UI and functionality done

app.py

@@ -1,18 +1,12 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
-#from unsloth import FastLanguageModel
 from peft import AutoPeftModelForCausalLM
 from transformers import 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("HuggingFaceH4/zephyr-7b-beta")
-#client = InferenceClient("halme/id2223_lora_model")
-
 
-def respond(message, history: list[tuple[str, str]], system_message, max_tokens, temperature, top_p,):
+def respond(message, history: list[tuple[str, str]], system_message, max_tokens, temperature, min_p,):
     messages = [{"role": "system", "content": system_message}]
 
     for val in history:
@@ -23,32 +17,10 @@ def respond(message, history: list[tuple[str, str]], system_message, max_tokens,
 
     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 """
-
-    """     model, tokenizer = FastLanguageModel.from_pretrained(
-        model_name = "halme/id2223_lora_model", # YOUR MODEL YOU USED FOR TRAINING
-        max_seq_length = max_tokens,
-        dtype = None,
-        load_in_4bit = True,
-    ) """
-
     model = AutoPeftModelForCausalLM.from_pretrained(
         "halme/id2223_lora_model", # YOUR MODEL YOU USED FOR TRAINING
     )
-    tokenizer = AutoTokenizer.from_pretrained("halme/id2223_lora_model") 
-
-    #FastLanguageModel.for_inference(model) # Enable native 2x faster inference
-
-    """messages = [
-        {"role": "user", "content": "Continue the fibonnaci sequence: 1, 1, 2, 3, 5, 8,"},
-    ] """
-
+    tokenizer = AutoTokenizer.from_pretrained("halme/id2223_lora_model")
     inputs = tokenizer.apply_chat_template(
         messages,
         tokenize = True,
@@ -56,12 +28,12 @@ def respond(message, history: list[tuple[str, str]], system_message, max_tokens,
         return_tensors = "pt",
     )
 
-    from transformers import TextStreamer
-    text_streamer = TextStreamer(tokenizer, skip_prompt = True)
-
-    yield model.generate(input_ids = inputs, streamer = text_streamer, max_new_tokens = 128,
-                    use_cache = True, temperature = 1.5, min_p = 0.1)
+    output = model.generate(input_ids = inputs, max_new_tokens = max_tokens,
+                    use_cache = True, temperature = temperature, min_p = min_p)
+    
+    response = tokenizer.batch_decode(output, skip_special_tokens=True)[0]
 
+    yield response.split('assistant')[-1]
 
 """
 For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
@@ -70,14 +42,14 @@ 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=1, maximum=2048, value=2048, step=1, label="Max new tokens"),
+        gr.Slider(minimum=0.1, maximum=4.0, value=1.5, step=0.1, label="Temperature"),
         gr.Slider(
             minimum=0.1,
             maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
+            value=0.99,
+            step=0.01,
+            label="Min-p",
         ),
     ],
 )