tocmo0nlord/axolotl
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

remove skipped test (#2002)

Browse Source 
* remove skipped test

* use mean_resizing_embeddings with qlora and added tokens

* use </s> as pad_token to prevent resize of embeddings

* make sure local hub test saves to a tmp dir

* use Path so concatenation works

* make sure to use tmp_ds_path for data files
This commit is contained in:
Wing Lian
2024-10-30 12:27:04 -04:00
committed by GitHub
parent 8c3a727f9d
commit 32c60765ef
2 changed files with 37 additions and 36 deletions
Download Patch File Download Diff File Expand all files Collapse all files

69
tests/test_datasets.py
View File

@@ -367,43 +367,44 @@ class TestDatasetPreparation(unittest.TestCase):
def test_load_local_hub_with_revision(self):
"""Verify that a local copy of a hub dataset can be loaded with a specific revision"""
with tempfile.TemporaryDirectory() as tmp_dir:
tmp_ds_path = Path("mhenrichsen/alpaca_2k_test")
tmp_ds_path.mkdir(parents=True, exist_ok=True)
snapshot_download(
repo_id="mhenrichsen/alpaca_2k_test",
repo_type="dataset",
local_dir=tmp_ds_path,
revision="d05c1cb",
)
with tempfile.TemporaryDirectory() as tmp_dir2:
tmp_ds_path = Path(tmp_dir2) / "mhenrichsen/alpaca_2k_test"
tmp_ds_path.mkdir(parents=True, exist_ok=True)
snapshot_download(
repo_id="mhenrichsen/alpaca_2k_test",
repo_type="dataset",
local_dir=tmp_ds_path,
revision="d05c1cb",
)
prepared_path = Path(tmp_dir) / "prepared"
cfg = DictDefault(
{
"tokenizer_config": "huggyllama/llama-7b",
"sequence_len": 1024,
"datasets": [
{
"path": "mhenrichsen/alpaca_2k_test",
"ds_type": "parquet",
"type": "alpaca",
"data_files": [
"mhenrichsen/alpaca_2k_test/alpaca_2000.parquet",
],
"revision": "d05c1cb",
},
],
}
)
prepared_path = Path(tmp_dir) / "prepared"
cfg = DictDefault(
{
"tokenizer_config": "huggyllama/llama-7b",
"sequence_len": 1024,
"datasets": [
{
"path": "mhenrichsen/alpaca_2k_test",
"ds_type": "parquet",
"type": "alpaca",
"data_files": [
f"{tmp_ds_path}/alpaca_2000.parquet",
],
"revision": "d05c1cb",
},
],
}
)
dataset, _ = load_tokenized_prepared_datasets(
self.tokenizer, cfg, prepared_path
)
dataset, _ = load_tokenized_prepared_datasets(
self.tokenizer, cfg, prepared_path
)
assert len(dataset) == 2000
assert "input_ids" in dataset.features
assert "attention_mask" in dataset.features
assert "labels" in dataset.features
shutil.rmtree(tmp_ds_path)
assert len(dataset) == 2000
assert "input_ids" in dataset.features
assert "attention_mask" in dataset.features
assert "labels" in dataset.features
shutil.rmtree(tmp_ds_path)
if __name__ == "__main__":