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

14
tests/e2e/test_mixtral.py
View File

@@ -9,8 +9,8 @@ import unittest
import torch
from transformers.utils import is_torch_bf16_gpu_available
from axolotl.cli import load_datasets
from axolotl.common.cli import TrainerCliArgs
from axolotl.cli.args import TrainerCliArgs
from axolotl.common.datasets import load_datasets
from axolotl.train import train
from axolotl.utils.config import normalize_config
from axolotl.utils.dict import DictDefault
@@ -73,7 +73,7 @@ class TestMixtral(unittest.TestCase):
cli_args = TrainerCliArgs()
dataset_meta = load_datasets(cfg=cfg, cli_args=cli_args)
model, _ = train(cfg=cfg, cli_args=cli_args, dataset_meta=dataset_meta)
model, _ = train(cfg=cfg, dataset_meta=dataset_meta)
assert (
model.base_model.model.model.layers[0].block_sparse_moe.gate.weight.dtype
== torch.float32
@@ -127,7 +127,7 @@ class TestMixtral(unittest.TestCase):
cli_args = TrainerCliArgs()
dataset_meta = load_datasets(cfg=cfg, cli_args=cli_args)
model, _ = train(cfg=cfg, cli_args=cli_args, dataset_meta=dataset_meta)
model, _ = train(cfg=cfg, dataset_meta=dataset_meta)
assert (
model.base_model.model.model.layers[0].block_sparse_moe.gate.weight.dtype
== torch.float32
@@ -184,7 +184,7 @@ class TestMixtral(unittest.TestCase):
cli_args = TrainerCliArgs()
dataset_meta = load_datasets(cfg=cfg, cli_args=cli_args)
model, _ = train(cfg=cfg, cli_args=cli_args, dataset_meta=dataset_meta)
model, _ = train(cfg=cfg, dataset_meta=dataset_meta)
assert (
model.base_model.model.model.layers[0].block_sparse_moe.gate.weight.dtype
== torch.float32
@@ -241,7 +241,7 @@ class TestMixtral(unittest.TestCase):
cli_args = TrainerCliArgs()
dataset_meta = load_datasets(cfg=cfg, cli_args=cli_args)
model, _ = train(cfg=cfg, cli_args=cli_args, dataset_meta=dataset_meta)
model, _ = train(cfg=cfg, dataset_meta=dataset_meta)
assert (
model.base_model.model.model.layers[0].block_sparse_moe.gate.weight.dtype
== torch.float32
@@ -285,5 +285,5 @@ class TestMixtral(unittest.TestCase):
cli_args = TrainerCliArgs()
dataset_meta = load_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(temp_dir, cfg)