mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-02-09 04:31:49 +08:00
206 lines
8.1 KiB
Python
206 lines
8.1 KiB
Python
# SPDX-FileCopyrightText: Copyright (c) 2022-2024 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 torch
|
|
|
|
import tensorrt_llm
|
|
import tensorrt_llm.models.redrafter
|
|
import tensorrt_llm.models.redrafter.redrafter_helper
|
|
from tensorrt_llm import Tensor
|
|
|
|
# from tensorrt_llm.tools.onnx_utils import to_onnx
|
|
|
|
sys.path.append(os.path.join(os.path.dirname(__file__), os.pardir, os.pardir))
|
|
from utils.util import create_session, run_session
|
|
|
|
T, F = True, False
|
|
|
|
|
|
class TestReDrafter(unittest.TestCase):
|
|
|
|
def setUp(self):
|
|
tensorrt_llm.logger.set_level('warning')
|
|
|
|
def test_beams2tree(self):
|
|
# test data
|
|
bs = 2
|
|
nb = 3
|
|
bl = 4
|
|
old_device = torch.get_default_device()
|
|
torch.set_default_device("cuda")
|
|
beams = torch.tensor(
|
|
[ # Assuming a batch of two sequences, each has 3 beams of 4 tokens.
|
|
[
|
|
[91, 92, 93, 95],
|
|
[91, 92, 94, 96],
|
|
[91, 92, 93, 97],
|
|
],
|
|
[
|
|
[93, 94, 95, 92],
|
|
[93, 95, 96, 93],
|
|
[93, 94, 97, 96],
|
|
],
|
|
],
|
|
dtype=torch.int32,
|
|
)
|
|
position_ids_base = torch.tensor([3, 10], dtype=torch.int32)
|
|
assert beams.shape == (bs, nb, bl)
|
|
assert position_ids_base.shape == (bs, )
|
|
# ref outputs
|
|
ref_active_flat_tokens = torch.tensor(
|
|
[91, 92, 93, 95, 94, 96, 97, 93, 94, 95, 92, 95, 96, 93, 97, 96],
|
|
dtype=torch.int32)
|
|
ref_active_token_indices = torch.tensor(
|
|
[[0, 1, 2, 3, 6, 7, 11, 0, 1], [0, 1, 2, 3, 5, 6, 7, 10, 11]],
|
|
dtype=torch.int32)
|
|
ref_draft_token_indices = torch.tensor(
|
|
[
|
|
[
|
|
[0, 1, 2, 3],
|
|
[0, 1, 4, 5],
|
|
[0, 1, 2, 6],
|
|
],
|
|
[
|
|
[0, 1, 2, 3],
|
|
[0, 4, 5, 6],
|
|
[0, 1, 7, 8],
|
|
],
|
|
],
|
|
dtype=torch.int32,
|
|
)
|
|
ref_total_lengths = torch.tensor([7, 9], dtype=torch.int32)
|
|
ref_max_len = ref_total_lengths.max().int()
|
|
ref_total_gen_len = ref_total_lengths.sum().int()
|
|
ref_position_offsets = ref_active_token_indices % bl
|
|
position_ids = ref_position_offsets + position_ids_base.unsqueeze(1)
|
|
ref_packed_position_ids = torch.concat(
|
|
[position_ids[b, :ref_total_lengths[b]] for b in range(bs)]).int()
|
|
ref_mask = torch.tensor([[[T, F, F, F, F, F, F, F, F],
|
|
[T, T, F, F, F, F, F, F, F],
|
|
[T, T, T, F, F, F, F, F, F],
|
|
[T, T, T, T, F, F, F, F, F],
|
|
[T, T, F, F, T, F, F, F, F],
|
|
[T, T, F, F, T, T, F, F, F],
|
|
[T, T, T, F, F, F, T, F, F],
|
|
[T, F, F, F, F, F, F, F, F],
|
|
[T, T, F, F, F, F, F, F, F]],
|
|
[[T, F, F, F, F, F, F, F, F],
|
|
[T, T, F, F, F, F, F, F, F],
|
|
[T, T, T, F, F, F, F, F, F],
|
|
[T, T, T, T, F, F, F, F, F],
|
|
[T, F, F, F, T, F, F, F, F],
|
|
[T, F, F, F, T, T, F, F, F],
|
|
[T, F, F, F, T, T, T, F, F],
|
|
[T, T, F, F, F, F, F, T, F],
|
|
[T, T, F, F, F, F, F, T, T]]])
|
|
|
|
# construct trt network
|
|
builder = tensorrt_llm.Builder()
|
|
network = builder.create_network()
|
|
with tensorrt_llm.net_guard(network):
|
|
beams_t = Tensor(name='beams',
|
|
shape=beams.shape,
|
|
dtype=tensorrt_llm.str_dtype_to_trt('int32'))
|
|
position_ids_base_t = Tensor(
|
|
name='position_ids_base',
|
|
shape=position_ids_base.shape,
|
|
dtype=tensorrt_llm.str_dtype_to_trt('int32'))
|
|
|
|
outputs = tensorrt_llm.models.redrafter.redrafter_helper._beams2tree(
|
|
beams_t, nb, bl, position_ids_base_t)
|
|
outputs[0].mark_output('active_tokens_flattened')
|
|
outputs[1].mark_output('draft_token_indices')
|
|
outputs[2].mark_output('spec_decoding_mask')
|
|
outputs[3].mark_output('position_offsets')
|
|
outputs[4].mark_output('packed_position_ids')
|
|
outputs[5].mark_output('total_lengths')
|
|
outputs[6].mark_output('max_gen_len')
|
|
outputs[7].mark_output('total_gen_len')
|
|
# save onnx
|
|
# model_path = 'b2t.onnx'
|
|
# to_onnx(network.trt_network, model_path)
|
|
session = create_session(
|
|
builder,
|
|
network,
|
|
precision='float32',
|
|
)
|
|
inputs = {
|
|
'beams': beams,
|
|
'position_ids_base': position_ids_base,
|
|
}
|
|
outputs = {
|
|
"active_tokens_flattened":
|
|
torch.empty((bs * nb * bl, ), dtype=torch.int32),
|
|
"draft_token_indices":
|
|
torch.empty((bs, nb, bl), dtype=torch.int32),
|
|
"spec_decoding_mask":
|
|
torch.empty((bs * nb * bl * nb * bl, ), dtype=torch.bool),
|
|
"position_offsets":
|
|
torch.empty((bs * nb * bl, ), dtype=torch.int32),
|
|
"packed_position_ids":
|
|
torch.empty((bs * nb * bl, ), dtype=torch.int32),
|
|
"total_lengths":
|
|
torch.empty((bs, ), dtype=torch.int32),
|
|
"max_len":
|
|
torch.empty((), dtype=torch.int32),
|
|
"total_gen_len":
|
|
torch.empty((), dtype=torch.int32),
|
|
}
|
|
outputs = run_session(session, inputs, outputs)
|
|
# # compare diff
|
|
torch.testing.assert_close(outputs['max_gen_len'],
|
|
ref_max_len,
|
|
rtol=0,
|
|
atol=0)
|
|
torch.testing.assert_close(outputs['total_gen_len'],
|
|
ref_total_gen_len,
|
|
rtol=0,
|
|
atol=0)
|
|
torch.testing.assert_close(outputs['total_lengths'],
|
|
ref_total_lengths,
|
|
rtol=0,
|
|
atol=0)
|
|
torch.testing.assert_close(
|
|
outputs['active_tokens_flattened'][:ref_total_gen_len],
|
|
ref_active_flat_tokens,
|
|
rtol=0,
|
|
atol=0)
|
|
torch.testing.assert_close(outputs['draft_token_indices'],
|
|
ref_draft_token_indices,
|
|
rtol=0,
|
|
atol=0)
|
|
torch.testing.assert_close(
|
|
outputs["spec_decoding_mask"][:bs * ref_max_len * ref_max_len].view(
|
|
bs, ref_max_len, ref_max_len),
|
|
ref_mask,
|
|
rtol=0,
|
|
atol=0)
|
|
torch.testing.assert_close(
|
|
outputs['position_offsets'][:bs * ref_max_len].view(
|
|
bs, ref_max_len),
|
|
ref_position_offsets,
|
|
rtol=0,
|
|
atol=0)
|
|
torch.testing.assert_close(
|
|
outputs['packed_position_ids'][:ref_total_gen_len],
|
|
ref_packed_position_ids,
|
|
rtol=0,
|
|
atol=0)
|
|
torch.set_default_device(old_device)
|
|
return
|