Spaces:
Sleeping
Sleeping
File size: 11,167 Bytes
ba79e72 f1145a6 ba79e72 32653f7 ba79e72 80ae5a7 ba79e72 80ae5a7 ba79e72 32653f7 ba79e72 e8ccc6c ba79e72 b70a00f ba79e72 80ae5a7 ba79e72 e38fd70 ba79e72 |
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 |
import datetime
import numpy as np
import pandas as pd
import re
import json
import os
import glob
import torch
import torch.nn.functional as F
from torch.optim import Adam
from tqdm import tqdm
from torch import nn
from transformers import BertModel
from transformers import AutoTokenizer
import argparse
def split_essay_to_sentence(origin_essay):
origin_essay_sentence = sum([[a.strip() for a in i.split('.')] for i in origin_essay.split('\n')], [])
essay_sent = [a for a in origin_essay_sentence if len(a) > 0]
return essay_sent
def get_first_extraction(text_sentence):
row_dict = {}
for row in tqdm(text_sentence):
question = 'what is the feeling?'
answer = question_answerer(question=question, context=row)
row_dict[row] = answer
return row_dict
def get_sent_labeldata():
label =pd.read_csv('./rawdata/sentimental_label.csv', encoding = 'cp949', header = None)
label[1] = label[1].apply(lambda x : re.findall(r'[๊ฐ-ํฃ]+', x)[0])
label_dict =label[label.index % 10 == 0].set_index(0).to_dict()[1]
emo2idx = {v : k for k, v in enumerate(label_dict.items())}
idx2emo = {v : k[1] for k, v in emo2idx.items()}
return emo2idx, idx2emo
def load_model():
class BertClassifier(nn.Module):
def __init__(self, dropout = 0.3):
super(BertClassifier, self).__init__()
self.bert= BertModel.from_pretrained('bert-base-multilingual-cased')
self.dropout = nn.Dropout(dropout)
self.linear = nn.Linear(768, 6)
self.relu = nn.ReLU()
def forward(self, input_id, mask):
_, pooled_output = self.bert(input_ids = input_id, attention_mask = mask, return_dict = False)
dropout_output = self.dropout(pooled_output)
linear_output = self.linear(dropout_output)
final_layer= self.relu(linear_output)
return final_layer
tokenizer = AutoTokenizer.from_pretrained('bert-base-multilingual-cased')
device = 'cuda' if torch.cuda.is_available() else 'cpu'
cls_model = BertClassifier()
criterion = nn.CrossEntropyLoss()
model_name = 'bert-base-multilingual-cased'
PATH = './model' + '/' + model_name + '_' + '2023102410'
print(PATH)
cls_model = torch.load(PATH)
#cls_model.load_state_dict(torch.load(PATH))
return tokenizer, cls_model
class myDataset_for_infer(torch.utils.data.Dataset):
def __init__(self, X):
self.X = X
def __len__(self):
return len(self.X)
def __getitem__(self,idx):
sentences = tokenizer(self.X[idx], return_tensors = 'pt', padding = 'max_length', max_length = 128, truncation = True)
return sentences
def infer_data(model, main_feeling_keyword):
#ds = myDataset_for_infer()
df_infer = myDataset_for_infer(main_feeling_keyword)
infer_dataloader = torch.utils.data.DataLoader(df_infer, batch_size= 16)
device = 'cuda' if torch.cuda.is_available() else 'cpu'
if device == 'cuda':
model = model.cuda()
result_list = []
with torch.no_grad():
for idx, infer_input in tqdm(enumerate(infer_dataloader)):
mask = infer_input['attention_mask'].to(device)
input_id = infer_input['input_ids'].squeeze(1).to(device)
output = model(input_id, mask)
result = np.argmax(F.softmax(output, dim=0).cpu(), axis=1).numpy()
result_list.extend(result)
return result_list
def get_word_emotion_pair(cls_model, origin_essay_sentence):
from konlpy.tag import Okt
okt = Okt()
#text = '๋๋ ์ ์๋ง๋ง ๋ฏธ์ํ์๊น'
def get_noun(text):
noun_list = [k for k, v in okt.pos(text) if (v == 'Noun' and len(k) > 1)]
return noun_list
def get_adj(text):
adj_list = [k for k, v in okt.pos(text) if (v == 'Adjective') and (len(k) > 1)]
return adj_list
def get_verb(text):
verb_list = [k for k, v in okt.pos(text) if (v == 'Verb') and (len(k) > 1)]
return verb_list
result_list = infer_data(cls_model, origin_essay_sentence)
final_result = pd.DataFrame(data = {'text': origin_essay_sentence , 'label' : result_list})
final_result['emotion'] = final_result['label'].map(idx2emo)
final_result['noun_list'] = final_result['text'].map(get_noun)
final_result['adj_list'] = final_result['text'].map(get_adj)
final_result['verb_list'] = final_result['text'].map(get_verb)
final_result['title'] = 'none'
file_made_dt = datetime.datetime.now()
file_made_dt_str = datetime.datetime.strftime(file_made_dt, '%Y%m%d_%H%M%d')
os.makedirs(f'./result/{file_made_dt_str}/', exist_ok = True)
final_result.to_csv(f"./result/{file_made_dt_str}/essay_result.csv", index = False)
return final_result, file_made_dt_str
def get_essay_base_analysis(file_made_dt_str):
essay1 = pd.read_csv(f"./result/{file_name_dt}/essay_result.csv")
essay1['noun_list_len'] = essay1['noun_list'].apply(lambda x : len(x))
essay1['noun_list_uniqlen'] = essay1['noun_list'].apply(lambda x : len(set(x)))
essay1['adj_list_len'] = essay1['adj_list'].apply(lambda x : len(x))
essay1['adj_list_uniqlen'] = essay1['adj_list'].apply(lambda x : len(set(x)))
essay1['vocab_all'] = essay1[['noun_list','adj_list']].apply(lambda x : sum((eval(x[0]),eval(x[1])), []), axis=1)
essay1['vocab_cnt'] = essay1['vocab_all'].apply(lambda x : len(x))
essay1['vocab_unique_cnt'] = essay1['vocab_all'].apply(lambda x : len(set(x)))
essay1['noun_list'] = essay1['noun_list'].apply(lambda x : eval(x))
essay1['adj_list'] = essay1['adj_list'].apply(lambda x : eval(x))
d = essay1.groupby('title')[['noun_list','adj_list']].sum([]).reset_index()
d['noun_cnt'] = d['noun_list'].apply(lambda x : len(set(x)))
d['adj_cnt'] = d['adj_list'].apply(lambda x : len(set(x)))
# ๋ฌธ์ฅ ๊ธฐ์ค ์ต๊ณ ๊ฐ์
essay_summary =essay1.groupby(['title'])['emotion'].value_counts().unstack(level =1)
emo_vocab_dict = {}
for k, v in essay1[['emotion','noun_list']].values:
for vocab in v:
if (k, 'noun', vocab) not in emo_vocab_dict:
emo_vocab_dict[(k, 'noun', vocab)] = 0
emo_vocab_dict[(k, 'noun', vocab)] += 1
for k, v in essay1[['emotion','adj_list']].values:
for vocab in v:
if (k, 'adj', vocab) not in emo_vocab_dict:
emo_vocab_dict[(k, 'adj', vocab)] = 0
emo_vocab_dict[(k, 'adj', vocab)] += 1
vocab_emo_cnt_dict = {}
for k, v in essay1[['emotion','noun_list']].values:
for vocab in v:
if (vocab, 'noun') not in vocab_emo_cnt_dict:
vocab_emo_cnt_dict[('noun', vocab)] = {}
if k not in vocab_emo_cnt_dict[( 'noun', vocab)]:
vocab_emo_cnt_dict[( 'noun', vocab)][k] = 0
vocab_emo_cnt_dict[('noun', vocab)][k] += 1
for k, v in essay1[['emotion','adj_list']].values:
for vocab in v:
if ('adj', vocab) not in vocab_emo_cnt_dict:
vocab_emo_cnt_dict[( 'adj', vocab)] = {}
if k not in vocab_emo_cnt_dict[( 'adj', vocab)]:
vocab_emo_cnt_dict[( 'adj', vocab)][k] = 0
vocab_emo_cnt_dict[('adj', vocab)][k] += 1
vocab_emo_cnt_df = pd.DataFrame(vocab_emo_cnt_dict).T
vocab_emo_cnt_df['total'] = vocab_emo_cnt_df.sum(axis=1)
# ๋จ์ด๋ณ ์ต๊ณ ๊ฐ์ ๋ฐ ๊ฐ์ ๊ฐ์
all_result=vocab_emo_cnt_df.sort_values(by = 'total', ascending = False)
# ๋จ์ด๋ณ ์ต๊ณ ๊ฐ์ ๋ฐ ๊ฐ์ ๊ฐ์ , ํ์ฉ์ฌ ํฌํจ ์
adj_result=vocab_emo_cnt_df.sort_values(by = 'total', ascending = False)
# ๋ช
์ฌ๋ง ์ฌ์ฉ ์
noun_result=vocab_emo_cnt_df[vocab_emo_cnt_df.index.get_level_values(0) == 'noun'].sort_values(by = 'total', ascending = False)
final_file_name = f"essay_all_vocab_result.csv"
adj_file_name = f"essay_adj_vocab_result.csv"
noun_file_name = f"essay_noun_vocab_result.csv"
os.makedirs(f'./result/{file_made_dt_str}/', exist_ok = True)
final_result.to_csv(f"./result/{file_made_dt_str}/essay_all_vocab_result.csv", index = False)
adj_result.to_csv(f"./result/{file_made_dt_str}/essay_adj_vocab_result.csv", index = False)
noun_result.to_csv(f"./result/{file_made_dt_str}/essay_noun_vocab_result.csv", index = False)
return final_result, adj_result, noun_result, essay_summary, file_made_dt_str
from transformers import pipeline
model_name = 'AlexKay/xlm-roberta-large-qa-multilingual-finedtuned-ru'
question_answerer = pipeline("question-answering", model=model_name)
class BertClassifier(nn.Module):
def __init__(self, dropout = 0.3):
super(BertClassifier, self).__init__()
self.bert= BertModel.from_pretrained('bert-base-multilingual-cased')
self.dropout = nn.Dropout(dropout)
self.linear = nn.Linear(768, 6)
self.relu = nn.ReLU()
def forward(self, input_id, mask):
_, pooled_output = self.bert(input_ids = input_id, attention_mask = mask, return_dict = False)
dropout_output = self.dropout(pooled_output)
linear_output = self.linear(dropout_output)
final_layer= self.relu(linear_output)
return final_layer
def all_process(origin_essay):
essay_sent =split_essay_to_sentence(origin_essay)
row_dict = {}
for row in tqdm(essay_sent):
question = 'what is the feeling?'
answer = question_answerer(question=question, context=row)
row_dict[row] = answer
emo2idx, idx2emo = get_sent_labeldata()
tokenizer, cls_model = load_model()
final_result, file_name_dt = get_word_emotion_pair(cls_model, essay_sent)
all_result, adj_result, noun_result, essay_summary, file_made_dt_str = get_essay_base_analysis(file_name_dt)
summary_result = pd.concat([adj_result, noun_result]).fillna(0).sort_values(by = 'total', ascending = False).fillna(0).reset_index()[:30]
with open(f'./result/{file_name_dt}/summary.json','w') as f:
json.dump( essay_summary.to_json(),f)
with open(f'./result/{file_made_dt_str}/all_result.json','w') as f:
json.dump( all_result.to_json(),f)
with open(f'./result/{file_made_dt_str}/adj_result.json','w') as f:
json.dump( adj_result.to_json(),f)
with open(f'./result/{file_made_dt_str}/noun_result.json','w') as f:
json.dump( noun_result.to_json(),f)
return essay_summary
import gradio as gr
outputs = [gr.Dataframe(row_count = (6, "dynamic"),
col_count=(2, "dynamic"),
label="Essay Summary based on Words")
#headers=['type','word','์ฌํ', '๋ถ๋
ธ', '๊ธฐ์จ', '๋ถ์', '์์ฒ', '๋นํฉ', 'total'])
]
#row_count = (10, "dynamic"),
#col_count=(9, "dynamic"),
#label="Results",
#headers=['type','word','์ฌํ', '๋ถ๋
ธ', '๊ธฐ์จ', '๋ถ์', '์์ฒ', '๋นํฉ', 'total'])
#]
iface = gr.Interface(
fn=all_process,
inputs = gr.Textbox(lines=2, placeholder= '๋น์ ์ ๊ธ์ ๋ฃ์ด๋ณด์ธ์'),
outputs = outputs,
)
iface.launch(share =True) |