Update app.py

app.py

@@ -1,15 +1,15 @@
 import gradio as gr
-from transformers import TFAutoModelForSequenceClassification, AutoTokenizer
 import tensorflow as tf
+from transformers import AutoTokenizer
 import numpy as np
 
-# Load model dan tokenizer dari Hugging Face Hub
-model_name = "jeanetrixsiee/javo_analisis_sentiment"
+# Load tokenizer dari Hugging Face repo
+tokenizer = AutoTokenizer.from_pretrained("jeanetrixsiee/javo_analisis_sentiment")
 
-model = TFAutoModelForSequenceClassification.from_pretrained(model_name)
-tokenizer = AutoTokenizer.from_pretrained(model_name)
+# Load model dari file .h5 secara langsung (bukan pakai TFAutoModel)
+model = tf.keras.models.load_model("keras_model")  # folder ini HARUS ada di root Spaces
 
-# Label mapping sesuai label2id yang kamu pakai
+# Pastikan label2id cocok
 id2label = {
     0: 'Negative',
     1: 'Neutral',
@@ -19,22 +19,18 @@ id2label = {
 }
 
 def predict_sentiment(text):
-    # Tokenisasi input
     inputs = tokenizer(text, return_tensors="tf", truncation=True, padding=True)
-    outputs = model(inputs)
+    outputs = model(inputs["input_ids"])
 
-    # Ambil hasil prediksi
-    probs = tf.nn.softmax(outputs.logits, axis=1).numpy()[0]
+    probs = tf.nn.softmax(outputs, axis=1).numpy()[0]
     predicted_index = int(np.argmax(probs))
     predicted_label = id2label[predicted_index]
 
-    # Buat dictionary label:probabilitas
     prob_dict = {id2label[i]: round(float(probs[i]) * 100, 2) for i in range(len(probs))}
     sorted_probs = dict(sorted(prob_dict.items(), key=lambda x: x[1], reverse=True))
 
     return f"Prediksi Sentimen: {predicted_label}", sorted_probs
 
-# Gradio UI
 interface = gr.Interface(
     fn=predict_sentiment,
     inputs=gr.Textbox(label="Masukkan Komentar YouTube"),
@@ -43,8 +39,7 @@ interface = gr.Interface(
         gr.Label(label="Probabilitas Tiap Label (%)")
     ],
     title="🔍 Analisis Sentimen Komentar YouTube",
-    description="Model ini memprediksi sentimen dari komentar YouTube dalam 5 kategori:\nVery Negative, Negative, Neutral, Positive, Very Positive.",
-    theme="default"
+    description="Model ini memprediksi sentimen dari komentar YouTube dalam 5 kategori:\nVery Negative, Negative, Neutral, Positive, Very Positive."
 )
 
 interface.launch()