33ecf4bcd8
Co-authored-by: Claude <noreply@anthropic.com> Co-authored-by: sglang-npu-bot <sglangnpu@163.com> Co-authored-by: Cherry_ming <136634645@qq.com>
79 lines
2.3 KiB
Python
79 lines
2.3 KiB
Python
import logging
|
|
import unittest
|
|
from types import SimpleNamespace
|
|
from urllib.parse import urlparse
|
|
|
|
from sglang.srt.utils import kill_process_tree
|
|
from sglang.test.ascend.test_ascend_utils import QWEN3_4B_GGUF_Q4_K_M_WEIGHTS_PATH
|
|
from sglang.test.ci.ci_register import register_npu_ci
|
|
from sglang.test.few_shot_gsm8k import run_eval as run_eval_few_shot_gsm8k
|
|
from sglang.test.test_utils import (
|
|
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
|
|
DEFAULT_URL_FOR_TEST,
|
|
CustomTestCase,
|
|
popen_launch_server,
|
|
)
|
|
|
|
register_npu_ci(est_time=300, suite="nightly-1-npu-a3", nightly=True)
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
TEST_MODEL_MATRIX = {
|
|
QWEN3_4B_GGUF_Q4_K_M_WEIGHTS_PATH: {
|
|
"accuracy": 0.80,
|
|
},
|
|
}
|
|
|
|
|
|
class TestAscendGGUF(CustomTestCase):
|
|
|
|
@classmethod
|
|
def setUpClass(cls):
|
|
cls.models = TEST_MODEL_MATRIX.keys()
|
|
cls.base_url = DEFAULT_URL_FOR_TEST
|
|
cls.url = urlparse(DEFAULT_URL_FOR_TEST)
|
|
cls.common_args = [
|
|
"--trust-remote-code",
|
|
"--mem-fraction-static",
|
|
0.8,
|
|
"--attention-backend",
|
|
"ascend",
|
|
]
|
|
|
|
def test_a_gsm8k(self):
|
|
for model in self.models:
|
|
with self.subTest(model=model):
|
|
logger.info(f"##=== Testing accuracy: {model} ===##")
|
|
|
|
process = popen_launch_server(
|
|
model,
|
|
self.base_url,
|
|
timeout=DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
|
|
other_args=[
|
|
*self.common_args,
|
|
],
|
|
)
|
|
|
|
try:
|
|
args = SimpleNamespace(
|
|
num_shots=5,
|
|
data_path=None,
|
|
num_questions=1319,
|
|
max_new_tokens=512,
|
|
parallel=128,
|
|
host=f"http://{self.url.hostname}",
|
|
port=int(self.url.port),
|
|
)
|
|
|
|
metrics = run_eval_few_shot_gsm8k(args)
|
|
self.assertGreaterEqual(
|
|
metrics["accuracy"],
|
|
TEST_MODEL_MATRIX[model]["accuracy"],
|
|
)
|
|
finally:
|
|
kill_process_tree(process.pid)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|