Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

add mistral e2e tests #649

Merged
merged 4 commits into from
Sep 29, 2023
Merged

add mistral e2e tests #649

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
a01ca0a
mistral e2e tests
winglian Sep 28, 2023
57bed92
make sure to enable flash attention for the e2e tests
winglian Sep 28, 2023
601808a
use latest transformers full sha
winglian Sep 28, 2023
af62286
uninstall first
winglian Sep 28, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions .github/workflows/tests.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -69,6 +69,7 @@ jobs:

- name: Install dependencies
run: |
pip3 uninstall -y transformers accelerate
pip3 install -U -e .[flash-attn]
pip3 install -r requirements-tests.txt

Expand Down
2 changes: 1 addition & 1 deletion requirements.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@ torch==2.0.1
auto-gptq
packaging
peft @ git+https://github.com/huggingface/peft.git
transformers @ git+https://github.com/huggingface/transformers.git@78dd120
transformers @ git+https://github.com/huggingface/transformers.git@5e11d72d4d0939138fbabfebe9a69d2061519547
bitsandbytes>=0.41.1
accelerate @ git+https://github.com/huggingface/accelerate@80da9cfb09bb3cc9f1b385cb55d6b90d025a5fd9
deepspeed
Expand Down
208 changes: 208 additions & 0 deletions tests/e2e/test_mistral.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,208 @@
"""
E2E tests for lora llama
"""

import logging
import os
import tempfile
import unittest
from pathlib import Path

from transformers.utils import is_torch_bf16_gpu_available

from axolotl.cli import load_datasets
from axolotl.common.cli import TrainerCliArgs
from axolotl.train import train
from axolotl.utils.config import normalize_config
from axolotl.utils.dict import DictDefault

LOG = logging.getLogger("axolotl.tests.e2e")
os.environ["WANDB_DISABLED"] = "true"


class TestMistral(unittest.TestCase):
"""
Test case for Llama models using LoRA
"""

def test_lora(self):
# pylint: disable=duplicate-code
output_dir = tempfile.mkdtemp()
cfg = DictDefault(
{
"base_model": "openaccess-ai-collective/tiny-mistral",
"base_model_config": "openaccess-ai-collective/tiny-mistral",
"flash_attention": True,
"sequence_len": 1024,
"load_in_8bit": True,
"adapter": "lora",
"lora_r": 32,
"lora_alpha": 64,
"lora_dropout": 0.05,
"lora_target_linear": True,
"val_set_size": 0.1,
"special_tokens": {
"unk_token": "<unk>",
"bos_token": "<s>",
"eos_token": "</s>",
},
"datasets": [
{
"path": "mhenrichsen/alpaca_2k_test",
"type": "alpaca",
},
],
"num_epochs": 2,
"micro_batch_size": 2,
"gradient_accumulation_steps": 1,
"output_dir": output_dir,
"learning_rate": 0.00001,
"optimizer": "adamw_torch",
"lr_scheduler": "cosine",
"max_steps": 20,
"save_steps": 10,
"eval_steps": 10,
}
)
normalize_config(cfg)
cli_args = TrainerCliArgs()
dataset_meta = load_datasets(cfg=cfg, cli_args=cli_args)

train(cfg=cfg, cli_args=cli_args, dataset_meta=dataset_meta)
assert (Path(output_dir) / "adapter_model.bin").exists()

def test_lora_packing(self):
# pylint: disable=duplicate-code
output_dir = tempfile.mkdtemp()
cfg = DictDefault(
{
"base_model": "openaccess-ai-collective/tiny-mistral",
"base_model_config": "openaccess-ai-collective/tiny-mistral",
"flash_attention": True,
"sample_packing": True,
"sequence_len": 1024,
"load_in_8bit": True,
"adapter": "lora",
"lora_r": 32,
"lora_alpha": 64,
"lora_dropout": 0.05,
"lora_target_linear": True,
"val_set_size": 0.1,
"special_tokens": {
"unk_token": "<unk>",
"bos_token": "<s>",
"eos_token": "</s>",
},
"datasets": [
{
"path": "mhenrichsen/alpaca_2k_test",
"type": "alpaca",
},
],
"num_epochs": 2,
"micro_batch_size": 2,
"gradient_accumulation_steps": 1,
"output_dir": output_dir,
"learning_rate": 0.00001,
"optimizer": "adamw_torch",
"lr_scheduler": "cosine",
"max_steps": 20,
"save_steps": 10,
"eval_steps": 10,
}
)
normalize_config(cfg)
cli_args = TrainerCliArgs()
dataset_meta = load_datasets(cfg=cfg, cli_args=cli_args)

train(cfg=cfg, cli_args=cli_args, dataset_meta=dataset_meta)
assert (Path(output_dir) / "adapter_model.bin").exists()

def test_ft(self):
# pylint: disable=duplicate-code
output_dir = tempfile.mkdtemp()
cfg = DictDefault(
{
"base_model": "openaccess-ai-collective/tiny-mistral",
"base_model_config": "openaccess-ai-collective/tiny-mistral",
"flash_attention": True,
"sequence_len": 1024,
"val_set_size": 0.1,
"special_tokens": {
"unk_token": "<unk>",
"bos_token": "<s>",
"eos_token": "</s>",
},
"datasets": [
{
"path": "mhenrichsen/alpaca_2k_test",
"type": "alpaca",
},
],
"num_epochs": 2,
"micro_batch_size": 2,
"gradient_accumulation_steps": 1,
"output_dir": output_dir,
"learning_rate": 0.00001,
"optimizer": "adamw_torch",
"lr_scheduler": "cosine",
"max_steps": 20,
"save_steps": 10,
"eval_steps": 10,
}
)
if is_torch_bf16_gpu_available():
cfg.bf16 = True
else:
cfg.fp16 = True
normalize_config(cfg)
cli_args = TrainerCliArgs()
dataset_meta = load_datasets(cfg=cfg, cli_args=cli_args)

train(cfg=cfg, cli_args=cli_args, dataset_meta=dataset_meta)
assert (Path(output_dir) / "pytorch_model.bin").exists()

def test_ft_packing(self):
# pylint: disable=duplicate-code
output_dir = tempfile.mkdtemp()
cfg = DictDefault(
{
"base_model": "openaccess-ai-collective/tiny-mistral",
"base_model_config": "openaccess-ai-collective/tiny-mistral",
"flash_attention": True,
"sample_packing": True,
"sequence_len": 1024,
"val_set_size": 0.1,
"special_tokens": {
"unk_token": "<unk>",
"bos_token": "<s>",
"eos_token": "</s>",
},
"datasets": [
{
"path": "mhenrichsen/alpaca_2k_test",
"type": "alpaca",
},
],
"num_epochs": 2,
"micro_batch_size": 2,
"gradient_accumulation_steps": 1,
"output_dir": output_dir,
"learning_rate": 0.00001,
"optimizer": "adamw_torch",
"lr_scheduler": "cosine",
"max_steps": 20,
"save_steps": 10,
"eval_steps": 10,
}
)
if is_torch_bf16_gpu_available():
cfg.bf16 = True
else:
cfg.fp16 = True
normalize_config(cfg)
cli_args = TrainerCliArgs()
dataset_meta = load_datasets(cfg=cfg, cli_args=cli_args)

train(cfg=cfg, cli_args=cli_args, dataset_meta=dataset_meta)
assert (Path(output_dir) / "pytorch_model.bin").exists()