mirror of
https://github.com/huggingface/text-generation-inference.git
synced 2025-04-20 14:22:08 +00:00
e4d31a40db
This PR bumps the client tests from `google/flan-t5-xxl` to `meta-llama/Llama-2-7b-chat-hf` to resolve issues when calling the endpoint and `google/flan-t5-xxl` is not available run with ```bash make python-client-tests clients/python/tests/test_client.py .............. [ 43%] clients/python/tests/test_errors.py .......... [ 75%] clients/python/tests/test_inference_api.py ...... [ 93%] clients/python/tests/test_types.py .. [100%] ``` **note `google/flan-t5-xxl` function is currently unused but still included in the `conftest.py`
62 lines
1.1 KiB
Python
62 lines
1.1 KiB
Python
import pytest
|
|
|
|
from text_generation import __version__
|
|
from huggingface_hub.utils import build_hf_headers
|
|
|
|
|
|
@pytest.fixture
|
|
def flan_t5_xxl():
|
|
return "google/flan-t5-xxl"
|
|
|
|
|
|
@pytest.fixture
|
|
def llama_7b():
|
|
return "meta-llama/Llama-2-7b-chat-hf"
|
|
|
|
|
|
@pytest.fixture
|
|
def fake_model():
|
|
return "fake/model"
|
|
|
|
|
|
@pytest.fixture
|
|
def unsupported_model():
|
|
return "gpt2"
|
|
|
|
|
|
@pytest.fixture
|
|
def base_url():
|
|
return "https://api-inference.huggingface.co/models"
|
|
|
|
|
|
@pytest.fixture
|
|
def bloom_url(base_url, bloom_model):
|
|
return f"{base_url}/{bloom_model}"
|
|
|
|
|
|
@pytest.fixture
|
|
def flan_t5_xxl_url(base_url, flan_t5_xxl):
|
|
return f"{base_url}/{flan_t5_xxl}"
|
|
|
|
|
|
@pytest.fixture
|
|
def llama_7b_url(base_url, llama_7b):
|
|
return f"{base_url}/{llama_7b}"
|
|
|
|
|
|
@pytest.fixture
|
|
def fake_url(base_url, fake_model):
|
|
return f"{base_url}/{fake_model}"
|
|
|
|
|
|
@pytest.fixture
|
|
def unsupported_url(base_url, unsupported_model):
|
|
return f"{base_url}/{unsupported_model}"
|
|
|
|
|
|
@pytest.fixture(scope="session")
|
|
def hf_headers():
|
|
return build_hf_headers(
|
|
library_name="text-generation-tests", library_version=__version__
|
|
)
|