add bot

Dockerfile

@@ -0,0 +1,14 @@
+# read the doc: https://huggingface.co/docs/hub/spaces-sdks-docker
+# you will also find guides on how best to write your Dockerfile
+
+FROM python:3.9
+
+WORKDIR /code
+
+COPY ./requirements.txt /code/requirements.txt
+
+RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
+
+COPY . .
+
+CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860"]

README.md

@@ -1,6 +1,6 @@
 ---
-title: Oci Ds 33b
-emoji: 💻
+title: OpenCodeInterpreter DS 33B
+emoji: 🧠
 colorFrom: indigo
 colorTo: gray
 sdk: docker
@@ -8,4 +8,4 @@ pinned: false
 license: mit
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+Check out the configuration reference at <https://huggingface.co/docs/hub/spaces-config-reference>

app.py

@@ -0,0 +1,104 @@
+import gradio as gr
+import torch
+from threading import Thread
+from typing import Iterator
+from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
+
+MAX_MAX_NEW_TOKENS = 1024
+MAX_INPUT_TOKEN_LENGTH = 2048
+
+# base_model_name = "m-a-p/OpenCodeInterpreter-DS-6.7B"
+base_model_name = "m-a-p/OpenCodeInterpreter-DS-33B"
+model = AutoModelForCausalLM.from_pretrained(base_model_name, torch_dtype=torch.bfloat16, device_map="cpu")
+
+tokenizer = AutoTokenizer.from_pretrained(base_model_name)
+
+def format_prompt(message, history):
+    system_prompt = "You are OpenCodeInterpreter, you are an expert programmer that helps to write code based on the user request, with concise explanations."
+    prompt = []
+    prompt.append({"role": "system", "content": system_prompt})
+    for user_prompt, bot_response in history:
+        prompt.extend([{"role": "user", "content": user_prompt}, {"role": "assistant", "content": bot_response}])
+    prompt.append({"role": "user", "content": message})
+    return prompt
+
+def generate(prompt: str, history: list[tuple[str, str]], max_new_tokens: int = 1024, temperature: float = 0.3,
+    top_p: float = 0.9, top_k: int = 50, repetition_penalty: float = 1 ) -> Iterator[str]:
+
+    temperature = float(temperature)
+    if temperature < 1e-2:
+        temperature = 1e-2
+    
+    formatted_prompt = []
+    formatted_prompt = format_prompt(prompt, history)
+    
+    input_ids = tokenizer.apply_chat_template(formatted_prompt, return_tensors="pt", add_generation_prompt=True)
+    if input_ids.shape[1] > MAX_INPUT_TOKEN_LENGTH:
+        input_ids = input_ids[:, -MAX_INPUT_TOKEN_LENGTH:]
+    input_ids = input_ids.to(model.device)
+
+    streamer = TextIteratorStreamer(tokenizer, timeout=15.0, skip_prompt=True, skip_special_tokens=True)
+    generation_kwargs    = dict({"input_ids": input_ids}, streamer=streamer, max_new_tokens=max_new_tokens, do_sample=False, top_p=top_p, top_k=top_k, 
+                              temperature=temperature, num_beams=1, repetition_penalty=repetition_penalty, eos_token_id=tokenizer.eos_token_id)
+
+    t = Thread(target=model.generate, kwargs=generation_kwargs  )
+    t.start()
+   
+    outputs = []
+    for chunk in streamer:
+        outputs.append(chunk)
+        yield "".join(outputs).replace("<|EOT|>","")
+        
+
+oci_chatbot = gr.Chatbot(layout="bubble", avatar_images=["user.png", "bot.png"], bubble_full_width=False, show_label=False, show_copy_button=True, likeable=True,)
+
+additional_inputs =  additional_inputs=[
+        gr.Slider(
+            label="Max new tokens",
+            minimum=1,
+            maximum=MAX_MAX_NEW_TOKENS,
+            step=1,
+            value=512,
+        ),
+        gr.Slider(
+            label="Temperature",
+            minimum=0,
+            maximum=1.0,
+            step=0.1,
+            value=0.3,
+        ),
+        gr.Slider(
+            label="Top-p",
+            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,
+        )]
+
+iface  = gr.ChatInterface(fn=generate,
+                        chatbot=oci_chatbot,
+                        additional_inputs=additional_inputs,
+                        description=" Running on CPU. The response may be slow for cpu environments. 🙏🏻",
+                        retry_btn=None,
+                        undo_btn=None
+                       )
+
+with gr.Blocks() as main:
+    gr.HTML("<center><h1>Tomoniai's Chat with OpenCodeInterpreter</h1></center>")
+    iface.render()
+    
+main.queue(max_size=10).launch(show_api=False)

requirements.txt

@@ -0,0 +1,9 @@
+accelerate==0.23.0
+bitsandbytes==0.41.1
+gradio==3.48.0
+protobuf==3.20.3
+scipy==1.11.2
+sentencepiece==0.1.99
+spaces==0.16.1
+torch==2.0.0
+transformers==4.36.2