TensorRT-LLMs/tests/functional/test_outer.py
2023-09-20 00:29:41 -07:00

64 lines
2.1 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 polygraphy.backend.trt import EngineFromNetwork, TrtRunner
import tensorrt_llm
from tensorrt_llm import Tensor
class TestFunctional(unittest.TestCase):
def setUp(self):
tensorrt_llm.logger.set_level('error')
def test_outer(self):
# test data
dtype = 'float32'
x_data = torch.arange(1., 5.)
y_data = torch.arange(1., 4.)
# 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_data.shape,
dtype=tensorrt_llm.str_dtype_to_trt(dtype))
y = Tensor(name='y',
shape=y_data.shape,
dtype=tensorrt_llm.str_dtype_to_trt(dtype))
output = tensorrt_llm.functional.outer(x, y).trt_tensor
output.name = 'output'
network.mark_output(output)
# trt run
build_engine = EngineFromNetwork((builder.trt_builder, net.trt_network))
with TrtRunner(build_engine) as runner:
outputs = runner.infer(feed_dict={
'x': x_data.numpy(),
'y': y_data.numpy()
})
# pytorch run
ref = torch.outer(x_data, y_data)
# compare diff
np.testing.assert_allclose(ref.cpu().numpy(), outputs['output'])