-
Notifications
You must be signed in to change notification settings - Fork 11
/
inference_schema_aug.py
323 lines (273 loc) · 11.9 KB
/
inference_schema_aug.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
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
import os
import json
import sys
import math
import torch
import argparse
# import textwrap
import transformers
from peft import PeftModel
from transformers import GenerationConfig
from llama_attn_replace import replace_llama_attn
from supervised_fine_tune import PROMPT_DICT
from tqdm import tqdm
# from queue import Queue
# from threading import Thread
# import gradio as gr
def parse_config():
parser = argparse.ArgumentParser(description='arg parser')
# parser.add_argument('--question', type=str, default="")
# parser.add_argument('--material', type=str, default="")
# parser.add_argument('--material_title', type=str, default="")
# parser.add_argument('--material_type', type=str, default="material")
parser.add_argument('--base_model', type=str, default="/data1/pretrained-models/llama-7b-hf")
parser.add_argument('--cache_dir', type=str, default="./cache")
parser.add_argument('--context_size', type=int, default=-1, help='context size during fine-tuning')
parser.add_argument('--flash_attn', type=bool, default=False, help='')
parser.add_argument('--temperature', type=float, default=0.6, help='')
parser.add_argument('--top_p', type=float, default=0.9, help='')
parser.add_argument('--max_gen_len', type=int, default=512, help='')
parser.add_argument('--input_data_file', type=str, default='input_data/', help='')
parser.add_argument('--output_data_file', type=str, default='output_data/', help='')
args = parser.parse_args()
return args
def generate_prompt(instruction, question, input_seg=None):
if input:
return PROMPT_DICT["prompt_input"].format(instruction=instruction, input_seg=input_seg, question=question)
else:
return PROMPT_DICT["prompt_no_input"].format(instruction=instruction)
# def format_prompt(material, message, material_type="book", material_title=""):
# if material_type == "paper":
# prompt = f"Below is a paper. Memorize the material and answer my question after the paper.\n {material} \n "
# elif material_type == "book":
# material_title = ", %s"%material_title if len(material_title)>0 else ""
# prompt = f"Below is some paragraphs in the book{material_title}. Memorize the content and answer my question after the book.\n {material} \n "
# else:
# prompt = f"Below is a material. Memorize the material and answer my question after the material. \n {material} \n "
# message = str(message).strip()
# prompt += f"Now the material ends. {message}"
# return prompt
# def read_txt_file(material_txt):
# if not material_txt.split(".")[-1]=='txt':
# raise ValueError("Only support txt or pdf file.")
# content = ""
# with open(material_txt) as f:
# for line in f.readlines():
# content += line
# return content
def build_generator(
item, model, tokenizer, temperature=0.6, top_p=0.9, max_gen_len=4096, use_cache=True
):
def response(item):
# def response(material, question, material_type="", material_title=None):
# material = read_txt_file(material)
# prompt = format_prompt(material, question, material_type, material_title)
prompt = generate_prompt(instruction = item["instruction"], input_seg = item["input_seg"], question = item["question"])
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
output = model.generate(
**inputs,
max_new_tokens=max_gen_len,
temperature=temperature,
top_p=top_p,
use_cache=use_cache
)
out = tokenizer.decode(output[0], skip_special_tokens=False, clean_up_tokenization_spaces=False)
out = out.split(prompt)[1].strip()
return out
return response
def main(args):
if args.flash_attn:
replace_llama_attn()
# Set RoPE scaling factor
config = transformers.AutoConfig.from_pretrained(
args.base_model,
cache_dir=args.cache_dir,
)
orig_ctx_len = getattr(config, "max_position_embeddings", None)
if orig_ctx_len and args.context_size > orig_ctx_len:
scaling_factor = float(math.ceil(args.context_size / orig_ctx_len))
config.rope_scaling = {"type": "linear", "factor": scaling_factor}
# Load model and tokenizer
model = transformers.AutoModelForCausalLM.from_pretrained(
args.base_model,
config=config,
cache_dir=args.cache_dir,
torch_dtype=torch.float16,
device_map="auto",
)
model.resize_token_embeddings(32001)
tokenizer = transformers.AutoTokenizer.from_pretrained(
args.base_model,
cache_dir=args.cache_dir,
model_max_length=args.context_size if args.context_size > orig_ctx_len else orig_ctx_len,
# padding_side="right",
padding_side="left",
use_fast=False,
)
model.eval()
if torch.__version__ >= "2" and sys.platform != "win32":
model = torch.compile(model)
with open(args.input_data_file, "r") as f:
test_data = json.load(f)
# import random
# test_data = random.sample(test_data, k=5)
test_data_pred = []
for i in tqdm(range(len(test_data))):
item = test_data[i]
new_item = {}
respond = build_generator(item, model, tokenizer, temperature=args.temperature, top_p=args.top_p,
max_gen_len=args.max_gen_len, use_cache=not args.flash_attn) # the temperature and top_p are highly different with previous alpaca exp, pay attention to this if there is sth wrong later
output = respond(item)
new_item["idx"] = i
new_item["table_id"] = test_data[i]["table_id"]
new_item["instruction"] = test_data[i]["instruction"]
new_item["input_seg"] = test_data[i]["input_seg"]
new_item["question"] = test_data[i]["question"]
new_item["target"] = test_data[i]["target"]
new_item["output_list"] = test_data[i]["output_list"]
new_item["output"] = test_data[i]["output"]
new_item["predict"] = output
test_data_pred.append(new_item)
# import pdb
# pdb.set_trace()
with open(args.output_data_file, "w") as f:
json.dump(test_data_pred, f, indent = 2)
# output = respond(args.material, args.question, args.material_type, args.material_title)
# print("output", output)
if __name__ == "__main__":
args = parse_config()
main(args)
# from dataclasses import dataclass, field
# import numpy as np
# import torch
# import transformers
# from transformers import GenerationConfig
# from train_llama2_long_context_reformat import ModelArguments, smart_tokenizer_and_embedding_resize, DEFAULT_PAD_TOKEN, DEFAULT_EOS_TOKEN, \
# DEFAULT_BOS_TOKEN, DEFAULT_UNK_TOKEN, PROMPT_DICT
# import json
# from tqdm import tqdm
# import math
# import argparse
# @dataclass
# class InferenceArguments:
# model_max_length: int = field(
# # default=512,
# # default=1024,
# default=1536,
# metadata={"help": "Maximum sequence length. Sequences will be right padded (and possibly truncated)."},
# )
# load_in_8bit: bool = field(
# default=False,
# metadata={"help": "Load the model in 8-bit mode."},
# )
# inference_dtype: torch.dtype = field(
# default=torch.float32,
# metadata={"help": "The dtype to use for inference."},
# )
# max_new_tokens: int = field(
# default=64,
# metadata={"help": "Maximum sequence length. Sequences will be right padded (and possibly truncated)."},
# )
# @dataclass
# class FileArguments:
# input_data_file: str = field(
# default="",
# metadata={"help": ""},
# )
# output_data_file: str = field(
# default="",
# metadata={"help": ""},
# )
# def batch_process(data_list, model, tokenizer, generation_config, batch_size, max_new_tokens):
# pred = []
# for i in tqdm(range(math.ceil(len(data_list)/batch_size))):
# if i != math.ceil(len(data_list)/batch_size) - 1:
# batch_data = data_list[i * batch_size: i * batch_size + batch_size]
# else:
# batch_data = data_list[i * batch_size:]
# batch_prompt =[generate_prompt(item["instruction"], item["input_seg"], item["question"]) for item in batch_data]
# inputs = tokenizer(batch_prompt,
# return_tensors="pt",
# padding="longest",
# max_length=tokenizer.model_max_length,
# truncation=True)
# outputs = model.generate(input_ids=inputs["input_ids"].cuda(), generation_config=generation_config, max_new_tokens = max_new_tokens)
# # import pdb
# # pdb.set_trace()
# # input_length = 1 if model.config.is_encoder_decoder else inputs.input_ids.shape[1]
# # generated_tokens = outputs.sequences[:, input_length:]
# # pred += tokenizer.batch_decode(generated_tokens, skip_special_tokens=False, clean_up_tokenization_spaces=False)[0]
# pred += tokenizer.batch_decode(outputs, skip_special_tokens=False, clean_up_tokenization_spaces=False)
# # import pdb
# # pdb.set_trace()
# return pred
# def inference(test_data, model_args, inference_args):
# # parser = transformers.HfArgumentParser((ModelArguments, InferenceArguments))
# # model_args, inference_args = parser.parse_args_into_dataclasses()
# model = transformers.AutoModelForCausalLM.from_pretrained(
# model_args.model_name_or_path,
# load_in_8bit=inference_args.load_in_8bit,
# torch_dtype=inference_args.inference_dtype,
# device_map="auto",
# )
# model.cuda()
# model.eval()
# generation_config = GenerationConfig(
# temperature=0.1,
# top_p=0.75,
# # num_beams=4,
# num_beams=1,
# # num_beams=2,
# )
# tokenizer = transformers.AutoTokenizer.from_pretrained(
# model_args.model_name_or_path,
# use_fast=False,
# model_max_length=inference_args.model_max_length,
# padding_side="left" ### important to add this in inference
# )
# if tokenizer.pad_token is None:
# smart_tokenizer_and_embedding_resize(
# special_tokens_dict=dict(pad_token=DEFAULT_PAD_TOKEN),
# tokenizer=tokenizer,
# model=model,
# )
# tokenizer.add_special_tokens(
# {
# "eos_token": DEFAULT_EOS_TOKEN,
# "bos_token": DEFAULT_BOS_TOKEN,
# "unk_token": DEFAULT_UNK_TOKEN,
# }
# )
# pred = batch_process(test_data, model, tokenizer, generation_config, 1, inference_args.max_new_tokens)
# new_test_list = []
# for i in tqdm(range(len(test_data))):
# # for i in tqdm(range(90, 101)):
# # for i in tqdm(range(3)):
# instruction = test_data[i]["instruction"]
# item = {}
# item["idx"] = i
# # item["table_id"] = test_data[i]["table_id"]
# # item["entity"] = test_data[i]["entity"]
# item["instruction"] = instruction
# # item["input"] = input
# item["input_seg"] = test_data[i]["input_seg"]
# # item["tokenizer_tensor_shape"] = inputs["input_ids"].shape
# item["output"] = test_data[i]["output"]
# item["predict"] = pred[i]
# new_test_list.append(item)
# return new_test_list
# if __name__ == "__main__":
# parser = transformers.HfArgumentParser((ModelArguments, InferenceArguments, FileArguments))
# model_args, inference_args, file_args = parser.parse_args_into_dataclasses()
# # num = 0
# # with open("/users/PAA0201/shubaobao/stanford_alpaca/table_all_tasks_fair/test/split_16_col_type/test_" + str(file_args.input_data_file_num) + ".json", "r") as f:
# with open(file_args.input_data_file, "r") as f:
# test_data = json.load(f)
# # import random
# # test_data = random.sample(test_data, k=5)
# test_list = inference(test_data, model_args, inference_args)
# # with open("/users/PAA0201/shubaobao/stanford_alpaca/table_all_tasks_fair/pred_ser_20000_seg/test_beam_search/test_" + str(file_args.output_data_file_num) + ".json", "w") as f:
# with open(file_args.output_data_file, "w") as f:
# json.dump(test_list, f, indent = 2)
# print("input file:", str(file_args.input_data_file))
# print("output file:", str(file_args.output_data_file))