sglang_v0.5.2/sglang/test/srt/test_dp_attention.py

171 lines
4.7 KiB
Python

import unittest
from types import SimpleNamespace
import requests
from sglang.srt.utils import kill_process_tree
from sglang.test.few_shot_gsm8k import run_eval as run_eval_few_shot_gsm8k
from sglang.test.run_eval import run_eval
from sglang.test.test_utils import (
DEFAULT_MLA_MODEL_NAME_FOR_TEST,
DEFAULT_MODEL_NAME_FOR_TEST_MLA,
DEFAULT_MODEL_NAME_FOR_TEST_MLA_NEXTN,
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
DEFAULT_URL_FOR_TEST,
CustomTestCase,
is_in_amd_ci,
popen_launch_server,
)
class TestDPAttentionDP2TP2(CustomTestCase):
@classmethod
def setUpClass(cls):
cls.model = DEFAULT_MLA_MODEL_NAME_FOR_TEST
cls.base_url = DEFAULT_URL_FOR_TEST
cls.process = popen_launch_server(
cls.model,
cls.base_url,
timeout=DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
other_args=[
"--trust-remote-code",
"--tp",
"2",
"--enable-dp-attention",
"--dp",
"2",
"--enable-torch-compile",
"--torch-compile-max-bs",
"2",
],
)
@classmethod
def tearDownClass(cls):
kill_process_tree(cls.process.pid)
def test_mgsm_en(self):
args = SimpleNamespace(
base_url=self.base_url,
model=self.model,
eval_name="mgsm_en",
num_examples=None,
num_threads=1024,
)
metrics = run_eval(args)
print(f"{metrics=}")
self.assertGreater(metrics["score"], 0.8)
class TestDPAttentionDP2TP2DeepseekV3MTP(CustomTestCase):
@classmethod
def setUpClass(cls):
cls.model = DEFAULT_MODEL_NAME_FOR_TEST_MLA
cls.base_url = DEFAULT_URL_FOR_TEST
other_args = [
"--trust-remote-code",
"--disable-radix",
"--speculative-algorithm",
"EAGLE",
"--speculative-num-steps",
"2",
"--speculative-eagle-topk",
"4",
"--speculative-num-draft-tokens",
"4",
"--speculative-draft-model-path",
DEFAULT_MODEL_NAME_FOR_TEST_MLA_NEXTN,
"--tp-size",
"2",
"--enable-dp-attention",
"--dp-size",
"2",
]
if not is_in_amd_ci():
other_args += ["--mem-frac", "0.7"]
cls.process = popen_launch_server(
cls.model,
cls.base_url,
timeout=DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
other_args=other_args,
)
@classmethod
def tearDownClass(cls):
kill_process_tree(cls.process.pid)
def test_gsm8k(self):
requests.get(self.base_url + "/flush_cache")
args = SimpleNamespace(
num_shots=5,
data_path=None,
num_questions=200,
max_new_tokens=512,
parallel=128,
host="http://127.0.0.1",
port=int(self.base_url.split(":")[-1]),
)
metrics = run_eval_few_shot_gsm8k(args)
print(metrics)
self.assertGreater(metrics["accuracy"], 0.60)
server_info = requests.get(self.base_url + "/get_server_info")
avg_spec_accept_length = server_info.json()["internal_states"][0][
"avg_spec_accept_length"
]
print(
f"###test_gsm8k (deepseek-v3 mtp + dp):\n"
f"accuracy={metrics['accuracy']=:.3f}\n"
f"{avg_spec_accept_length=:.3f}\n"
)
self.assertGreater(avg_spec_accept_length, 2.5)
class TestDPAttentionMinimumTokenLoadBalance(CustomTestCase):
@classmethod
def setUpClass(cls):
cls.model = DEFAULT_MLA_MODEL_NAME_FOR_TEST
cls.base_url = DEFAULT_URL_FOR_TEST
cls.process = popen_launch_server(
cls.model,
cls.base_url,
timeout=DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
other_args=[
"--trust-remote-code",
"--tp",
"2",
"--enable-dp-attention",
"--dp",
"2",
"--enable-torch-compile",
"--torch-compile-max-bs",
"2",
"--load-balance-method",
"minimum_tokens",
],
)
@classmethod
def tearDownClass(cls):
kill_process_tree(cls.process.pid)
def test_mgsm_en(self):
args = SimpleNamespace(
base_url=self.base_url,
model=self.model,
eval_name="mgsm_en",
num_examples=None,
num_threads=1024,
)
metrics = run_eval(args)
print(f"{metrics=}")
self.assertGreater(metrics["score"], 0.8)
if __name__ == "__main__":
unittest.main()