Basic demo

app.py

@@ -0,0 +1,27 @@
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+import gradio as gr
+
+# Load the fine-tuned model and tokenizer from Hugging Face
+model_name = "yeshavyas27/moondream-ft"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForSequenceClassification.from_pretrained(model_name, trust_remote_code=True)
+
+# Define the inference function
+def predict(input_text):
+    inputs = tokenizer(input_text, return_tensors="pt", truncation=True, padding=True)
+    outputs = model(**inputs)
+    predictions = outputs.logits.argmax(dim=-1)
+    return predictions.item()
+
+# Set up the Gradio interface
+iface = gr.Interface(
+    fn=predict,
+    inputs=gr.inputs.Textbox(lines=2, placeholder="Enter text here..."),
+    outputs="text",
+    title="Moondream Model Inference",
+    description="Enter some text to get predictions from the fine-tuned Moondream model."
+)
+
+# Run the app
+if __name__ == "__main__":
+    iface.launch()