Create app.py

app.py

@@ -0,0 +1,47 @@
+import gradio as gr
+from transformers import AutoTokenizer
+import onnxruntime as ort
+import numpy as np
+
+# Load tokenizer and ONNX quantized model
+tokenizer = AutoTokenizer.from_pretrained("onnx/")
+session = ort.InferenceSession("onnx/model_quantized.onnx")
+
+# Softmax function
+def softmax(x):
+    e_x = np.exp(x - np.max(x))
+    return e_x / e_x.sum()
+
+# Prediction function
+def classify_sentiment(text):
+    # Tokenize the input text
+    inputs = tokenizer(text, return_tensors="np")
+    #print(inputs)
+    # Run inference
+    outputs = session.run(None, {
+        "input_ids": inputs["input_ids"],
+        "attention_mask": inputs["attention_mask"]
+    })
+
+    # Process logits
+    logits = outputs[0][0]
+    probs = softmax(logits)
+    pred_class = int(np.argmax(probs))
+    
+    label_map = {0: "Negative", 1: "Positive"}
+    print(label_map[pred_class])
+    return label_map[pred_class]
+     
+
+# Gradio Interface
+interface = gr.Interface(
+    fn=classify_sentiment,
+    inputs=gr.Textbox(lines=2, placeholder="Enter text here..."),
+    outputs='label',
+    title="Sentiment Classifier",
+    description="Enter a sentence to classify its sentiment",
+)
+
+# Launch the app
+if __name__ == "__main__":
+    interface.launch(share=True)