mirror of https://github.com/hpcaitech/ColossalAI
You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
276 lines
10 KiB
276 lines
10 KiB
1 year ago
|
import argparse
|
||
|
import time
|
||
11 months ago
|
from contextlib import nullcontext
|
||
1 year ago
|
|
||
|
import torch
|
||
|
import torch.distributed as dist
|
||
|
import transformers
|
||
11 months ago
|
from transformers import AutoTokenizer, GenerationConfig
|
||
10 months ago
|
from vllm import LLM, SamplingParams
|
||
1 year ago
|
|
||
|
import colossalai
|
||
11 months ago
|
from colossalai.accelerator import get_accelerator
|
||
10 months ago
|
from colossalai.inference.config import InferenceConfig
|
||
|
from colossalai.inference.core.engine import InferenceEngine
|
||
1 year ago
|
from colossalai.testing import clear_cache_before_run, rerun_if_address_is_in_use, spawn
|
||
|
|
||
|
GIGABYTE = 1024**3
|
||
|
MEGABYTE = 1024 * 1024
|
||
|
|
||
|
CONFIG_MAP = {
|
||
|
"toy": transformers.LlamaConfig(num_hidden_layers=4),
|
||
|
"llama-7b": transformers.LlamaConfig(
|
||
|
hidden_size=4096,
|
||
|
intermediate_size=11008,
|
||
|
num_attention_heads=32,
|
||
|
num_hidden_layers=32,
|
||
|
num_key_value_heads=32,
|
||
|
max_position_embeddings=2048,
|
||
|
),
|
||
|
"llama-13b": transformers.LlamaConfig(
|
||
|
hidden_size=5120,
|
||
|
intermediate_size=13824,
|
||
|
num_attention_heads=40,
|
||
|
num_hidden_layers=40,
|
||
|
num_key_value_heads=40,
|
||
|
max_position_embeddings=2048,
|
||
|
),
|
||
|
"llama2-7b": transformers.LlamaConfig(
|
||
|
hidden_size=4096,
|
||
|
intermediate_size=11008,
|
||
|
num_attention_heads=32,
|
||
|
num_hidden_layers=32,
|
||
|
num_key_value_heads=32,
|
||
|
max_position_embeddings=4096,
|
||
|
),
|
||
|
"llama2-13b": transformers.LlamaConfig(
|
||
|
hidden_size=5120,
|
||
|
intermediate_size=13824,
|
||
|
num_attention_heads=40,
|
||
|
num_hidden_layers=40,
|
||
|
num_key_value_heads=40,
|
||
|
max_position_embeddings=4096,
|
||
|
),
|
||
7 months ago
|
"llama3-8b": transformers.LlamaConfig(
|
||
|
hidden_size=4096,
|
||
|
intermediate_size=14336,
|
||
|
num_attention_heads=32,
|
||
|
num_hidden_layers=32,
|
||
|
num_key_value_heads=8,
|
||
|
max_position_embeddings=8192,
|
||
|
),
|
||
|
"llama3-70b": transformers.LlamaConfig(
|
||
|
hidden_size=8192,
|
||
|
intermediate_size=28672,
|
||
|
num_attention_heads=64,
|
||
|
num_hidden_layers=80,
|
||
|
num_key_value_heads=8,
|
||
|
max_position_embeddings=8192,
|
||
|
),
|
||
1 year ago
|
}
|
||
|
|
||
|
|
||
|
def data_gen(batch_size: int = 4, seq_len: int = 512):
|
||
11 months ago
|
input_ids = torch.randint(10, 30000, (batch_size, seq_len), device=get_accelerator().get_current_device())
|
||
11 months ago
|
return input_ids
|
||
1 year ago
|
|
||
|
|
||
10 months ago
|
def print_details_info(model_config, args, whole_end2end, total_token_num):
|
||
1 year ago
|
msg: str = ""
|
||
|
|
||
|
if dist.get_rank() == 0:
|
||
|
msg += "-------Perf Summary-------\n"
|
||
10 months ago
|
whole_avg_latency = whole_end2end / (total_token_num)
|
||
1 year ago
|
num_layers = getattr(model_config, "num_layers", model_config.num_hidden_layers)
|
||
7 months ago
|
num_parameters = num_layers * model_config.hidden_size * model_config.hidden_size * 12
|
||
1 year ago
|
if args.dtype in ["fp16", "bf16"]:
|
||
|
num_bytes = 2
|
||
|
else:
|
||
|
num_bytes = 4
|
||
|
|
||
|
msg += f"Whole batch end2end time: {whole_end2end * 1000:.2f} ms\n"
|
||
|
msg += f"Whole batch per token latency: {whole_avg_latency * 1000:.2f} ms\n"
|
||
10 months ago
|
msg += f"Throughput: {total_token_num / whole_end2end:.2f} tokens/s\n"
|
||
1 year ago
|
msg += f"Flops: {num_parameters * num_bytes / whole_avg_latency / 1e12:.2f} TFLOPS\n"
|
||
|
|
||
|
if torch.cuda.is_available():
|
||
11 months ago
|
msg += f"-------Memory Summary Device:{get_accelerator().current_device()}-------\n"
|
||
|
msg += f"Max memory allocated: {get_accelerator().max_memory_allocated() / GIGABYTE:.2f} GB\n"
|
||
|
msg += f"Max memory reserved: {get_accelerator().max_memory_reserved() / GIGABYTE:.2f} GB\n"
|
||
1 year ago
|
|
||
|
print(msg)
|
||
|
|
||
|
|
||
|
def benchmark_inference(args):
|
||
11 months ago
|
with torch.no_grad():
|
||
|
config = CONFIG_MAP[args.model]
|
||
|
config.pad_token_id = config.eos_token_id
|
||
10 months ago
|
|
||
7 months ago
|
if args.mode != "vllm":
|
||
|
if args.test_random_weight:
|
||
|
model = transformers.LlamaForCausalLM(config).cuda()
|
||
|
tokenizer = AutoTokenizer.from_pretrained("hf-internal-testing/llama-tokenizer")
|
||
|
else:
|
||
|
assert args.model_path, "When testing pretrained weights, the model path must be provided.'"
|
||
|
model = transformers.LlamaForCausalLM.from_pretrained(args.model_path).cuda()
|
||
|
tokenizer = AutoTokenizer.from_pretrained(args.model_path)
|
||
|
|
||
|
model = model.eval()
|
||
11 months ago
|
|
||
7 months ago
|
if args.dtype == "fp16":
|
||
|
model = model.half()
|
||
|
elif args.dtype == "bf16":
|
||
|
model = model.to(torch.bfloat16)
|
||
|
|
||
|
generation_config = GenerationConfig(
|
||
|
pad_token_id=tokenizer.pad_token_id,
|
||
|
max_length=args.seq_len + args.output_len,
|
||
|
# max_new_tokens=args.max_output_len,
|
||
|
)
|
||
11 months ago
|
|
||
10 months ago
|
if args.continous_batching:
|
||
|
mbsz = args.mbsz
|
||
|
else:
|
||
|
mbsz = args.batch_size
|
||
10 months ago
|
if args.mode == "colossalai":
|
||
11 months ago
|
inference_config = InferenceConfig(
|
||
|
dtype=args.dtype,
|
||
|
max_batch_size=mbsz,
|
||
|
max_input_len=args.seq_len,
|
||
|
max_output_len=args.output_len,
|
||
|
prefill_ratio=1.2,
|
||
10 months ago
|
block_size=32,
|
||
7 months ago
|
tp_size=args.tp_size,
|
||
8 months ago
|
use_cuda_kernel=True,
|
||
11 months ago
|
)
|
||
|
engine = InferenceEngine(model, tokenizer, inference_config, verbose=True)
|
||
10 months ago
|
elif args.mode == "vllm":
|
||
|
engine = LLM(
|
||
|
model=args.model_path,
|
||
9 months ago
|
tokenizer="hf-internal-testing/llama-tokenizer",
|
||
10 months ago
|
max_num_seqs=mbsz,
|
||
|
dtype="float16",
|
||
|
enforce_eager=True,
|
||
|
)
|
||
|
|
||
|
sampling_params = SamplingParams(
|
||
|
max_tokens=args.output_len,
|
||
|
)
|
||
11 months ago
|
else:
|
||
|
engine = model
|
||
|
|
||
|
data = data_gen(mbsz, args.seq_len)
|
||
10 months ago
|
|
||
8 months ago
|
if args.mode == "colossalai" or args.mode == "vllm":
|
||
|
data = data.tolist()
|
||
10 months ago
|
|
||
11 months ago
|
N_WARMUP_STEPS = 2
|
||
|
|
||
|
ctx = (
|
||
|
torch.profiler.profile(
|
||
|
record_shapes=True,
|
||
|
with_stack=True,
|
||
|
with_modules=True,
|
||
|
activities=[
|
||
|
torch.profiler.ProfilerActivity.CPU,
|
||
|
torch.profiler.ProfilerActivity.CUDA,
|
||
|
],
|
||
|
schedule=torch.profiler.schedule(wait=0, warmup=N_WARMUP_STEPS, active=1),
|
||
10 months ago
|
on_trace_ready=torch.profiler.tensorboard_trace_handler(f"./tb_log_{args.batch_size}_" + args.mode),
|
||
11 months ago
|
)
|
||
|
if args.profile
|
||
|
else nullcontext()
|
||
|
)
|
||
|
|
||
|
with ctx:
|
||
|
for _ in range(N_WARMUP_STEPS):
|
||
10 months ago
|
if args.mode == "colossalai":
|
||
10 months ago
|
engine.generate(prompts_token_ids=data, generation_config=generation_config)
|
||
10 months ago
|
elif args.mode == "vllm":
|
||
|
engine.generate(prompt_token_ids=data, sampling_params=sampling_params)
|
||
11 months ago
|
else:
|
||
|
engine.generate(data, generation_config=generation_config)
|
||
|
if args.profile:
|
||
|
ctx.step()
|
||
|
|
||
|
if args.nsys:
|
||
|
torch.cuda.cudart().cudaProfilerStart()
|
||
|
|
||
|
torch.cuda.synchronize()
|
||
|
|
||
|
whole_end2end = time.perf_counter()
|
||
10 months ago
|
|
||
|
if args.mode == "colossalai":
|
||
11 months ago
|
for _ in range(args.batch_size // mbsz):
|
||
10 months ago
|
output, output_tokens_list = engine.generate(
|
||
|
prompts_token_ids=data, generation_config=generation_config, return_token_ids=True
|
||
|
)
|
||
|
elif args.mode == "vllm":
|
||
|
for _ in range(args.batch_size // mbsz):
|
||
|
output = engine.generate(prompt_token_ids=data, sampling_params=sampling_params)
|
||
11 months ago
|
else:
|
||
|
for _ in range(args.batch_size // mbsz):
|
||
10 months ago
|
output = engine.generate(data, generation_config=generation_config)
|
||
|
|
||
11 months ago
|
whole_end2end = time.perf_counter() - whole_end2end
|
||
10 months ago
|
|
||
|
if args.mode == "colossalai":
|
||
|
total_token_num = sum([len(output_tokens) for output_tokens in output_tokens_list])
|
||
|
elif args.mode == "vllm":
|
||
|
total_token_num = sum([len(out.outputs[0].token_ids) for out in output])
|
||
|
else:
|
||
|
total_token_num = sum([len(out) for out in output])
|
||
|
|
||
|
print("total_token_num: ", total_token_num)
|
||
11 months ago
|
if args.nsys:
|
||
|
torch.cuda.cudart().cudaProfilerStop()
|
||
|
if args.profile:
|
||
|
ctx.step()
|
||
9 months ago
|
print(f"config:batch_size {args.batch_size}, input_len{ args.seq_len}, output_len {args.output_len}")
|
||
7 months ago
|
print_details_info(config, args, whole_end2end, total_token_num)
|
||
1 year ago
|
|
||
|
|
||
|
def hybrid_inference(rank, world_size, port, args):
|
||
7 months ago
|
colossalai.launch(rank=rank, world_size=world_size, host="localhost", port=port, backend="nccl")
|
||
1 year ago
|
benchmark_inference(args)
|
||
|
|
||
|
|
||
|
@rerun_if_address_is_in_use()
|
||
|
@clear_cache_before_run()
|
||
|
def benchmark(args):
|
||
7 months ago
|
spawn(hybrid_inference, nprocs=args.tp_size, args=args)
|
||
1 year ago
|
|
||
|
|
||
|
if __name__ == "__main__":
|
||
|
parser = argparse.ArgumentParser()
|
||
|
parser.add_argument(
|
||
|
"-m",
|
||
|
"--model",
|
||
|
default="toy",
|
||
|
help="the size of model",
|
||
7 months ago
|
choices=["toy", "llama-7b", "llama-13b", "llama2-7b", "llama2-13b", "llama3-8b", "llama3-70b"],
|
||
1 year ago
|
)
|
||
10 months ago
|
parser.add_argument("--model_path", type=str, default=None, help="The pretrained weights path")
|
||
1 year ago
|
parser.add_argument("-b", "--batch_size", type=int, default=8, help="batch size")
|
||
11 months ago
|
parser.add_argument("--mbsz", type=int, default=8, help="batch size for one step")
|
||
|
parser.add_argument("-s", "--seq_len", type=int, default=8, help="input sequence length")
|
||
7 months ago
|
parser.add_argument("--tp_size", type=int, default=1, help="Tensor Parallelism size")
|
||
1 year ago
|
parser.add_argument("--output_len", type=int, default=128, help="Output length")
|
||
11 months ago
|
parser.add_argument("--dtype", type=str, default="fp16", help="data type", choices=["fp16", "fp32", "bf16"])
|
||
10 months ago
|
parser.add_argument(
|
||
|
"--test_random_weight", default=False, action="store_true", help="whether to test random weight"
|
||
|
)
|
||
11 months ago
|
parser.add_argument("--profile", default=False, action="store_true", help="enable torch profiler")
|
||
|
parser.add_argument("--nsys", default=False, action="store_true", help="enable nsys profiler")
|
||
|
parser.add_argument(
|
||
|
"--mode",
|
||
10 months ago
|
default="colossalai",
|
||
|
choices=["colossalai", "transformers", "vllm"],
|
||
11 months ago
|
help="decide which inference framework to run",
|
||
|
)
|
||
10 months ago
|
parser.add_argument(
|
||
|
"-cb", "--continous_batching", default=False, action="store_true", help="enable continous batching"
|
||
|
)
|
||
1 year ago
|
args = parser.parse_args()
|
||
|
benchmark(args)
|