423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2026-01-13 01:20:35 +08:00
Code Issues Packages Projects Releases Wiki Activity

[v1] add sft (#9752)

Browse Source
This commit is contained in:
Yaowei Zheng
2026-01-12 03:15:01 +08:00
committed by GitHub
parent 4d3621e3d3
commit 958b9c3468
29 changed files with 439 additions and 305 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
tests_v1/plugins/data_plugins/test_converter.py
View File

@@ -24,8 +24,8 @@ from llamafactory.v1.plugins.data_plugins.converter import DataConverterPlugin
@pytest.mark.parametrize("num_samples", [16])
def test_alpaca_converter(num_samples: int):
data_args = DataArguments(dataset="llamafactory/v1-dataset-info/tiny-supervised-dataset.yaml")
data_engine = DataEngine(data_args)
data_args = DataArguments(train_dataset="llamafactory/v1-dataset-info/tiny-supervised-dataset.yaml")
data_engine = DataEngine(data_args.train_dataset)
original_data = load_dataset("llamafactory/tiny-supervised-dataset", split="train")
indexes = random.choices(range(len(data_engine)), k=num_samples)
for index in indexes:
@@ -73,8 +73,8 @@ def test_sharegpt_converter():
@pytest.mark.parametrize("num_samples", [16])
def test_pair_converter(num_samples: int):
data_args = DataArguments(dataset="llamafactory/v1-dataset-info/orca-dpo-pairs.yaml")
data_engine = DataEngine(data_args)
data_args = DataArguments(train_dataset="llamafactory/v1-dataset-info/orca-dpo-pairs.yaml")
data_engine = DataEngine(data_args.train_dataset)
original_data = load_dataset("HuggingFaceH4/orca_dpo_pairs", split="train_prefs")
indexes = random.choices(range(len(data_engine)), k=num_samples)
for index in indexes: