forked from PaddlePaddle/PaddleNLP
-
Notifications
You must be signed in to change notification settings - Fork 0
/
merge_tp_and_pp_params.py
161 lines (130 loc) · 6.32 KB
/
merge_tp_and_pp_params.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
# Copyright (c) 2024 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 importlib
import os
import re
import paddle
from paddlenlp.transformers import AutoConfig
from paddlenlp.transformers.auto.modeling import MAPPING_NAMES
from paddlenlp.utils.log import logger
def parse_arguments():
import argparse
parser = argparse.ArgumentParser()
parser.add_argument("--model_name_or_path", default=None, required=True, help="The directory of model.")
parser.add_argument("--device", type=str, default="gpu", help="Device")
parser.add_argument("--pipeline_parallel_degree", "--pp", type=int, required=True, help="pp degree")
parser.add_argument("--tensor_parallel_degree", "--tp", type=int, required=True, help="tp degree")
return parser.parse_args()
def validate_model_file(path: str, tp_degree: int, pp_degree: int) -> None:
files = os.listdir(path)
pattern = r"model_state\.tp0\d*_pp0\d*\.pdparams|model_state\.tp0\d*\.pdparams|model_state\.pp0\d*\.pdparams"
if pp_degree == 0:
target_files = [f"model_state.tp{tp:0>2d}.pdparams" for tp in range(tp_degree)]
elif tp_degree == 0:
target_files = [f"model_state.pp{pp:0>2d}.pdparams" for pp in range(pp_degree)]
else:
target_files = [
f"model_state.tp{tp:0>2d}_pp{pp:0>2d}.pdparams" for tp in range(tp_degree) for pp in range(pp_degree)
]
exist_required_files = []
for file in files:
if re.match(pattern, file):
exist_required_files.append(file)
missing_files = set(target_files) - set(exist_required_files)
if len(missing_files) > 0:
raise FileNotFoundError(f"Please check your pp/tp degree, missing files {list(missing_files)}")
def load_tp_params(tp_degree, path):
tp_state_dict_list = []
for tp in range(tp_degree):
tp_state_dict = {}
tmp = paddle.load(os.path.join(path, f"model_state.tp{tp:0>2d}.pdparams"), return_numpy=True)
for k, v in tmp.items():
tp_state_dict[k] = v
tp_state_dict_list.append(tp_state_dict)
return tp_state_dict_list
def load_tp_and_pp_params(tp_degree, pp_degree, path):
tp_state_dict_list = []
for tp in range(tp_degree):
tp_state_dict = {}
for pp in range(pp_degree):
tmp = paddle.load(os.path.join(path, f"model_state.tp{tp:0>2d}_pp{pp:0>2d}.pdparams"), return_numpy=True)
for k, v in tmp.items():
tp_state_dict[k] = v
tp_state_dict_list.append(tp_state_dict)
return tp_state_dict_list
def load_pp_params(pp_degree, path):
pp_state_dict = {}
for pp in range(pp_degree):
tmp = paddle.load(os.path.join(path, f"model_state.pp{pp:0>2d}.pdparams"), return_numpy=True)
for k, v in tmp.items():
pp_state_dict[k] = v
return pp_state_dict
def merge_tensor_parallel(model_class, state_dict_list, config) -> None:
"""the entry of converting config and converting model file
Args:
input_dir (str | None): the input dir which contains `pytorch_model.bin` and `config.json` file
config (PretrainedConfig): the PretrainedConfig instance of model
"""
name_action_mappings = model_class._get_tensor_parallel_mappings(config, is_split=False)
state_keys_map = model_class._resolve_prefix_keys(name_action_mappings.keys(), state_dict_list[0].keys())
for k, v in state_keys_map.items():
name_action_mappings[v] = name_action_mappings.pop(k)
state_dict_to_save = {}
for key in state_dict_list[0].keys():
tensor = state_dict_list[0][key]
if key in name_action_mappings:
ret = [x[key] for x in state_dict_list]
action = name_action_mappings.pop(key)
tensor = action(ret)
state_dict_to_save[key] = tensor
if len(name_action_mappings) > 0:
for x in name_action_mappings.keys():
logger.warning(f"key <{x}> need to merge tensor parallel but we can't find in model state.")
logger.info("Finally, we merging state dict to fellowing tensors.")
for k, v in state_dict_to_save.items():
logger.info(f"{k}, {v.shape}, {v.dtype}")
return state_dict_to_save
def main():
args = parse_arguments()
paddle.set_device(args.device)
config = AutoConfig.from_pretrained(args.model_name_or_path)
init_class = config["architectures"][0]
if args.pipeline_parallel_degree > 1:
# using pp
import_class = importlib.import_module(f"paddlenlp.transformers.{MAPPING_NAMES[init_class[:-15]]}.modeling_pp")
else:
# tp only
import_class = importlib.import_module(f"paddlenlp.transformers.{MAPPING_NAMES[init_class[:-11]]}.modeling")
model_class = getattr(import_class, init_class)
validate_model_file(args.model_name_or_path, args.tensor_parallel_degree, args.pipeline_parallel_degree)
if args.tensor_parallel_degree > 1:
if args.pipeline_parallel_degree > 1:
tp_state_dict_list = load_tp_and_pp_params(
args.tensor_parallel_degree, args.pipeline_parallel_degree, args.model_name_or_path
)
else:
tp_state_dict_list = load_tp_params(args.tensor_parallel_degree, args.model_name_or_path)
state_dict_to_save = merge_tensor_parallel(
model_class=model_class, state_dict_list=tp_state_dict_list, config=config
)
logger.info("Saving")
paddle.save(state_dict_to_save, os.path.join(args.model_name_or_path, "model_state.pdparams"))
elif args.pipeline_parallel_degree > 1:
state_dict_to_save = load_pp_params(args.pipeline_parallel_degree, args.model_name_or_path)
logger.info("Saving")
paddle.save(state_dict_to_save, os.path.join(args.model_name_or_path, "model_state.pdparams"))
else:
logger.info("No need to merge since config.tensor_parallel_degree <= 1.")
if __name__ == "__main__":
main()