mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-14 06:27:45 +08:00
4bb65f216f
* Update TensorRT-LLM --------- Co-authored-by: meghagarwal <16129366+megha95@users.noreply.github.com> Co-authored-by: Shixiaowei02 <39303645+Shixiaowei02@users.noreply.github.com>
93 lines
3.4 KiB
Python
93 lines
3.4 KiB
Python
# SPDX-FileCopyrightText: Copyright (c) 2024 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 os
|
|
import sys
|
|
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
|
|
|
|
sys.path.append(os.path.join(os.path.dirname(__file__), '..'))
|
|
from utils.util import unittest_name_func
|
|
|
|
|
|
class TestFunctional(unittest.TestCase):
|
|
|
|
def setUp(self):
|
|
tensorrt_llm.logger.set_level('error')
|
|
|
|
@parameterized.expand([
|
|
('int32', (1, )),
|
|
('int32', (256, )),
|
|
('int32', (256, )),
|
|
('float32', (3, 16)),
|
|
('float32', (3, 16)),
|
|
('float32', (3, 16)),
|
|
('float16', (5, 6, 8)),
|
|
('float16', (5, 6, 8)),
|
|
('float16', (5, 6, 8)),
|
|
],
|
|
name_func=unittest_name_func)
|
|
def test_masked_select(self, dtype, input_shape):
|
|
dtype = 'float32'
|
|
mask_shape = input_shape
|
|
|
|
input_data = torch.rand(
|
|
input_shape, dtype=tensorrt_llm._utils.str_dtype_to_torch(dtype))
|
|
mask_data = torch.randint(2, mask_shape).to(torch.bool)
|
|
|
|
source_shape = (mask_data.nonzero().shape[0])
|
|
source_data = torch.rand(
|
|
source_shape, dtype=tensorrt_llm._utils.str_dtype_to_torch(dtype))
|
|
|
|
builder = tensorrt_llm.Builder()
|
|
net = builder.create_network()
|
|
with tensorrt_llm.net_guard(net):
|
|
network = tensorrt_llm.default_trtnet()
|
|
x = Tensor(name='input',
|
|
shape=input_data.shape,
|
|
dtype=tensorrt_llm.str_dtype_to_trt(dtype))
|
|
y = Tensor(name='mask',
|
|
shape=mask_data.shape,
|
|
dtype=tensorrt_llm.str_dtype_to_trt('bool'))
|
|
source = Tensor(name='source',
|
|
shape=source_data.shape,
|
|
dtype=tensorrt_llm.str_dtype_to_trt(dtype))
|
|
output = tensorrt_llm.functional.masked_scatter(x, y,
|
|
source).trt_tensor
|
|
output.name = 'output'
|
|
network.mark_output(output)
|
|
|
|
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={
|
|
'input': input_data.numpy(),
|
|
'mask': mask_data.numpy(),
|
|
'source': source_data.numpy()
|
|
})
|
|
input_data = input_data.cuda()
|
|
ref = input_data.masked_scatter_(mask_data.cuda(), source_data.cuda())
|
|
np.testing.assert_allclose(ref.cpu().numpy(),
|
|
outputs['output'],
|
|
atol=1e-5)
|