Update app.py

app.py

@@ -6,22 +6,23 @@ import sentencepiece
 from tokenization_yi import YiTokenizer
 
 
+from transformers import AutoModelForCausalLM, GPTQConfig, AutoTokenizer, AutoModelForCausalLM
+import torch
+import os
+import gradio as gr
+import sentencepiece
+from tokenization_yi import YiTokenizer
+
 os.environ['PYTORCH_CUDA_ALLOC_CONF'] = 'max_split_size_mb:50'
 model_id = "TheBloke/Yi-34B-200K-Llamafied-GPTQ"
 
-gptq_config = GPTQConfig( bits=4, exllama_config={"version": 2})
-tokenizer = YiTokenizer.from_pretrained("./")
-model = AutoModelForCausalLM.from_pretrained( model_id, device_map="auto", torch_dtype="auto", trust_remote_code=True, quantization_config=gptq_config)
+gptq_config = GPTQConfig(bits=4, exllama_config={"version": 2})
+tokenizer = YiTokenizer.from_pretrained("./") #self-tokenizer method
+model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", torch_dtype="auto", trust_remote_code=True, quantization_config=gptq_config)
 
-def run(message, chat_history, max_new_tokens=4056, temperature=3.5, top_p=0.9, top_k=800):
-    prompt = get_prompt(message, chat_history)
+def predict(message, max_new_tokens=4056, temperature=3.5, top_p=0.9, top_k=800):
+    prompt = message.strip()
     input_ids = tokenizer.encode(prompt, return_tensors='pt')
-    
-    print("Input IDs:", input_ids)  # Debug print
-    print("Input IDs shape:", input_ids.shape)  # Debug print
-    if input_ids.shape[1] == 0:
-        raise ValueError("The input is empty after tokenization.")
-
     input_ids = input_ids.to(model.device)
     response_ids = model.generate(
         input_ids,
@@ -30,206 +31,34 @@ def run(message, chat_history, max_new_tokens=4056, temperature=3.5, top_p=0.9,
         top_p=top_p,              
         top_k=top_k,              
         pad_token_id=tokenizer.eos_token_id,
-        do_sample=True            
-
+        do_sample=True
     )
-
     response = tokenizer.decode(response_ids[:, input_ids.shape[-1]:][0], skip_special_tokens=True)
     return response
 
-def get_prompt(message, chat_history):
-    texts = []
-
-    do_strip = False
-    for user_input, response in chat_history:
-        user_input = user_input.strip() if do_strip else user_input
-        do_strip = True
-        texts.append(f" {response.strip()} {user_input} ")
-    message = message.strip() if do_strip else message
-    texts.append(f"{message}")
-    return ''.join(texts)
-
 DESCRIPTION = """
 # 👋🏻Welcome to 🙋🏻‍♂️Tonic's🧑🏻‍🚀YI-200K🚀"
 You can use this Space to test out the current model [Tonic/YI](https://huggingface.co/01-ai/Yi-34B)
 You can also use 🧑🏻‍🚀YI-200K🚀 by cloning this space. 🧬🔬🔍 Simply click here: <a style="display:inline-block" href="https://huggingface.co/spaces/Tonic1/YiTonic?duplicate=true"><img src="https://img.shields.io/badge/-Duplicate%20Space-blue?labelColor=white&style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAAXNSR0IArs4c6QAAAP5JREFUOE+lk7FqAkEURY+ltunEgFXS2sZGIbXfEPdLlnxJyDdYB62sbbUKpLbVNhyYFzbrrA74YJlh9r079973psed0cvUD4A+4HoCjsA85X0Dfn/RBLBgBDxnQPfAEJgBY+A9gALA4tcbamSzS4xq4FOQAJgCDwV2CPKV8tZAJcAjMMkUe1vX+U+SMhfAJEHasQIWmXNN3abzDwHUrgcRGmYcgKe0bxrblHEB4E/pndMazNpSZGcsZdBlYJcEL9Afo75molJyM2FxmPgmgPqlWNLGfwZGG6UiyEvLzHYDmoPkDDiNm9JR9uboiONcBXrpY1qmgs21x1QwyZcpvxt9NS09PlsPAAAAAElFTkSuQmCC&logoWidth=14" alt="Duplicate Space"></a></h3> 
 Join us : 🌟TeamTonic🌟 is always making cool demos! Join our active builder's🛠️community on 👻Discord: [Discord](https://discord.gg/nXx5wbX9) On 🤗Huggingface: [TeamTonic](https://huggingface.co/TeamTonic) & [MultiTransformer](https://huggingface.co/MultiTransformer) On 🌐Github: [Polytonic](https://github.com/tonic-ai) & contribute to 🌟 [PolyGPT](https://github.com/tonic-ai/polygpt-alpha)
 """
-
-MAX_MAX_NEW_TOKENS = 4056
-DEFAULT_MAX_NEW_TOKENS = 1256
-MAX_INPUT_TOKEN_LENGTH = 120000
-
-def clear_and_save_textbox(message): return '', message
-
-def display_input(message, history=[]):
-    history.append((message, ''))
-    return history
-
-def delete_prev_fn(history=[]):
-    try:
-        message, _ = history.pop()
-    except IndexError:
-        message = ''
-    return history, message or ''
-
-def generate(message, history_with_input, max_new_tokens, temperature, top_p, top_k):
-    if int(max_new_tokens) > MAX_MAX_NEW_TOKENS:
-        raise ValueError
-
-    history = history_with_input[:-1]
-    response = run(message, history, max_new_tokens, temperature, top_p, top_k)
-    yield history + [(message, response)]
-
-
-def process_example(message):
-    generator = generate(message, [], 1024, 2.5, 0.95, 900)
-    for x in generator:
-        pass
-    return '', x
-
-def check_input_token_length(message, chat_history):
-    input_token_length = len(message) + len(chat_history)
-    if input_token_length > MAX_INPUT_TOKEN_LENGTH:
-        raise gr.Error(f"The accumulated input is too long ({input_token_length} > {MAX_INPUT_TOKEN_LENGTH}). Clear your chat history and try again.")
-
 with gr.Blocks(theme='ParityError/Anime') as demo:
     gr.Markdown(DESCRIPTION)
-
-
-    
     with gr.Group():
+        textbox = gr.Textbox(placeholder='Enter your message here', label='Your Message', lines=2)
+        submit_button = gr.Button('Submit', variant='primary')
         chatbot = gr.Chatbot(label='TonicYi-30B-200K')
-        with gr.Row():
-            textbox = gr.Textbox(
-                container=False,
-                show_label=False,
-                placeholder='As the dawn approached, they leant in and said',
-                scale=10
-            )
-            submit_button = gr.Button('Submit', variant='primary', scale=1, min_width=0)
-
-    with gr.Row():
-        retry_button = gr.Button('Retry', variant='secondary')
-        undo_button = gr.Button('Undo', variant='secondary')
-        clear_button = gr.Button('Clear', variant='secondary')
-
-    saved_input = gr.State()
 
     with gr.Accordion(label='Advanced options', open=False):
-#       system_prompt = gr.Textbox(label='System prompt', value=DEFAULT_SYSTEM_PROMPT, lines=5, interactive=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=0.1)
+        max_new_tokens = gr.Slider(label='Max New Tokens', minimum=1, maximum=2056, step=1, value=980)
+        temperature = gr.Slider(label='Temperature', minimum=0.1, maximum=4.0, step=0.1, value=1.2)
         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=10)
-
-    textbox.submit(
-        fn=clear_and_save_textbox,
-        inputs=textbox,
-        outputs=[textbox, saved_input],
-        api_name=False,
-        queue=False,
-    ).then(
-        fn=display_input,
-        inputs=[saved_input, chatbot],
-        outputs=chatbot,
-        api_name=False,
-        queue=False,
-    ).then(
-        fn=check_input_token_length,
-        inputs=[saved_input, chatbot],
-        api_name=False,
-        queue=False,
-    ).success(
-        fn=generate,
-        inputs=[
-            saved_input,
-            chatbot,
-            max_new_tokens,
-            temperature,
-            top_p,
-            top_k,
-        ],
-        outputs=chatbot,
-        api_name="Generate",
-    )
-
-    button_event_preprocess = submit_button.click(
-        fn=clear_and_save_textbox,
-        inputs=textbox,
-        outputs=[textbox, saved_input],
-        api_name=False,
-        queue=False,
-    ).then(
-        fn=display_input,
-        inputs=[saved_input, chatbot],
-        outputs=chatbot,
-        api_name=False,
-        queue=False,
-    ).then(
-        fn=check_input_token_length,
-        inputs=[saved_input, chatbot],
-        api_name=False,
-        queue=False,
-    ).success(
-        fn=generate,
-        inputs=[
-            saved_input,
-            chatbot,
-            max_new_tokens,
-            temperature,
-            top_p,
-            top_k,
-        ],
-        outputs=chatbot,
-        api_name="Cgenerate",
-    )
-
-    retry_button.click(
-        fn=delete_prev_fn,
-        inputs=chatbot,
-        outputs=[chatbot, saved_input],
-        api_name=False,
-        queue=False,
-    ).then(
-        fn=display_input,
-        inputs=[saved_input, chatbot],
-        outputs=chatbot,
-        api_name=False,
-        queue=False,
-    ).then(
-        fn=generate,
-        inputs=[
-            saved_input,
-            chatbot,
-            max_new_tokens,
-            temperature,
-            top_p,
-            top_k,
-        ],
-        outputs=chatbot,
-        api_name=False,
-    )
-
-    undo_button.click(
-        fn=delete_prev_fn,
-        inputs=chatbot,
-        outputs=[chatbot, saved_input],
-        api_name=False,
-        queue=False,
-    ).then(
-        fn=lambda x: x,
-        inputs=[saved_input],
-        outputs=textbox,
-        api_name=False,
-        queue=False,
-    )
+        top_k = gr.Slider(label='Top-K', minimum=1, maximum=1000, step=1, value=900)
 
-    clear_button.click(
-        fn=lambda: ([], ''),
-        outputs=[chatbot, saved_input],
-        queue=False,
-        api_name=False,
+    submit_button.click(
+        fn=predict,
+        inputs=[textbox, max_new_tokens, temperature, top_p, top_k],
+        outputs=chatbot
     )
 
-demo.queue().launch(show_api=True)
+demo.launch(concurrency_limit=5)