Duplicate from zhtet/RegBotBeta

Co-authored-by: zwea htet <zhtet@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,34 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,7 @@
+venv
+data/__pycache__
+models/__pycache__
+.env
+__pycache__
+vectorStores
+.vscode

Dockerfile

@@ -0,0 +1,35 @@
+FROM python:3.9
+
+WORKDIR /docker
+
+ADD . /docker
+
+COPY requirements.txt requirements.txt
+RUN pip install --upgrade pip
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY . .
+
+# Set up a new user named "user" with user ID 1000
+RUN useradd -m -u 1000 user
+
+# Switch to the "user" user
+USER user
+
+# Set home to the user's home directory
+ENV HOME=/home/user \
+    PATH=/home/user/.local/bin:$PATH
+
+# Set the working directory to the user's home directory
+WORKDIR $HOME/app
+
+# Set the working directory to the user's home directory
+COPY --chown=user . $HOME/app
+
+EXPOSE 8501
+
+HEALTHCHECK CMD --fail http://localhost:8501/_stcore/health
+
+ENTRYPOINT ["streamlit", "run", "app.py", "--server.port=8501", "--server.address=0.0.0.0"]
+# CMD [ "streamlit" , "run", "app.py"]
+# CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]

README.md

@@ -0,0 +1,13 @@
+---
+title: RegBotBeta
+emoji: 😻
+colorFrom: green
+colorTo: red
+sdk: docker
+app_file: app.py
+app_port: 8501
+pinned: false
+duplicated_from: zhtet/RegBotBeta
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,87 @@
+# https://docs.streamlit.io/knowledge-base/tutorials/build-conversational-apps
+
+import os
+import time
+
+import openai
+import requests
+import streamlit as st
+
+from models import bloom
+from utils.util import *
+
+# from streamlit_chat import message
+
+
+st.title("Welcome to RegBotBeta")
+st.header("Powered by `LlamaIndex🦙` and `OpenAI API`")
+
+if "messages" not in st.session_state:
+    st.session_state.messages = []
+
+index = None
+
+api_key = st.text_input("Enter your OpenAI API key here:", type="password")
+
+if api_key:
+    resp = validate(api_key)
+    if "error" in resp.json():
+        st.info("Invalid Token! Try again.")
+    else:
+        st.info("Success")
+        os.environ["OPENAI_API_KEY"] = api_key
+        openai.api_key = api_key
+        with st.spinner("Initializing vector index ..."):
+            index = create_index(bloom)
+
+st.write("---")
+if index:
+    # Display chat messages from history on app rerun
+    for message in st.session_state.messages:
+        with st.chat_message(message["role"]):
+            st.markdown(message["content"])
+
+    if prompt := st.chat_input("Say something"):
+        # Display user message in chat message container
+        st.chat_message("user").markdown(prompt)
+
+        # Add user message to chat history
+        st.session_state.messages.append({"role": "user", "content": prompt})
+
+        with st.spinner("Processing your query..."):
+            bot_response = get_response(index, prompt)
+
+        print("bot: ", bot_response)
+
+        # Display assistant response in chat message container
+        with st.chat_message("assistant"):
+            message_placeholder = st.empty()
+            full_response = ""
+
+            # simulate the chatbot "thinking" before responding
+            # (or stream its response)
+            for chunk in bot_response.split():
+                full_response += chunk + " "
+                time.sleep(0.05)
+
+                # add a blinking cursor to simulate typing
+                message_placeholder.markdown(full_response + "▌")
+
+            message_placeholder.markdown(full_response)
+            # st.markdown(response)
+
+        # Add assistant response to chat history
+        st.session_state.messages.append(
+            {"role": "assistant", "content": full_response}
+        )
+
+        # Scroll to the bottom of the chat container
+        # st.markdown(
+        #     """
+        #     <script>
+        #     const chatContainer = document.getElementsByClassName("css-1n76uvr")[0];
+        #     chatContainer.scrollTop = chatContainer.scrollHeight;
+        #     </script>
+        #     """,
+        #     unsafe_allow_html=True,
+        # )

models/bloom.py

@@ -0,0 +1,107 @@
+import os
+import pickle
+from json import dumps, loads
+
+import numpy as np
+import openai
+import pandas as pd
+from dotenv import load_dotenv
+from huggingface_hub import HfFileSystem
+from llama_index import (
+    Document,
+    GPTVectorStoreIndex,
+    LLMPredictor,
+    PromptHelper,
+    ServiceContext,
+    StorageContext,
+    load_index_from_storage,
+)
+from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
+
+from utils.customLLM import CustomLLM
+
+load_dotenv()
+openai.api_key = os.getenv("OPENAI_API_KEY")
+fs = HfFileSystem()
+
+# get model
+# model_name = "bigscience/bloom-560m"
+# tokenizer = AutoTokenizer.from_pretrained(model_name)
+# model = AutoModelForCausalLM.from_pretrained(model_name, config='T5Config')
+
+# define prompt helper
+# set maximum input size
+context_window = 2048
+# set number of output tokens
+num_output = 525
+# set maximum chunk overlap
+chunk_overlap_ratio = 0.2
+prompt_helper = PromptHelper(context_window, num_output, chunk_overlap_ratio)
+
+# create a pipeline
+# pl = pipeline(
+#     model=model,
+#     tokenizer=tokenizer,
+#     task="text-generation",
+#     # device=0, # GPU device number
+#     # max_length=512,
+#     do_sample=True,
+#     top_p=0.95,
+#     top_k=50,
+#     temperature=0.7
+# )
+
+# define llm
+llm_predictor = LLMPredictor(llm=CustomLLM())
+service_context = ServiceContext.from_defaults(
+    llm_predictor=llm_predictor, prompt_helper=prompt_helper
+)
+
+
+def prepare_data(file_path: str):
+    df = pd.read_json(file_path)
+    df = df.replace(to_replace="", value=np.nan).dropna(axis=0)  # remove null values
+
+    parsed = loads(df.to_json(orient="records"))
+
+    documents = []
+    for item in parsed:
+        document = Document(
+            text=item["paragraphText"],
+            doc_id=item["_id"]["$oid"],
+            extra_info={
+                "chapter": item["chapter"],
+                "article": item["article"],
+                "title": item["title"],
+            },
+        )
+        documents.append(document)
+
+    return documents
+
+
+def initialize_index(index_name):
+    file_path = f"./vectorStores/{index_name}"
+    if os.path.exists(file_path):
+        # rebuild storage context
+        storage_context = StorageContext.from_defaults(persist_dir=file_path)
+
+        # local load index access
+        index = load_index_from_storage(storage_context)
+
+        # huggingface repo load access
+        # with fs.open(file_path, "r") as file:
+        #     index = pickle.loads(file.readlines())
+        return index
+    else:
+        documents = prepare_data(r"./assets/regItems.json")
+        index = GPTVectorStoreIndex.from_documents(
+            documents, service_context=service_context
+        )
+        # local write access
+        index.storage_context.persist(file_path)
+
+        # huggingface repo write access
+        # with fs.open(file_path, "w") as file:
+        #     file.write(pickle.dumps(index))
+        return index

requirements.txt

@@ -0,0 +1,12 @@
+llama_index
+torch
+transformers
+panda
+numpy
+langchain
+openai
+faiss-cpu
+python-dotenv
+streamlit>=1.24.0
+huggingface_hub
+xformers

utils/customLLM.py

@@ -0,0 +1,38 @@
+from typing import Any, List, Mapping, Optional
+
+from langchain.llms.base import LLM
+from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
+
+model_name = "bigscience/bloom-560m" 
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name, config='T5Config')
+
+pl = pipeline(
+    model=model,
+    tokenizer=tokenizer,
+    task="text-generation",
+    # device=0, # GPU device number
+    # max_length=512,
+    do_sample=True,
+    top_p=0.95,
+    top_k=50,
+    temperature=0.7
+)
+
+class CustomLLM(LLM):
+    pipeline = pl
+
+    def _call(self, prompt: str, stop: Optional[List[str]] = None) -> str:
+        prompt_length = len(prompt)
+        response = self.pipeline(prompt, max_new_tokens=525)[0]["generated_text"]
+
+        # only return newly generated tokens
+        return response[prompt_length:]
+
+    @property
+    def _identifying_params(self) -> Mapping[str, Any]:
+        return {"name_of_model": self.model_name}
+
+    @property
+    def _llm_type(self) -> str:
+        return "custom"

utils/util.py

@@ -0,0 +1,27 @@
+import requests
+
+
+def validate(token: str):
+    api_endpoint = "https://api.openai.com/v1/chat/completions"
+    api_key = token
+
+    headers = {"Content-Type": "application/json", "Authorization": f"Bearer {api_key}"}
+
+    messages = [{"role": "user", "content": "Say this is a test!"}]
+
+    data = {"model": "gpt-3.5-turbo", "messages": messages}
+
+    response = requests.post(api_endpoint, json=data, headers=headers)
+    return response
+
+
+def create_index(model):
+    index = model.initialize_index("bloomLlama")
+    return index
+
+
+def get_response(vector_index, query_str):
+    print("query_str: ", query_str)
+    query_engine = vector_index.as_query_engine()
+    response = query_engine.query(query_str)
+    return str(response)