biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Misc] Replace os environ to monkeypatch in test suite (#14516)

Browse Source 
Signed-off-by: sibi <85477603+t-sibiraj@users.noreply.github.com>
Signed-off-by: Aaron Pham <contact@aarnphm.xyz>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
Co-authored-by: Aaron Pham <contact@aarnphm.xyz>
This commit is contained in:
Sibi
2025-03-17 11:35:57 +08:00
committed by GitHub
parent 1e799b7ec1
commit a73e183e36
43 changed files with 1900 additions and 1658 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tests/entrypoints/llm/test_accuracy.py
View File

@@ -49,7 +49,7 @@ TPU_TP_TEST_STR = "" #"tensor_parallel_size=4"
@pytest.mark.skipif(not current_platform.is_cuda()
and not current_platform.is_tpu(),
reason="V1 is currently only supported on CUDA and TPU")
def test_lm_eval_accuracy_v1_engine(monkeypatch):
def test_lm_eval_accuracy_v1_engine(monkeypatch: pytest.MonkeyPatch):
"""Run with the V1 Engine."""
with monkeypatch.context() as m:
@@ -67,7 +67,7 @@ def test_lm_eval_accuracy_v1_engine(monkeypatch):
run_test(more_args)
def test_lm_eval_accuracy_v0_engine(monkeypatch):
def test_lm_eval_accuracy_v0_engine(monkeypatch: pytest.MonkeyPatch):
"""Run with the V0 Engine."""
with monkeypatch.context() as m:

49
tests/entrypoints/offline_mode/test_offline_mode.py
View File

@@ -53,32 +53,37 @@ def cache_models():
@pytest.mark.skip_global_cleanup
@pytest.mark.usefixtures("cache_models")
def test_offline_mode(monkeypatch):
def test_offline_mode(monkeypatch: pytest.MonkeyPatch):
# Set HF to offline mode and ensure we can still construct an LLM
try:
monkeypatch.setenv("HF_HUB_OFFLINE", "1")
monkeypatch.setenv("VLLM_NO_USAGE_STATS", "1")
with monkeypatch.context() as m:
try:
m.setenv("HF_HUB_OFFLINE", "1")
m.setenv("VLLM_NO_USAGE_STATS", "1")
def disable_connect(*args, **kwargs):
raise RuntimeError("No http calls allowed")
def disable_connect(*args, **kwargs):
raise RuntimeError("No http calls allowed")
monkeypatch.setattr(urllib3.connection.HTTPConnection, "connect",
disable_connect)
monkeypatch.setattr(urllib3.connection.HTTPSConnection, "connect",
disable_connect)
m.setattr(
urllib3.connection.HTTPConnection,
"connect",
disable_connect,
)
m.setattr(
urllib3.connection.HTTPSConnection,
"connect",
disable_connect,
)
# Need to re-import huggingface_hub and friends to setup offline mode
_re_import_modules()
# Cached model files should be used in offline mode
for model_config in MODEL_CONFIGS:
LLM(**model_config)
finally:
# Reset the environment after the test
# NB: Assuming tests are run in online mode
monkeypatch.delenv("HF_HUB_OFFLINE")
monkeypatch.delenv("VLLM_NO_USAGE_STATS")
_re_import_modules()
pass
# Need to re-import huggingface_hub
# and friends to setup offline mode
_re_import_modules()
# Cached model files should be used in offline mode
for model_config in MODEL_CONFIGS:
LLM(**model_config)
finally:
# Reset the environment after the test
# NB: Assuming tests are run in online mode
_re_import_modules()
def _re_import_modules():

5
tests/entrypoints/openai/correctness/test_lmeval.py
View File

@@ -70,7 +70,7 @@ def run_test(more_args):
@pytest.mark.skipif(not current_platform.is_cuda()
and not current_platform.is_tpu(),
reason="V1 currently only supported on CUDA and TPU")
def test_lm_eval_accuracy_v1_engine(monkeypatch):
def test_lm_eval_accuracy_v1_engine(monkeypatch: pytest.MonkeyPatch):
"""Run with the V1 Engine."""
with monkeypatch.context() as m:
@@ -85,7 +85,8 @@ def test_lm_eval_accuracy_v1_engine(monkeypatch):
@pytest.mark.parametrize("more_args", MORE_ARGS_LIST)
def test_lm_eval_accuracy_v0_engine(monkeypatch, more_args):
def test_lm_eval_accuracy_v0_engine(monkeypatch: pytest.MonkeyPatch,
more_args):
"""Run with the V0 Engine."""
with monkeypatch.context() as m: