encode input

app.py

@@ -1,6 +1,7 @@
 from flask import Flask, request, jsonify
 from flask_cors import CORS
 import logging
+from pyvi import ViTokenizer
 
 # Cấu hình logging cho app
 logging.basicConfig(level=logging.INFO)
@@ -41,6 +42,7 @@ def predict():
     # Gọi hàm dự đoán từ model
     try:
         if sentiment_type == "3sentiment":
+            text_to_predict = ViTokenizer.tokenize(text_to_predict)
             sentiment, score = predict_sentiment_3sentiment(text_to_predict)
         elif sentiment_type == "5sentiment":
             sentiment, score = predict_sentiment_5sentiment(text_to_predict)

model.py

@@ -5,11 +5,12 @@ from transformers import AutoModelForSequenceClassification, AutoTokenizer
 import torch.nn.functional as F
 import logging
 
+
 # Cấu hình logging
 logging.basicConfig(level=logging.INFO)
 logger = logging.getLogger(__name__)
 
-model_name_3sentiment = "phongnt251199/phobert-sentiment-reviews-v5"
+model_name_3sentiment = "phongnt251199/phobert-sentiment-reviews-v6"
 model_name_5sentiment = "phongnt251199/phobert-sentiment-reviews-v4"
 device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
 logger.info(f"Device being used: {device}")

requirements.txt

@@ -5,3 +5,4 @@ openpyxl
 torch
 transformers
 gunicorn
+pyvi