app.py

# import gradio as gr
import tensorflow as tf
import numpy as np
from keras.models import load_model
from tensorflow.keras.preprocessing.text import Tokenizer
import pickle
from tensorflow.keras.preprocessing.sequence import pad_sequences
import os
from pathlib import Path
import pandas as pd
import plotly.express as px
import keras
import unicodedata as ud

from underthesea import word_tokenize

from phoBERT import BERT_predict

# Load tokenizer
# fp = Path(__file__).with_name('tokenizer.pkl')
# with open(fp,mode="rb") as f:
#     tokenizer = pickle.load(f)

#Load LSTM
#fp = Path(__file__).with_name('lstm_model.h5')
LSTM_model = tf.keras.models.load_model('lstm_model.tf')

#Load GRU
#fp = Path(__file__).with_name('gru_model.h5')
GRU_model = tf.keras.models.load_model('gru_model.tf')


def tokenizer_pad(tokenizer,comment_text,max_length=200):
   
    comment_text = word_tokenize(comment_text, format="text")
    comment_text = [comment_text]
    tokenized_text = tokenizer.texts_to_sequences(comment_text)

    padded_sequences = pad_sequences(sequences=tokenized_text,maxlen=max_length,padding="post",truncating="post")

    return padded_sequences

def LSTM_predict(x):
    # x = tokenizer_pad(tokenizer=tokenizer,comment_text=x)

    
    pred_proba = LSTM_model.predict([x])[0]

    pred_proba = [round(i,2) for i in pred_proba]

    #print(pred_proba)

    return pred_proba

def GRU_predict(x):
    # x = tokenizer_pad(tokenizer=tokenizer,comment_text=x)

    
    pred_proba = GRU_model.predict([x])[0]

    pred_proba = [round(i,2) for i in pred_proba]

    #print(pred_proba)

    return pred_proba

def plot(result):
  label = ['độc hại', 'cực kì độc hại', 'tục tĩu', 'đe dọa', 'xúc phạm', 'thù ghét cá nhân']
  data = pd.DataFrame()
  data['Nhãn'] = label
  data['Điểm'] = result

  #print(data)

  p = px.bar(data, x='Nhãn', y='Điểm', color='Nhãn', range_y=[0, 1] )
  return p
  pass

def judge(x):

  label = ['độc hại', 'cực kì độc hại', 'tục tĩu', 'đe dọa', 'xúc phạm', 'thù ghét cá nhân']
  result = []
  judge_result = []

  x = ud.normalize('NFKC', x)
  x = word_tokenize(x, format="text")

  lstm_pred = LSTM_predict(x)
  gru_pred = GRU_predict(x)
  #bert_pred = BERT_predict(x)
  #print(result)
  
  return_result = 'Result'
  result_lstm = np.round(lstm_pred, 2)
  result_gru = np.round(gru_pred, 2)
  #result_bert = np.round(bert_pred, 2)

  for i in range(6):
    result.append((result_lstm[i]+result_gru[i])/2)
  
  return (result)

def judgePlus(x):

  label = ['độc hại', 'cực kì độc hại', 'tục tĩu', 'đe dọa', 'xúc phạm', 'thù ghét cá nhân']
  result = []
  judge_result = []

  x = ud.normalize('NFKC', x)
  x = word_tokenize(x, format="text")

  lstm_pred = LSTM_predict(x)
  gru_pred = GRU_predict(x)
  try:
    bert_pred = BERT_predict(x)
  except:
    bert_pred = np.average([lstm_pred, gru_pred], axis=0)
  
  
  return_result = 'Result'
  result_lstm = np.round(lstm_pred, 2)
  result_gru = np.round(gru_pred, 2)
  result_bert = np.round(bert_pred, 2)
  #result_bert = np.round(bert_pred, 2)
  
  for i in range(6):
    result.append((result_lstm[i]+result_gru[i]+result_bert[i])/3)
  
  return (result)