mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-14 06:27:45 +08:00
655524dd82
* Update TensorRT-LLM --------- Co-authored-by: Bhuvanesh Sridharan <bhuvan.sridharan@gmail.com> Co-authored-by: Shixiaowei02 <39303645+Shixiaowei02@users.noreply.github.com>
125 lines
4.3 KiB
Python
Executable File
125 lines
4.3 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 platform
|
|
import shutil
|
|
import sys
|
|
import typing
|
|
from pathlib import Path
|
|
|
|
from build_engines_utils import run_command, wincopy
|
|
|
|
resources_dir = Path(
|
|
__file__).parent.parent.parent.parent.parent / "examples/chatglm"
|
|
sys.path.insert(0, str(resources_dir))
|
|
|
|
engine_target_path = Path(__file__).parent.parent / "models/rt_engine"
|
|
|
|
|
|
def convert_ckpt(model_dir: str, output_dir: str, world_size: int):
|
|
convert_cmd = [
|
|
sys.executable,
|
|
str(resources_dir / "convert_checkpoint.py"), "--dtype=float16",
|
|
f"--model_dir={model_dir}", f"--output_dir={output_dir}",
|
|
f"--tp_size={world_size}"
|
|
]
|
|
print("Running: " + " ".join(convert_cmd))
|
|
run_command(convert_cmd)
|
|
|
|
|
|
def build_engine(ckpt_dir: str, engine_dir: str):
|
|
build_cmd = [
|
|
"trtllm-build", f"--checkpoint_dir={ckpt_dir}",
|
|
f"--output_dir={engine_dir}", "--log_level=error", "--max_batch_size=2",
|
|
"--max_beam_width=4", "--max_input_len=512", "--max_output_len=512",
|
|
"--gpt_attention_plugin=float16", "--gemm_plugin=float16",
|
|
"--builder_opt=0", "--remove_input_padding=disable",
|
|
"--paged_kv_cache=disable"
|
|
]
|
|
print("Running: " + " ".join(build_cmd))
|
|
run_command(build_cmd)
|
|
|
|
|
|
def build_engines(model_cache: typing.Optional[str] = None,
|
|
world_size: int = 1):
|
|
Path(engine_target_path).mkdir(parents=True, exist_ok=True)
|
|
|
|
run_command(
|
|
["pip", "install", "-r",
|
|
str(resources_dir) + "/requirements.txt"],
|
|
cwd=resources_dir)
|
|
|
|
for model_name in ["chatglm_6b", "chatglm2_6b", "chatglm3_6b"]:
|
|
# Get original model
|
|
model_cache_dir = Path(model_cache) / model_name.replace("_", "-")
|
|
if model_cache_dir.is_dir():
|
|
print("Copy model from model_cache")
|
|
if platform.system() == "Windows":
|
|
wincopy(source=str(model_cache_dir),
|
|
dest=model_name,
|
|
isdir=True,
|
|
cwd=resources_dir)
|
|
else:
|
|
run_command(
|
|
["rsync", "-av", str(model_cache_dir), "."],
|
|
cwd=resources_dir)
|
|
shutil.move(resources_dir / model_name.replace("_", "-"),
|
|
resources_dir / model_name)
|
|
else:
|
|
print("Clone model from HF")
|
|
run_command(
|
|
[
|
|
"git", "clone",
|
|
f"https://huggingface.co/THUDM/{model_name.replace('_', '-')}",
|
|
model_name
|
|
],
|
|
cwd=resources_dir,
|
|
)
|
|
|
|
# Build engines
|
|
print(f"Building {model_name}")
|
|
weight_dir = Path(resources_dir) / model_name
|
|
ckpt_dir = Path(resources_dir) / "trt_ckpt" / model_name
|
|
trt_dir = Path(resources_dir) / "trt_engines" / model_name
|
|
|
|
# fix remained error in chatglm_6b, hope to remove this in the future
|
|
if model_name == "chatglm_6b":
|
|
shutil.copy(
|
|
Path(resources_dir) / "tokenization_chatglm.py",
|
|
weight_dir,
|
|
)
|
|
|
|
convert_ckpt(weight_dir, ckpt_dir, world_size)
|
|
build_engine(ckpt_dir, trt_dir)
|
|
shutil.move(trt_dir, engine_target_path)
|
|
|
|
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()))
|