NLPSentimentAnalysis / src /streamlit_app.py
EnYa32's picture
Update src/streamlit_app.py
964ecc7 verified
Raw History Blame Contribute Delete
2.55 kB
import re
from pathlib import Path
import joblib
import streamlit as st
# -------------------------
# Page config
# -------------------------
st.set_page_config(
page_title='Movie Review Sentiment (LinearSVC + TF-IDF)',
page_icon='🎬',
layout='centered'
)
st.title('🎬 Movie Review Sentiment Analysis')
st.write(
'Classify Rotten Tomatoes review phrases into 5 sentiment classes using a trained '
'LinearSVC + TF-IDF pipeline.'
)
# -------------------------
# Paths (HF friendly)
# -------------------------
BASE_DIR = Path(__file__).resolve().parent
MODEL_PATH = BASE_DIR / 'sentiment_pipeline.pkl'
LABELS_PATH = BASE_DIR / 'label_names.pkl'
# -------------------------
# Helpers
# -------------------------
def clean_text(text: str) -> str:
if not isinstance(text, str):
return ''
text = text.lower()
text = re.sub(r'<[^>]+>', ' ', text)
text = re.sub(r'\d+', ' ', text)
text = re.sub(r'[\n\r]', ' ', text)
text = re.sub(r"[^a-z\s']", ' ', text)
text = re.sub(r'\s+', ' ', text)
return text.strip()
@st.cache_resource
def load_model():
if not MODEL_PATH.exists():
raise FileNotFoundError(
f'Model not found: {MODEL_PATH.name}. Put it in the repo root (same folder as app.py).'
)
return joblib.load(MODEL_PATH)
@st.cache_resource
def load_labels():
if LABELS_PATH.exists():
return joblib.load(LABELS_PATH)
return {
0: 'negative',
1: 'somewhat negative',
2: 'neutral',
3: 'somewhat positive',
4: 'positive'
}
model = load_model()
label_names = load_labels()
# -------------------------
# UI
# -------------------------
st.subheader('Enter a phrase')
default_text = 'An intermittently pleasing but mostly routine effort.'
user_text = st.text_area('Movie review phrase', value=default_text, height=120)
col1, col2 = st.columns([1, 1])
with col1:
run = st.button('Predict', type='primary')
with col2:
st.write('')
if run:
cleaned = clean_text(user_text)
if cleaned.strip() == '':
st.warning('Please enter some text (not empty).')
else:
pred = int(model.predict([cleaned])[0])
label = label_names.get(pred, str(pred))
st.success(f'Prediction: **{pred} — {label}**')
with st.expander('Show cleaned text'):
st.code(cleaned)
st.markdown('---')
st.caption(
'Model: TF-IDF (1–2 grams) + LinearSVC | Labels: 0=negative, 1=somewhat negative, '
'2=neutral, 3=somewhat positive, 4=positive'
)