wehub-resource-sync
211 lines
7.1 KiB
Python
211 lines
7.1 KiB
Python
# Copyright (c) 2025 PaddlePaddle Authors. All Rights Reserved.
|
|
#
|
|
# 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
|
|
|
|
from dist_llama_inference_model import LlamaInferenceModel
|
|
|
|
import paddle
|
|
import paddle.distributed as dist
|
|
from paddle import LazyGuard
|
|
from paddle.distributed import fleet
|
|
|
|
|
|
class Config:
|
|
vocab_size = 32000
|
|
hidden_size = 4096
|
|
intermediate_size = 11008
|
|
max_position_embeddings = 2048
|
|
seq_length = 2048
|
|
num_hidden_layers = 2
|
|
num_attention_heads = 32
|
|
num_key_value_heads = 32
|
|
initializer_range = 0.02
|
|
rms_norm_eps = 1e-6
|
|
use_cache = True
|
|
use_flash_attention = False
|
|
sequence_parallel = False
|
|
rope = True
|
|
recompute = False
|
|
recompute_granularity = None
|
|
use_lazy_init = False
|
|
rope_theta = 10000
|
|
tensor_parallel_degree = 1
|
|
tensor_parallel_rank = 0
|
|
dtype = 'bfloat16'
|
|
|
|
|
|
class TestLlamaExportAndPredict:
|
|
def __init__(self):
|
|
self.config = Config()
|
|
self.dp = int(os.getenv("dp")) if os.getenv("dp") is not None else 1
|
|
self.mp = int(os.getenv("mp")) if os.getenv("dp") is not None else 1
|
|
self.pp = int(os.getenv("pp")) if os.getenv("dp") is not None else 1
|
|
if os.getenv("use_sp") == "true":
|
|
self.config.sequence_parallel = True
|
|
if os.getenv("recompute") == "true":
|
|
self.config.recompute = True
|
|
if os.getenv("use_lazy_init") == "true":
|
|
self.config.use_lazy_init = True
|
|
|
|
self.init_dist_env()
|
|
|
|
def init_dist_env(self):
|
|
tensor_parallel_degree = paddle.distributed.get_world_size()
|
|
|
|
self.mp = paddle.distributed.get_world_size()
|
|
self.config.tensor_parallel_degree = self.mp
|
|
self.tensor_parallel_rank = dist.get_rank()
|
|
if tensor_parallel_degree > 1:
|
|
strategy = fleet.DistributedStrategy()
|
|
strategy.hybrid_configs = {
|
|
"dp_degree": 1,
|
|
"mp_degree": tensor_parallel_degree,
|
|
"pp_degree": 1,
|
|
"sharding_degree": 1,
|
|
}
|
|
fleet.init(is_collective=True, strategy=strategy)
|
|
|
|
def run_export(self):
|
|
if self.config.use_lazy_init:
|
|
with LazyGuard():
|
|
model = LlamaInferenceModel(self.config)
|
|
for param in model.parameters():
|
|
assert not param._is_initialized()
|
|
param.initialize()
|
|
else:
|
|
model = LlamaInferenceModel(self.config)
|
|
|
|
cache_kvs = []
|
|
for i in range(self.config.num_hidden_layers):
|
|
cache_kvs.append(
|
|
paddle.static.InputSpec(
|
|
shape=[
|
|
None,
|
|
self.config.num_key_value_heads // self.mp,
|
|
None,
|
|
None,
|
|
],
|
|
dtype=self.config.dtype,
|
|
name=f"key_caches_{i}",
|
|
)
|
|
)
|
|
cache_kvs.append(
|
|
paddle.static.InputSpec(
|
|
shape=[
|
|
None,
|
|
self.config.num_key_value_heads // self.mp,
|
|
None,
|
|
None,
|
|
],
|
|
dtype=self.config.dtype,
|
|
name=f"value_caches_{i}",
|
|
)
|
|
)
|
|
|
|
precache_input_spec = None
|
|
|
|
input_spec = [
|
|
paddle.static.InputSpec(
|
|
shape=[None, None], dtype="int64", name="input_ids"
|
|
), # input_ids
|
|
None, # attention_mask
|
|
None, # inputs_embeds
|
|
False, # use_cache
|
|
cache_kvs, # cache_kvs
|
|
None, # pre_caches,
|
|
paddle.static.InputSpec(
|
|
shape=[None, 1], dtype="int32", name="seq_lens_encoder"
|
|
), # seq_len_encoder
|
|
paddle.static.InputSpec(
|
|
shape=[None, 1], dtype="int32", name="seq_lens_decoder"
|
|
), # seq_len_decoder
|
|
None, # past_key_values
|
|
False, # output_attentions
|
|
False, # output_hidden_states
|
|
False, # return_dict
|
|
]
|
|
|
|
model = paddle.jit.to_static(model, input_spec=input_spec)
|
|
paddle.jit.save(
|
|
model, "./inference_model_unitest", skip_prune_program=True
|
|
)
|
|
|
|
def run_inference(self):
|
|
infer_model_path = "./inference_model_unitest"
|
|
if paddle.base.framework.use_pir_api():
|
|
config = paddle.inference.Config(
|
|
infer_model_path + ".json", infer_model_path + ".pdiparams"
|
|
)
|
|
else:
|
|
config = paddle.inference.Config(
|
|
infer_model_path + ".pdmodel", infer_model_path + ".pdiparams"
|
|
)
|
|
config.switch_ir_optim(False)
|
|
device_id = int(os.environ.get("FLAGS_selected_gpus", 0))
|
|
config.enable_use_gpu(100, device_id)
|
|
config.enable_new_executor()
|
|
predictor = paddle.inference.create_predictor(config)
|
|
|
|
model_inputs = {}
|
|
model_inputs["input_ids"] = paddle.to_tensor(
|
|
[[10002] * 10], dtype='int64'
|
|
)
|
|
model_inputs["seq_lens_encoder"] = paddle.to_tensor(
|
|
[10], dtype='int32'
|
|
).reshape((-1, 1))
|
|
model_inputs["seq_lens_decoder"] = paddle.to_tensor(
|
|
[0], dtype='int32'
|
|
).reshape((-1, 1))
|
|
for i in range(self.config.num_hidden_layers):
|
|
model_inputs[f"key_caches_{i}"] = paddle.full(
|
|
shape=[
|
|
96,
|
|
self.config.num_key_value_heads,
|
|
64,
|
|
self.config.hidden_size // self.config.num_attention_heads,
|
|
],
|
|
fill_value=0,
|
|
dtype=self.config.dtype,
|
|
)
|
|
model_inputs[f"value_caches_{i}"] = paddle.full(
|
|
shape=[
|
|
96,
|
|
self.config.num_key_value_heads,
|
|
64,
|
|
self.config.hidden_size // self.config.num_attention_heads,
|
|
],
|
|
fill_value=0,
|
|
dtype=self.config.dtype,
|
|
)
|
|
|
|
for name in predictor.get_input_names():
|
|
input_tensor = predictor.get_input_handle(name)
|
|
input_tensor.share_external_data(model_inputs[name])
|
|
|
|
predictor.run()
|
|
outputs_handle = predictor.get_output_handle(
|
|
predictor.get_output_names()[0]
|
|
)
|
|
result = outputs_handle.copy_to_cpu()
|
|
|
|
def run_test_cases(self):
|
|
self.run_export()
|
|
self.run_inference()
|
|
os.system("rm -rf ./inference_model_unitest*") # 删除模型文件
|
|
|
|
|
|
if __name__ == '__main__':
|
|
TestLlamaExportAndPredict().run_test_cases()
|