable=duplicate-code
|
|
self.tokenizer = AutoTokenizer.from_pretrained("huggyllama/llama-7b")
|
|
self.tokenizer.add_special_tokens(
|
|
{
|
|
"bos_token": "<s>",
|
|
"eos_token": "</s>",
|
|
"unk_token": "<unk>",
|
|
}
|
|
)
|
|
|
|
def test_increments_attention(self):
|
|
prompter = AlpacaPrompter("chat")
|
|
strat = AlpacaPromptTokenizingStrategy(
|
|
prompter,
|
|
self.tokenizer,
|
|
False,
|
|
2048,
|
|
)
|
|
dateset = load_dataset(
|
|
"json",
|
|
data_files=str(Path(__file__).parent / "fixtures/alpaca/alpaca.json"),
|
|
)["train"]
|
|
dataset = Dataset.from_list(list(TokenizedPromptDataset(strat, dateset)))
|
|
|
|
constant_len_dataset = ConstantLengthDataset(
|
|
self.tokenizer,
|
|
[dataset],
|
|
seq_length=2048,
|
|
)
|
|
packed_dataset = Dataset.from_list(list(constant_len_dataset))
|
|
example = packed_dataset[0]
|
|
next_bos_index = (
|
|
example["input_ids"][1:].index(self.tokenizer.bos_token_id) + 1
|
|
) # add one since we sliced
|
|
|
|
# first example doesn't have mask reset
|
|
assert example["input_ids"][0] == self.tokenizer.bos_token_id
|
|
assert example["attention_mask"][0] == 1
|
|
assert example["position_ids"][0] == 0
|
|
assert example["position_ids"][1] == 1
|
|
|
|
# but subsequent one does
|
|
assert example["input_ids"][next_bos_index] == self.tokenizer.bos_token_id
|
|
assert example["attention_mask"][next_bos_index] == 2
|
|
assert example["position_ids"][next_bos_index] == 0
|
|
assert example["position_ids"][next_bos_index + 1] == 1
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|