mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-14 06:27:45 +08:00
f430a4b447
* Update TensorRT-LLM --------- Co-authored-by: IbrahimAmin <ibrahimamin532@gmail.com> Co-authored-by: Fabian Joswig <fjosw@users.noreply.github.com> Co-authored-by: Pzzzzz <hello-cd.plus@hotmail.com> Co-authored-by: CoderHam <hemant@cohere.com> Co-authored-by: Konstantin Lopuhin <kostia.lopuhin@gmail.com>
256 lines
9.0 KiB
Python
Executable File
256 lines
9.0 KiB
Python
Executable File
#!/usr/bin/env python3
|
|
# 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 argparse
|
|
import os
|
|
import platform
|
|
import sys
|
|
from pathlib import Path
|
|
from typing import Optional
|
|
|
|
from build_engines_utils import run_command, wincopy
|
|
|
|
|
|
def convert_ckpt(model_dir: str,
|
|
output_dir: str,
|
|
*args,
|
|
world_size: int = 1,
|
|
dtype: str = 'float16'):
|
|
convert_cmd = [
|
|
sys.executable, "examples/gpt/convert_checkpoint.py",
|
|
f"--model_dir={model_dir}", f"--output_dir={output_dir}",
|
|
f"--dtype={dtype}", f"--tp_size={world_size}"
|
|
] + list(args)
|
|
run_command(convert_cmd)
|
|
|
|
|
|
def build_engine(
|
|
checkpoint_dir: str,
|
|
engine_dir: str,
|
|
*args,
|
|
max_input_len: int = 256,
|
|
max_output_len: int = 128,
|
|
):
|
|
build_cmd = [
|
|
"trtllm-build", '--log_level=error',
|
|
f'--checkpoint_dir={checkpoint_dir}', f'--output_dir={engine_dir}',
|
|
'--max_batch_size=64', f'--max_input_len={max_input_len}',
|
|
f'--max_output_len={max_output_len}', '--max_beam_width=2',
|
|
'--builder_opt=0'
|
|
]
|
|
legacy_args = [
|
|
"--gpt_attention_plugin=disable",
|
|
"--context_fmha=disable",
|
|
"--paged_kv_cache=disable",
|
|
"--remove_input_padding=disable",
|
|
"--enable_xqa=disable",
|
|
]
|
|
build_cmd = build_cmd + legacy_args + list(args)
|
|
run_command(build_cmd)
|
|
|
|
|
|
def build_engines(model_cache: Optional[str] = None, world_size: int = 1):
|
|
# TODO add support of Pipeline parallelism to GPT
|
|
tp_size = world_size
|
|
pp_size = 1
|
|
|
|
resources_dir = Path(__file__).parent.resolve().parent
|
|
models_dir = resources_dir / 'models'
|
|
model_name = 'gpt2'
|
|
|
|
# Clone or update the model directory without lfs
|
|
hf_dir = models_dir / model_name
|
|
if hf_dir.exists():
|
|
assert hf_dir.is_dir()
|
|
run_command(["git", "pull"], cwd=hf_dir)
|
|
else:
|
|
if platform.system() == "Windows":
|
|
url_prefix = ""
|
|
else:
|
|
url_prefix = "file://"
|
|
|
|
model_url = url_prefix + str(
|
|
Path(model_cache) /
|
|
model_name) if model_cache else "https://huggingface.co/gpt2"
|
|
run_command([
|
|
"git", "clone", model_url, "--single-branch", "--no-local",
|
|
model_name
|
|
],
|
|
cwd=hf_dir.parent,
|
|
env={
|
|
**os.environ, "GIT_LFS_SKIP_SMUDGE": "1"
|
|
})
|
|
|
|
assert hf_dir.is_dir()
|
|
|
|
# Download the model file
|
|
model_file_name = "pytorch_model.bin"
|
|
if model_cache:
|
|
if platform.system() == "Windows":
|
|
wincopy(source=str(
|
|
Path(model_cache) / model_name / model_file_name),
|
|
dest=model_file_name,
|
|
isdir=False,
|
|
cwd=hf_dir)
|
|
else:
|
|
run_command([
|
|
"rsync", "-av",
|
|
str(Path(model_cache) / model_name / model_file_name), "."
|
|
],
|
|
cwd=hf_dir)
|
|
else:
|
|
run_command(["git", "lfs", "pull", "--include", model_file_name],
|
|
cwd=hf_dir)
|
|
|
|
safetensor_file = hf_dir / "model.safetensors"
|
|
has_safetensor = safetensor_file.exists()
|
|
if has_safetensor:
|
|
safetensor_file.rename(str(safetensor_file) + ".bak")
|
|
|
|
assert (hf_dir / model_file_name).is_file()
|
|
|
|
ckpt_dir = models_dir / 'c-model' / model_name
|
|
engine_dir = models_dir / 'rt_engine' / model_name
|
|
|
|
tp_pp_dir = f"tp{tp_size}-pp{pp_size}-gpu"
|
|
tp_dir = f"{world_size}-gpu"
|
|
|
|
print("\nConverting to fp32")
|
|
fp32_ckpt_dir = ckpt_dir / 'fp32' / tp_dir
|
|
convert_ckpt(str(hf_dir),
|
|
str(fp32_ckpt_dir),
|
|
world_size=tp_size,
|
|
dtype='float32')
|
|
|
|
print("\nBuilding fp32 engines")
|
|
build_engine(str(fp32_ckpt_dir),
|
|
str(engine_dir / 'fp32-default' / tp_pp_dir))
|
|
build_engine(str(fp32_ckpt_dir),
|
|
str(engine_dir / 'fp32-plugin' / tp_pp_dir),
|
|
'--gpt_attention_plugin=float32', '--context_fmha=enable',
|
|
'--context_fmha_fp32_acc=enable')
|
|
|
|
print("\nConverting to fp16")
|
|
fp16_ckpt_dir = ckpt_dir / 'fp16' / tp_dir
|
|
convert_ckpt(str(hf_dir),
|
|
str(fp16_ckpt_dir),
|
|
world_size=tp_size,
|
|
dtype='float16')
|
|
|
|
print("\nBuilding fp16 engines")
|
|
build_engine(str(fp16_ckpt_dir),
|
|
str(engine_dir / 'fp16-default' / tp_pp_dir))
|
|
build_engine(str(fp16_ckpt_dir),
|
|
str(engine_dir / 'fp16-plugin' / tp_pp_dir),
|
|
'--gpt_attention_plugin=float16')
|
|
build_engine(str(fp16_ckpt_dir),
|
|
str(engine_dir / 'fp16-plugin-packed' / tp_pp_dir),
|
|
'--gpt_attention_plugin=float16',
|
|
'--remove_input_padding=enable')
|
|
|
|
# this engine can be use for in-flight batching
|
|
ifb_args = [
|
|
'--gpt_attention_plugin=float16',
|
|
'--remove_input_padding=enable',
|
|
'--paged_kv_cache=enable',
|
|
'--context_fmha=enable',
|
|
'--context_fmha_fp32_acc=enable',
|
|
'--max_num_tokens=10000',
|
|
'--use_paged_context_fmha=enable',
|
|
]
|
|
build_engine(str(fp16_ckpt_dir),
|
|
str(engine_dir / 'fp16-plugin-packed-paged' / tp_pp_dir),
|
|
*ifb_args)
|
|
build_engine(
|
|
str(fp16_ckpt_dir),
|
|
str(engine_dir / 'fp16-plugin-packed-paged-draft-tokens' / tp_pp_dir),
|
|
'--max_draft_len=5',
|
|
'--speculative_decoding_mode=draft_tokens_external', *ifb_args)
|
|
build_engine(
|
|
str(fp16_ckpt_dir),
|
|
str(engine_dir / 'fp16-plugin-packed-paged-nprofiles' / tp_pp_dir),
|
|
'--multiple_profiles=enable', *ifb_args)
|
|
build_engine(str(fp16_ckpt_dir),
|
|
str(engine_dir / 'fp16-plugin-packed-paged-in128' / tp_pp_dir),
|
|
*ifb_args,
|
|
max_input_len=128)
|
|
|
|
# Build the target model with return accepted token logits
|
|
# Build with '--max_draft_len', '--speculative_decoding_mode' and '--gather_generation_logits'
|
|
build_engine(
|
|
str(fp16_ckpt_dir),
|
|
str(engine_dir /
|
|
'fp16-plugin-packed-paged-return-accepted-tokens-logits' /
|
|
tp_pp_dir), '--max_draft_len=5',
|
|
'--speculative_decoding_mode=draft_tokens_external',
|
|
'--gather_generation_logits', *ifb_args)
|
|
|
|
# We build almost the same engine twice. But this engine has gather_all_token_logits
|
|
# to extract logits from python runtime and uses context FMHA for generation to match draft model executions,
|
|
# which uses context FMHA for draft tokens prediction.
|
|
# Currently the gather_all_token_logits is not supported with target model of speculative decoding
|
|
build_engine(
|
|
str(fp16_ckpt_dir),
|
|
str(engine_dir / 'fp16-plugin-packed-paged-gather' / tp_pp_dir),
|
|
'--gather_all_token_logits', *ifb_args)
|
|
|
|
build_engine(
|
|
str(fp16_ckpt_dir),
|
|
str(engine_dir / 'fp16-plugin-packed-paged-la-decoding' / tp_pp_dir),
|
|
'--max_draft_len=64', '--speculative_decoding_mode=lookahead_decoding',
|
|
*ifb_args)
|
|
|
|
# build engine with lora enabled
|
|
build_engine(str(fp16_ckpt_dir),
|
|
str(engine_dir / "fp16-plugin-packed-paged-lora" / tp_pp_dir),
|
|
"--lora_target_modules=attn_qkv", '--lora_plugin=float16',
|
|
*ifb_args)
|
|
|
|
llm_datasets_root = Path(model_cache) / "datasets"
|
|
calib_dataset = llm_datasets_root / "cimec/lambada/"
|
|
print("\nConverting to fp16 SQ")
|
|
fp16_sq_ckpt_dir = ckpt_dir / 'fp16-sq' / tp_dir
|
|
convert_ckpt(str(hf_dir),
|
|
str(fp16_sq_ckpt_dir),
|
|
"--smoothquant=0.5",
|
|
f"--calib_dataset={calib_dataset}",
|
|
world_size=tp_size,
|
|
dtype='float16')
|
|
|
|
print("\nBuilding fp16 SQ engines")
|
|
build_engine(str(fp16_sq_ckpt_dir),
|
|
str(engine_dir / 'fp16-plugin-packed-paged-sq' / tp_pp_dir),
|
|
*ifb_args)
|
|
|
|
if has_safetensor:
|
|
Path(str(safetensor_file) + ".bak").rename(safetensor_file)
|
|
|
|
print("Done.")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
parser = argparse.ArgumentParser()
|
|
parser.add_argument("--model_cache",
|
|
type=str,
|
|
help="Directory where models are stored")
|
|
|
|
parser.add_argument('--world_size',
|
|
type=int,
|
|
default=1,
|
|
help='world size, only support tensor parallelism now')
|
|
|
|
build_engines(**vars(parser.parse_args()))
|