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

CLI cleanup and documentation (#2244)

Browse Source 
* CLI init refactor

* fix

* cleanup and (partial) docs

* Adding documentation and continuing cleanup (in progress)

* remove finetune.py script

* continued cleanup and documentation

* pytest fixes

* review comments

* fix

* Fix

* typing fixes

* make sure the batch dataset patcher for multipack is always loaded when handling datasets

* review comments

* fix

---------

Co-authored-by: Dan Saunders <dan@axolotl.ai>
Co-authored-by: Wing Lian <wing@axolotl.ai>
This commit is contained in:
Dan Saunders
2025-01-13 12:55:29 -05:00
committed by GitHub
parent f89e962119
commit 1ed4de73b6
60 changed files with 1269 additions and 1259 deletions
Download Patch File Download Diff File Expand all files Collapse all files

32
tests/e2e/test_dpo.py
View File

@@ -9,8 +9,8 @@ from pathlib import Path
import pytest
from axolotl.cli import load_rl_datasets
from axolotl.common.cli import TrainerCliArgs
from axolotl.cli.args import TrainerCliArgs
from axolotl.common.datasets import load_preference_datasets
from axolotl.train import train
from axolotl.utils.config import normalize_config
from axolotl.utils.dict import DictDefault
@@ -65,9 +65,9 @@ class TestDPOLlamaLora(unittest.TestCase):
)
normalize_config(cfg)
cli_args = TrainerCliArgs()
dataset_meta = load_rl_datasets(cfg=cfg, cli_args=cli_args)
dataset_meta = load_preference_datasets(cfg=cfg, cli_args=cli_args)
train(cfg=cfg, cli_args=cli_args, dataset_meta=dataset_meta)
train(cfg=cfg, dataset_meta=dataset_meta)
check_model_output_exists(Path(temp_dir) / "checkpoint-20", cfg)
@with_temp_dir
@@ -110,9 +110,9 @@ class TestDPOLlamaLora(unittest.TestCase):
)
normalize_config(cfg)
cli_args = TrainerCliArgs()
dataset_meta = load_rl_datasets(cfg=cfg, cli_args=cli_args)
dataset_meta = load_preference_datasets(cfg=cfg, cli_args=cli_args)
train(cfg=cfg, cli_args=cli_args, dataset_meta=dataset_meta)
train(cfg=cfg, dataset_meta=dataset_meta)
check_model_output_exists(Path(temp_dir) / "checkpoint-20", cfg)
@with_temp_dir
@@ -155,9 +155,9 @@ class TestDPOLlamaLora(unittest.TestCase):
)
normalize_config(cfg)
cli_args = TrainerCliArgs()
dataset_meta = load_rl_datasets(cfg=cfg, cli_args=cli_args)
dataset_meta = load_preference_datasets(cfg=cfg, cli_args=cli_args)
train(cfg=cfg, cli_args=cli_args, dataset_meta=dataset_meta)
train(cfg=cfg, dataset_meta=dataset_meta)
check_model_output_exists(Path(temp_dir) / "checkpoint-20", cfg)
@pytest.mark.skip("kto_pair no longer supported in trl")
@@ -200,9 +200,9 @@ class TestDPOLlamaLora(unittest.TestCase):
)
normalize_config(cfg)
cli_args = TrainerCliArgs()
dataset_meta = load_rl_datasets(cfg=cfg, cli_args=cli_args)
dataset_meta = load_preference_datasets(cfg=cfg, cli_args=cli_args)
train(cfg=cfg, cli_args=cli_args, dataset_meta=dataset_meta)
train(cfg=cfg, dataset_meta=dataset_meta)
check_model_output_exists(Path(temp_dir) / "checkpoint-20", cfg)
@with_temp_dir
@@ -244,9 +244,9 @@ class TestDPOLlamaLora(unittest.TestCase):
)
normalize_config(cfg)
cli_args = TrainerCliArgs()
dataset_meta = load_rl_datasets(cfg=cfg, cli_args=cli_args)
dataset_meta = load_preference_datasets(cfg=cfg, cli_args=cli_args)
train(cfg=cfg, cli_args=cli_args, dataset_meta=dataset_meta)
train(cfg=cfg, dataset_meta=dataset_meta)
check_model_output_exists(Path(temp_dir) / "checkpoint-20", cfg)
@with_temp_dir
@@ -291,9 +291,9 @@ class TestDPOLlamaLora(unittest.TestCase):
)
normalize_config(cfg)
cli_args = TrainerCliArgs()
dataset_meta = load_rl_datasets(cfg=cfg, cli_args=cli_args)
dataset_meta = load_preference_datasets(cfg=cfg, cli_args=cli_args)
train(cfg=cfg, cli_args=cli_args, dataset_meta=dataset_meta)
train(cfg=cfg, dataset_meta=dataset_meta)
check_model_output_exists(Path(temp_dir) / "checkpoint-20", cfg)
@pytest.mark.skip(reason="Fix the implementation")
@@ -355,7 +355,7 @@ class TestDPOLlamaLora(unittest.TestCase):
)
normalize_config(cfg)
cli_args = TrainerCliArgs()
dataset_meta = load_rl_datasets(cfg=cfg, cli_args=cli_args)
dataset_meta = load_preference_datasets(cfg=cfg, cli_args=cli_args)
train(cfg=cfg, cli_args=cli_args, dataset_meta=dataset_meta)
train(cfg=cfg, dataset_meta=dataset_meta)
check_model_output_exists(Path(temp_dir) / "checkpoint-20", cfg)