mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-14 06:27:45 +08:00
67 lines
2.4 KiB
Python
67 lines
2.4 KiB
Python
# SPDX-FileCopyrightText: Copyright (c) 2022-2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
|
|
# SPDX-License-Identifier: Apache-2.0
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
import unittest
|
|
|
|
import numpy as np
|
|
import torch
|
|
from parameterized import parameterized
|
|
from polygraphy.backend.trt import CreateConfig, EngineFromNetwork, TrtRunner
|
|
|
|
import tensorrt_llm
|
|
from tensorrt_llm import Tensor
|
|
|
|
|
|
class TestFunctional(unittest.TestCase):
|
|
|
|
def setUp(self):
|
|
tensorrt_llm.logger.set_level('error')
|
|
|
|
@parameterized.expand([('float32', ), ('float16', )])
|
|
def test_group_norm(self, dtype):
|
|
# test data
|
|
num_channels = 6
|
|
num_groups = 3
|
|
x_shape = (2, num_channels, 3, 3)
|
|
x_data = torch.rand(x_shape,
|
|
dtype=tensorrt_llm._utils.str_dtype_to_torch(dtype))
|
|
|
|
# construct trt network
|
|
builder = tensorrt_llm.Builder()
|
|
net = builder.create_network()
|
|
with tensorrt_llm.net_guard(net):
|
|
network = tensorrt_llm.default_trtnet()
|
|
x = Tensor(name='x',
|
|
shape=x_shape,
|
|
dtype=tensorrt_llm.str_dtype_to_trt(dtype))
|
|
output = tensorrt_llm.functional.group_norm(x,
|
|
num_groups).trt_tensor
|
|
output.name = 'output'
|
|
network.mark_output(output)
|
|
|
|
# trt run
|
|
build_engine = EngineFromNetwork(
|
|
(builder.trt_builder, net.trt_network),
|
|
config=CreateConfig(fp16=(dtype == 'float16')))
|
|
with TrtRunner(build_engine) as runner:
|
|
outputs = runner.infer(feed_dict={'x': x_data.numpy()})
|
|
|
|
# pytorch run
|
|
ref = torch.nn.functional.group_norm(x_data.cuda(), num_groups)
|
|
|
|
# compare diff
|
|
np.testing.assert_allclose(ref.cpu().numpy(),
|
|
outputs['output'],
|
|
atol=1e-2)
|