Update app.py

app.py

@@ -1,64 +1,79 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
-
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-
-
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
-
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
-
-
-if __name__ == "__main__":
-    demo.launch()
+import numpy as np
+from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline
+
+def load_model(model_name):
+    tokenizer = AutoTokenizer.from_pretrained(model_name)
+    model = AutoModelForSequenceClassification.from_pretrained(model_name)
+    return pipeline('text-classification', model=model, tokenizer=tokenizer, truncation=True, max_length=512, top_k=4)
+
+classifier = load_model("ngocminhta/authscan-baseline")
+classifier2 = load_model("ngocminhta/authscan-baseline-machine")
+
+TEXT_CLASS_MAPPING_MACHINE = {
+    'LABEL_0': 'Gemini 1.5 Pro',
+    'LABEL_1': 'Gemini 2.0 Experimental',
+    'LABEL_2': 'GPT-4o Mini',
+    'LABEL_3': 'Llama 3.1 8B'
+}
+
+TEXT_CLASS_MAPPING = {
+    'LABEL_0': 'Human-Written',
+    'LABEL_1': 'Machine-Generated'
+}
+
+def update_language(language):
+    if language == 'java':
+        return gr.update(language='python')
+    return gr.update(language=language)
+
+def process_result_detection_tab(text, language): 
+    result = classifier(f"Language: {language}\n\n{text}")[0]
+    result_machine = classifier2(f"Language: {language}\n\n{text}")[0]
+
+    labels = [TEXT_CLASS_MAPPING[x['label']] for x in result]
+    labels_machine = [TEXT_CLASS_MAPPING_MACHINE[x['label']] for x in result_machine]
+
+    scores = list(np.array([x['score'] for x in result]))
+    scores_machine = list(np.array([x['score'] for x in result_machine]))
+    final_results = dict(zip(labels, scores))
+    if max(final_results, key=final_results.get) == 'Machine-Generated':
+        final_results_machine = dict(zip(labels_machine, scores_machine))
+    else:
+        final_results_machine = None
+    return final_results, final_results_machine
+
+def clear_detection_tab():
+    return "", gr.update(interactive=False)
+
+with gr.Blocks() as demo:
+    gr.Markdown("""<h1><center>LLM-DetectAIve</center></h1>""")
+
+    with gr.Row():
+        language = gr.Dropdown(
+            choices=["c", "cpp", "java", "python"],
+            label="Select Programming Language",
+            value="python"
+        )
+
+    with gr.Row():
+        input_text = gr.Code(
+          label="Enter code here",
+          language="python",
+          elem_id="code_input",
+        )
+
+    with gr.Row():
+        check_button = gr.Button("Check Origin", variant="primary")
+        clear_button = gr.Button("Clear", variant="stop")
+
+    out = gr.Label(label='Result')
+    out_machine = gr.Label(label='Detailed Information')
+    # When language is changed, update the code component's language
+    language.change(update_language, inputs=language, outputs=input_text)
+
+    check_button.click(process_result_detection_tab, inputs=[input_text, language], outputs=[out, out_machine])
+    # out_machine.change(lambda x: gr.update(visible=True) if out_machine else gr.update(visible=False), inputs=out_machine, outputs=out_machine)
+    clear_button.click(clear_detection_tab, inputs=[], outputs=[input_text, check_button])
+
+demo.launch(debug=True)