Create app.py

app.py

@@ -0,0 +1,41 @@
+import gradio as gr
+from transformers import pipeline
+
+# The pipeline will automatically load the model and tokenizer
+# from the current directory where you've put the files.
+try:
+    classifier = pipeline("text-classification", model="./", tokenizer="./")
+
+    def classify_text(text):
+        """
+        Classifies a single piece of text and returns a human-readable prediction.
+        """
+        if not text:
+            return "Please enter some text to classify."
+            
+        result = classifier(text)[0]
+        label = "Hate Speech" if result['label'] == 'LABEL_1' else "Not Hate Speech"
+        score = result['score']
+        return f"Prediction: {label}\nConfidence: {score:.4f}"
+
+    # Create the Gradio interface
+    iface = gr.Interface(
+        fn=classify_text,
+        inputs=gr.Textbox(lines=5, placeholder="Enter a comment in English or Hindi..."),
+        outputs=gr.Textbox(label="Result"),
+        title="Multilingual Hate Speech Classifier",
+        description="A model to classify comments in Hindi and English."
+    )
+    
+    iface.launch()
+    
+except Exception as e:
+    # A simple error message box for the user
+    gr.Interface(
+        lambda x: f"An error occurred: {e}",
+        inputs="text",
+        outputs="text",
+        title="Error Loading Model",
+        description="There was an issue loading the model. Please check your files and dependencies."
+    ).launch()
+