update from gr.ChatInterface to gr.Chatbot to enable chat_history to be actually passed through

app.py

@@ -86,61 +86,104 @@ def generate(
         outputs.append(text)
         yield "".join(outputs)
 
-
-chat_interface = gr.ChatInterface(
-    fn=generate,
-    additional_inputs=[
-        gr.Textbox(label="System prompt", lines=6),
-        gr.Slider(
-            label="Max new tokens",
-            minimum=1,
-            maximum=MAX_MAX_NEW_TOKENS,
-            step=1,
-            value=DEFAULT_MAX_NEW_TOKENS,
-        ),
-        gr.Slider(
-            label="Temperature",
-            minimum=0.1,
-            maximum=4.0,
-            step=0.1,
-            value=0.6,
-        ),
-        gr.Slider(
-            label="Top-p (nucleus sampling)",
-            minimum=0.05,
-            maximum=1.0,
-            step=0.05,
-            value=0.9,
-        ),
-        gr.Slider(
-            label="Top-k",
-            minimum=1,
-            maximum=1000,
-            step=1,
-            value=50,
-        ),
-        gr.Slider(
-            label="Repetition penalty",
-            minimum=1.0,
-            maximum=2.0,
-            step=0.05,
-            value=1.2,
-        ),
-    ],
-    stop_btn=None,
-    examples=[
-        ["Ninawezaje kupata usingizi haraka?"],
-        ["Bosi wangu anadhibiti sana, nifanye nini?"],
-        ["Je, ni vipindi gani muhimu katika historia vya kujua kuvihusu?"],
-        ["Ni kazi gani nzuri ikiwa ninataka kupata pesa lakini pia kufurahiya?"],
-        ["Nivae nini kwenye harusi?"],
-    ],
-)
+examples = [
+    ["Ninawezaje kupata usingizi haraka?"],
+    ["Bosi wangu anadhibiti sana, nifanye nini?"],
+    ["Je, ni vipindi gani muhimu katika historia vya kujua kuvihusu?"],
+    ["Ni kazi gani nzuri ikiwa ninataka kupata pesa lakini pia kufurahiya?"],
+    ["Nivae nini kwenye harusi?"],
+]
+
+# chat_interface = gr.ChatInterface(
+#     fn=generate,
+#     additional_inputs=[
+#         gr.Textbox(label="System prompt", lines=6),
+#         gr.Slider(
+#             label="Max new tokens",
+#             minimum=1,
+#             maximum=MAX_MAX_NEW_TOKENS,
+#             step=1,
+#             value=DEFAULT_MAX_NEW_TOKENS,
+#         ),
+#         gr.Slider(
+#             label="Temperature",
+#             minimum=0.1,
+#             maximum=4.0,
+#             step=0.1,
+#             value=0.6,
+#         ),
+#         gr.Slider(
+#             label="Top-p (nucleus sampling)",
+#             minimum=0.05,
+#             maximum=1.0,
+#             step=0.05,
+#             value=0.9,
+#         ),
+#         gr.Slider(
+#             label="Top-k",
+#             minimum=1,
+#             maximum=1000,
+#             step=1,
+#             value=50,
+#         ),
+#         gr.Slider(
+#             label="Repetition penalty",
+#             minimum=1.0,
+#             maximum=2.0,
+#             step=0.05,
+#             value=1.2,
+#         ),
+#     ],
+#     stop_btn=None,
+#     examples=[
+#         ["Ninawezaje kupata usingizi haraka?"],
+#         ["Bosi wangu anadhibiti sana, nifanye nini?"],
+#         ["Je, ni vipindi gani muhimu katika historia vya kujua kuvihusu?"],
+#         ["Ni kazi gani nzuri ikiwa ninataka kupata pesa lakini pia kufurahiya?"],
+#         ["Nivae nini kwenye harusi?"],
+#     ],
+# )
 
 with gr.Blocks(css="style.css") as demo:
     gr.Markdown(DESCRIPTION)
-    gr.DuplicateButton(value="Duplicate Space for private use", elem_id="duplicate-button")
-    chat_interface.render()
+    chatbot = gr.Chatbot()
+    msg = gr.Textbox(label="Enter your message")
+    submit_btn = gr.Button("Submit")
+    clear = gr.Button("Clear")
+
+    def user(user_message, history):
+        return "", history + [[user_message, None]]
+
+    def bot(history, max_new_tokens, temperature, top_p, top_k, repetition_penalty):
+        user_message = history[-1][0]
+        chat_history = [(msg[0], msg[1]) for msg in history[:-1]]
+        bot_message = ""
+        for response in generate(user_message, chat_history, max_new_tokens, temperature, top_p, top_k, repetition_penalty):
+            bot_message = response
+            history[-1][1] = bot_message
+            yield history
+
+    gr.Examples(examples=examples, inputs=[msg], label="Examples")
+
+    with gr.Accordion("Advanced Options", open=False):
+        max_new_tokens = gr.Slider(label="Max new tokens", minimum=1, maximum=MAX_MAX_NEW_TOKENS, step=1, value=DEFAULT_MAX_NEW_TOKENS)
+        temperature = gr.Slider(label="Temperature", minimum=0.1, maximum=4.0, step=0.1, value=0.6)
+        top_p = gr.Slider(label="Top-p (nucleus sampling)", minimum=0.05, maximum=1.0, step=0.05, value=0.9)
+        top_k = gr.Slider(label="Top-k", minimum=1, maximum=1000, step=1, value=50)
+        repetition_penalty = gr.Slider(label="Repetition penalty", minimum=1.0, maximum=2.0, step=0.05, value=1.2)
+
+    submit_btn.click(user, [msg, chatbot], [msg, chatbot], queue=False).then(
+        bot,
+        [chatbot, max_new_tokens, temperature, top_p, top_k, repetition_penalty],
+        chatbot,
+    )
+    msg.submit(user, [msg, chatbot], [msg, chatbot], queue=False).then(
+        bot,
+        [chatbot, max_new_tokens, temperature, top_p, top_k, repetition_penalty],
+        chatbot,
+    )
+    clear.click(lambda: None, None, chatbot, queue=False)
+    
     gr.Markdown(LICENSE)
 
 if __name__ == "__main__":