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

✨feat:add task name filter #20

Merged
merged 2 commits into from
Jun 27, 2023
Merged

✨feat:add task name filter #20

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6efadfa
✨feat:add task name filter
Jun 26, 2023
fb28a53
🐞fix:make blue happy
Jun 26, 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
19 changes: 5 additions & 14 deletions mteb-zh/readme.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -59,38 +59,29 @@ pip install -r requirements.txt
```
2. 运行评测脚本
```bash
# model_type: m3e_base, erlangshen, uer, d_meta_soul, openai, text2vec ...
python run_mteb_zh <model_type> <model_name: Optional>
python run_mteb_zh.py --model-type <model_type> --model-id <model_id: Optional>
```
3. 查看帮助
```bash
python run_mteb_zh --help
python run_mteb_zh.py --help
```

### 示例

评测 M3E-base 模型
```bash
python run_mteb_zh sentence-transformer moka-ai/m3e-base
python run_mteb_zh --model-type sentence_transformer --model-id moka-ai/m3e-base
```

评测 UER 模型
```bash
python run_mteb_zh uer uer/sbert-base-chinese-nli
python run_mteb_zh --model-type sentence_transformer --model-id uer/sbert-base-chinese-nli
```

评测 ErLangShen 模型
```bash
python run_mteb_zh erlangshen
python run_mteb_zh --model-type erlangshen
```
case ModelType.m3e_small:
return SentenceTransformer('moka-ai/m3e-small')
case ModelType.m3e_base:
return SentenceTransformer('moka-ai/m3e-base')
case ModelType.d_meta_soul:
return SentenceTransformer('DMetaSoul/sbert-chinese-general-v2')
case ModelType.uer:
return SentenceTransformer('uer/sbert-base-chinese-nli')

## 评测你的模型

Expand Down
22 changes: 16 additions & 6 deletions mteb-zh/run_mteb_zh.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,24 +30,34 @@
]


def filter_by_name(name: str):
return [task for task in default_tasks if task.description['name'] == name] # type: ignore


def filter_by_type(task_type: TaskType):
if task_type is TaskType.All:
return default_tasks
else:
return [task for task in default_tasks if task.description['type'] == task_type.value] # type: ignore


def main(
model_type: Annotated[ModelType, typer.Option()],
model_id: str | None = None,
model_name: str | None = None,
task_type: TaskType = TaskType.Classification,
task_name: str | None = None,
output_folder: Path = Path('results'),
):
output_folder = Path(output_folder)
model = load_model(model_type, model_id)

if task_type is TaskType.All:
tasks = default_tasks
if task_name:
tasks = filter_by_name(task_name)
else:
tasks = [task for task in default_tasks if task.description['type'] == task_type.value] # type: ignore
tasks = filter_by_type(task_type)

evaluation = MTEB(tasks=tasks)
if model_name is None:
model_name = model_type.value + (f'-{model_id.replace("/", "-")}' if model_id else '')
model_name = model_type.value + (f'-{model_id.replace("/", "-")}' if model_id else '')
evaluation.run(model, output_folder=str(output_folder / model_name))


Expand Down