src/streamlit_app.py

import streamlit as st
import pandas as pd
import re
import tensorflow as tf
import tensorflow_hub as tf_hub
from nltk.corpus import stopwords
from nltk.tokenize import word_tokenize
from tensorflow.keras.models import load_model
from Sastrawi.Stemmer.StemmerFactory import StemmerFactory
import nltk

# Use /tmp for NLTK data (writable in Hugging Face Spaces)
nltk_data_dir = "/tmp/nltk_data"
nltk.data.path.append(nltk_data_dir)

# Download the stopwords and punkt resources
nltk.download('stopwords', download_dir=nltk_data_dir)
nltk.download('punkt_tab', download_dir=nltk_data_dir)

# Load the trained model
model = tf.keras.models.load_model('src/model_final.keras',
                                   custom_objects={'KerasLayer': tf_hub.KerasLayer})
# Load stopwords
# Define Stopwords
stpwds_id = list(set(stopwords.words('indonesian')))
stpwds_id.append('oh')

# Define Stemming
stemmer = StemmerFactory().create_stemmer()

# Create A Function for Text Preprocessing

def text_preprocessing(text):
  # Case folding
  text = text.lower()

  # Mention removal
  text = re.sub("@[A-Za-z0-9_]+", " ", text)

  # Hashtags removal
  text = re.sub("#[A-Za-z0-9_]+", " ", text)

  # Newline removal (\n)
  text = re.sub(r"\\n", " ",text)

  # Whitespace removal
  text = text.strip()

  # URL removal
  text = re.sub(r"http\S+", " ", text)
  text = re.sub(r"www.\S+", " ", text)

  # Non-letter removal (such as emoticon, symbol (like μ, $, 兀), etc
  text = re.sub("[^A-Za-z\s']", " ", text)

  # Tokenization
  tokens = word_tokenize(text)

  # Stopwords removal
  tokens = [word for word in tokens if word not in stpwds_id]

  # Stemming
  tokens = [stemmer.stem(word) for word in tokens]

  # Combining Tokens
  text = ' '.join(tokens)

  return text
    
hub_layer = tf_hub.KerasLayer(
    "https://www.kaggle.com/models/google/nnlm/TensorFlow2/id-dim128-with-normalization/1",
    input_shape=[],
    dtype=tf.string,
    trainable=False
)
# Define the Streamlit interface
st.title('Sentiment Analysis App')

# Get user input
user_input = st.text_area("Enter the text for sentiment analysis:")

if st.button('Analyze'):
    if user_input:
        # Preprocess the input text
        processed_text = text_preprocessing(user_input)
        data_inf = hub_layer([processed_text]) 
        prediction = model.predict(data_inf)
        sentiment = "Positive" if prediction[0] > 0.5 else "Negative"

        # Display the result
        st.write(f"Sentiment: {sentiment}")
    else:
        st.write("Please enter some text.")