[FIRST COMMIT]

app.py

@@ -0,0 +1,47 @@
+import streamlit as st
+import pickle
+import pandas as pd
+
+# Load model dan vectorizer
+@st.cache_resource
+def load_model(model_choice):
+    with open(f'models/{model_choice}_model.pkl', 'rb') as f:
+        model = pickle.load(f)
+    with open('models/tfidf_vectorizer.pkl', 'rb') as f:
+        vectorizer = pickle.load(f)
+    return model, vectorizer
+
+label_names = ['susu', 'kacang', 'telur', 'makanan_laut', 'gandum']
+
+# UI
+st.title("🚀 Multi-label Food Classification")
+st.write("Masukkan teks untuk memprediksi kemungkinan alergi makanan.")
+
+model_choice = st.selectbox(
+    "Pilih model:",
+    options=["KNN", "SVM", "RF"]
+)
+
+model, vectorizer = load_model(model_choice)
+
+user_input = st.text_area("🧾 Masukkan teks makanan:")
+
+threshold = st.slider("Threshold prediksi (default 0.5):", 0.0, 1.0, 0.5)
+
+if st.button("Prediksi"):
+    if user_input.strip() != "":
+        user_vector = vectorizer.transform([user_input])
+
+        if hasattr(model, "predict_proba"):
+            user_proba = model.predict_proba(user_vector)
+            user_proba = [p[0][1] for p in user_proba]  # probability class 1
+        else:
+            user_proba = model.predict(user_vector)[0]
+            user_proba = [float(val) for val in user_proba]
+
+        st.subheader(f"Hasil Prediksi ({model_choice}):")
+        for label, proba in zip(label_names, user_proba):
+            status = "✅ Ada" if proba >= threshold else "❌ Tidak Ada"
+            st.write(f"- **{label}**: {status} ({proba:.2f})")
+    else:
+        st.warning("Masukkan teks terlebih dahulu!")

requirements.txt

@@ -0,0 +1,5 @@
+streamlit
+scikit-learn
+pandas
+pickle5
+numpy