Add app.py file

app.py

@@ -0,0 +1,48 @@
+import gradio as gr
+import torch
+from transformers import AutoModelForSequenceClassification, AutoTokenizer
+
+# Load your fine-tuned model from the Hugging Face Hub
+model_name = "Wisaba/emotion_roberta_weighted"
+
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForSequenceClassification.from_pretrained(model_name)
+
+# Move to GPU if available
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+model.to(device)
+
+# Labels
+emotion_labels = ["sadness", "joy", "love", "anger", "fear", "surprise"]
+
+def classify_emotion(text):
+    # 1. Tokenize
+    inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=128)
+    inputs = {k: v.to(device) for k, v in inputs.items()}
+
+    # 2. Predict
+    with torch.no_grad():
+        outputs = model(**inputs)
+
+    # 3. Get Label
+    logits = outputs.logits
+    predicted_class_id = torch.argmax(logits, dim=-1).item()
+    return emotion_labels[predicted_class_id]
+
+# Define the Gradio Interface
+iface = gr.Interface(
+    fn=classify_emotion,
+    inputs=gr.Textbox(lines=2, placeholder="Type how you feel...", label="Text Input"),
+    outputs=gr.Textbox(label="Predicted Emotion"),
+    title="Emotion Analysis (RoBERTa)",
+    description="This model classifies text into 6 emotions: Sadness, Joy, Love, Anger, Fear, Surprise.",
+    examples=[
+        ["I am feeling so lonely and sad today."],
+        ["I'm incredibly excited about the new project!"],
+        ["Why did you do that? I'm so mad at you!"]
+    ]
+)
+
+# Launch
+if __name__ == "__main__":
+    iface.launch()