File size: 4,911 Bytes
6ef99d1 |
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 |
import json
import re
import torch
import numpy as np
from collections import defaultdict
class summarization_data:
def __init__(self, raw_data, tokenizer, domain_adaption=False, wwm_prob=0.1):
data = self.select_data(raw_data)
self.data = self.data_clean(data)
self.len = len(self.data)
self.tokenizer = tokenizer
self.padding_max_len = "max_length"
self.domain_adaption = domain_adaption
self.wwm_prob = wwm_prob
def select_data(self, raw_data):
data = list()
for item in raw_data:
del item['pub_time']
del item['labels']
data.append(item)
return data
def data_clean(self, data):
for item in data:
item["text"] = re.sub(r"[?|!+?|:|(|)]|\\|-|/.*?/|http\S+", "", item["text"].lower())
item["title"] = re.sub(r"[?|!+?|:|(|)]|\\|-|/.*?/|http\S+", "", item["title"].lower())
return data
def __getitem__(self, index):
if self.domain_adaption:
tokenized_text = self.tokenizer(
self.data[index]["text"],
add_special_tokens=True,
padding="max_length",
return_token_type_ids=True,
truncation=True,
)
text_mask = tokenized_text['attention_mask']
input_ids, labels = self._word_masking(self.tokenizer, tokenized_text, self.wwm_prob)
return {
'input_ids': torch.tensor(input_ids),
'attention_mask': torch.tensor(text_mask),
'labels': torch.tensor(labels)
}
else:
tokenized_text = self.tokenizer(
"summarize:"+self.data[index]["text"],
add_special_tokens=True,
padding="max_length",
return_token_type_ids=True,
truncation=True,
)
text_ids = tokenized_text['input_ids']
text_mask = tokenized_text['attention_mask']
tokenized_title = self.tokenizer(
self.data[index]["title"],
padding="max_length",
return_token_type_ids=True,
truncation=True,
)
title_ids = tokenized_title['input_ids']
return {
'input_ids': torch.tensor(text_ids),
'attention_mask': torch.tensor(text_mask),
'labels': torch.tensor(title_ids)
}
def _word_masking(self, tokenizer, tokenized_inputs, wwm_prob):
# randomly mask words
input_ids = tokenized_inputs["input_ids"]
mask = np.random.binomial(1, wwm_prob, (len(input_ids),))
labels = list()
for idx in np.where(mask == 1)[0]:
#add special sentinel tokens
sentinel_token = tokenizer.additional_special_tokens[input_ids[idx] % 100]
labels.append(tokenizer(sentinel_token).input_ids[0])
labels.append(input_ids[idx])
input_ids[idx] = tokenizer(sentinel_token).input_ids[0]
return input_ids, labels
def __len__(self):
return self.len
# from tqdm.auto import tqdm
# from transformers import T5TokenizerFast, T5ForConditionalGeneration
# from transformers import DataCollatorForSeq2Seq
# from torch.utils.data import DataLoader, random_split
# from FTE_NLP.model.summarization_dataset_v1 import *
# from torch.optim import AdamW
# from transformers import get_scheduler
# from FTE_NLP.utils.post_process import *
#
# json_filename = '../data/raw_EDT/Trading_benchmark/evaluate_news_test.json'
# with open(json_filename) as data_file:
# test_data = json.loads(data_file.read())
#
# model_checkpoint = "t5-small"
# tokenizer = T5TokenizerFast.from_pretrained(model_checkpoint, model_max_length=512)
# model = T5ForConditionalGeneration.from_pretrained(model_checkpoint)
#
# all_dataset = summarization_data(test_data, tokenizer, domain_adaption=True, wwm_prob=0.1)
# train_dataset, eval_dataset = random_split(all_dataset, [7, 3], generator=torch.Generator().manual_seed(42))
#
# # data_collator = DataCollatorForSeq2Seq(tokenizer, model,label_pad_token_id=0)
# data_collator = DataCollatorForSeq2Seq(tokenizer,model)
# # pass data to dataloader
# train_params = {'batch_size': 2, 'shuffle': False, 'num_workers': 0}
# train_loader = DataLoader(train_dataset, collate_fn=data_collator, **train_params)
#
# eval_params = {'batch_size': 2, 'shuffle': False, 'num_workers': 0}
# eval_loader = DataLoader(eval_dataset, collate_fn=data_collator, **train_params)
#
#
# for item in train_loader:
# print(item)
# print("input id numbers:",item["input_ids"][0])
# print("input id: ",tokenizer.decode(item["input_ids"][0]))
# print("labels id number:",item["labels"][0])
# print("labels:",tokenizer.decode(item["labels"][0],skip_special_tokens=False))
# break |