Removing system message and loading FlanT5 base

app.py

@@ -1,64 +1,53 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
 
-"""
-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("google/flan-t5-xxl")
+# Load Flan-T5-base model and tokenizer from Hugging Face
+model_name = "google/flan-t5-base"
+model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
+tokenizer = AutoTokenizer.from_pretrained(model_name)
 
 
 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,
+    # Prepare the input by concatenating the history into a dialogue format
+    input_text = ""
+    for user_msg, bot_msg in history:
+        input_text += f"User: {user_msg} Assistant: {bot_msg} "
+    input_text += f"User: {message}"
+
+    # Tokenize the input text
+    inputs = tokenizer(input_text, return_tensors="pt", truncation=True)
+
+    # Generate the response using Flan-T5-base
+    output_tokens = model.generate(
+        inputs["input_ids"],
+        max_length=max_tokens,
         temperature=temperature,
         top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+        do_sample=True,
+    )
 
-        response += token
-        yield response
+    # Decode and return the assistant's response
+    response = tokenizer.decode(output_tokens[0], skip_special_tokens=True)
+    yield response
 
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+# Define the Gradio interface
 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)",
-        ),
+        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"),
     ],
 )
 
-
 if __name__ == "__main__":
     demo.launch()
+
+