updates to main app and requirement.txts to deploy to HF Spaces

app.py

@@ -1,147 +1,355 @@
-import io
-import random
-from typing import List, Tuple
+import os
+import re
+import sys
+import logging
+import nest_asyncio
+#import time
 
-import aiohttp
 import panel as pn
-from PIL import Image
-from transformers import CLIPModel, CLIPProcessor
+import tiktoken
+import chromadb
 
-pn.extension(design="bootstrap", sizing_mode="stretch_width")
+from llama_index.core import (
+    Settings, 
+    VectorStoreIndex, 
+    PromptTemplate, 
+    PromptHelper, 
+    StorageContext
+)
+from llama_index.core.text_splitter import SentenceSplitter
+from llama_index.llms.openai import OpenAI
+from llama_index.embeddings.huggingface import HuggingFaceEmbedding
+from llama_index.readers.web import SimpleWebPageReader
+from llama_index.vector_stores.chroma import ChromaVectorStore
 
-ICON_URLS = {
-    "brand-github": "https://github.com/holoviz/panel",
-    "brand-twitter": "https://twitter.com/Panel_Org",
-    "brand-linkedin": "https://www.linkedin.com/company/panel-org",
-    "message-circle": "https://discourse.holoviz.org/",
-    "brand-discord": "https://discord.gg/AXRHnJU6sP",
-}
+nest_asyncio.apply()
 
+FORMAT = "%(asctime)s | %(levelname)s | %(name)s | %(message)s"
 
-async def random_url(_):
-    pet = random.choice(["cat", "dog"])
-    api_url = f"https://api.the{pet}api.com/v1/images/search"
-    async with aiohttp.ClientSession() as session:
-        async with session.get(api_url) as resp:
-            return (await resp.json())[0]["url"]
+@pn.cache
+def get_logger(name, format_=FORMAT, level=logging.INFO):
+    logger = logging.getLogger(name)
 
+    logger.handlers.clear()
 
-@pn.cache
-def load_processor_model(
-    processor_name: str, model_name: str
-) -> Tuple[CLIPProcessor, CLIPModel]:
-    processor = CLIPProcessor.from_pretrained(processor_name)
-    model = CLIPModel.from_pretrained(model_name)
-    return processor, model
+    handler = logging.StreamHandler()
+    handler.setStream(sys.stdout)
+    formatter = logging.Formatter(format_)
+    handler.setFormatter(formatter)
+    logger.addHandler(handler)
+    logger.propagate = False
+
+    logger.setLevel(level)
+    logger.info("Logger successfully configured")
+    return logger
+
+####################
+# Global Constants #
+####################
+
+pn.extension("codeeditor", sizing_mode="stretch_width")
+
+TTL = 1800  # 30 minutes
+ACCENT = "#2EB872"
+THEME = pn.config.theme
+
+CHAT_GPT_LOGO = "https://upload.wikimedia.org/wikipedia/commons/thumb/0/04/ChatGPT_logo.svg/512px-ChatGPT_logo.svg.png"
+CHAT_GPT_URL = "https://chat.openai.com/"
+LLAMA_INDEX_LOGO = "https://asset.brandfetch.io/id6a4s3gXI/idncpUsO_z.jpeg"
+LLAMA_INDEX_URL = "https://www.llamaindex.ai/"
+
+LLM_VERSION = "gpt-3.5-turbo-1106"
+
+pn.chat.ChatMessage.default_avatars.update(
+    {
+        "assistant": CHAT_GPT_LOGO,
+        "user": "🦙",
+    }
+)
+pn.chat.ChatMessage.show_reaction_icons = False
+
+EXPLANATION = f"""
+## ScaleUp - (Level up your Python abilities)
+---
+
+**ScaleUp** is a powerful Python coding assistant app that leverages `OpenAI` and `LlamaIndex` to provide an interactive, 
+AI-powered learning experience.
+
+It acts as a virtual mentor, offering expert guidance, contextually relevant responses, and an integrated code editor for writing and testing Python code.
+
+### Key Features:
+
+- **Expert Python Guidance**: Get insightful and accurate answers to your Python queries.
+- **Interactive Code Editor**: Write and test your code, with suggestions and code snippets from the AI.
+- **Context-Aware Responses**: Responses are tailored based on your provided information and a comprehensive knowledge base.
+- **Streaming Responses**: Receive real-time, up-to-date responses as the AI generates them.
+
+## OpenAI GPT
+---
+We are using the OpenAI `{LLM_VERSION}` to power the coding assistant.
 
+## Getting Started
+---
 
-async def open_image_url(image_url: str) -> Image:
-    async with aiohttp.ClientSession() as session:
-        async with session.get(image_url) as resp:
-            return Image.open(io.BytesIO(await resp.read()))
+Ask your Python-related questions, share your code snippets, or request guidance on specific topics.
 
+The AI will respond with detailed explanations, code examples, and insightful suggestions to help you learn and improve your Python skills.
+"""
 
-def get_similarity_scores(class_items: List[str], image: Image) -> List[float]:
-    processor, model = load_processor_model(
-        "openai/clip-vit-base-patch32", "openai/clip-vit-base-patch32"
+SYSTEM_PROMPT = (
+    "You are an expert Python developer with years of experience writing Python code and teaching Python to other programmers. "
+    "You have vast experience mentoring people who are learning Python. "
+    "I want you to be my mentor while I learn Python myself. "
+    "Your goal is to provide insightful, accurate, and concise answers to questions in this domain. "
+    "When generating code, please explicitly state the sources you reference.\n\n"
+    "Here is some context related to the query:\n"
+    "-----------------------------------------\n"
+    "{context_str}\n"
+    "-----------------------------------------\n"
+    "Considering the above information, please respond to the following inquiry with detailed references to applicable principles, "
+    "libraries, design patterns, or debugging methodology where appropriate:\n\n"
+    "Question: {query_str}\n\n"
+    "Answer succinctly, and ensure your response is understandable to someone with extreme enthusiasm to learn Python programming."
+)
+
+# URL's for context with RAG Based Data 
+URLS = [
+    "https://thewhitetulip.gitbook.io/py",
+    "https://docs.python.org/3/tutorial/",
+    "https://awesomepython.org/",
+    "https://awesome-python.com/",
+]
+
+##########################################
+# Data Processing and handling functions #
+##########################################
+
+USER_CONTENT_FORMAT = """
+Request:
+{content}
+Code:
+```python
+{code}
+```
+""".strip()
+
+DEFAULT_CODE_EXAMPLE = """
+print("Hello World")
+""".strip()
+
+# Sample Python programming questions
+EXAMPLE_QUESTIONS = f"""
+## Python Programming Questions
+
+### Basic
+
+- Write a Python function to find the maximum of three numbers.
+- Write a Python program to reverse a string.
+- Write a Python program to check if a given number is prime or not.
+- Write a Python program to find the factorial of a number.
+- Write a Python program to check if a string is a palindrome or not.
+- Write a Python program to find the largest number in a list. 
+- Write a Python program to find the sum of all numbers in a list.
+- Write a Python program to find the second largest number in a list.
+- Write a Python program to remove duplicates from a list.
+- Write a Python program to implement a simple calculator.
+- Write a Python program to check if a string is a palindrome. 
+- Write a Python program to find the Fibonacci sequence up to a given number.
+- Write a Python program to Solve the Fizbuzz Algorithm in the most simple way you can think of ...
+
+### Advanced
+
+- Write a Python program to sort a list of dictionaries by a specific value.
+- Write a Python program to implement a binary search algorithm.
+- Write a Python program to implement a merge sort algorithm.
+- Write a Python program to implement a linked list data structure.
+- Write a Python program to implement a binary tree data structure.
+- Implement an LRU (Least Recently Used) Cache.
+- Write a function to check if a binary tree is balanced. 
+- Implement a stack using two queues. 
+- Write a function to calculate the factorial of a number recursively.
+- Implement a depth-first search (DFS) algorithm to traverse a graph.  
+
+"""
+
+def _powered_by():
+    """Returns a component describing the frameworks powering the chat ui."""
+    params = {"height": 40, "sizing_mode": "fixed", "margin": (0, 10)}
+    return pn.Column(
+        pn.pane.Markdown("### AI Powered By", margin=(10, 5, 10, 0)),
+        pn.Row(
+            pn.pane.Image(LLAMA_INDEX_LOGO, link_url=LLAMA_INDEX_URL, **params),
+            pn.pane.Image(CHAT_GPT_LOGO, link_url=CHAT_GPT_URL, **params),
+            align="center",
+        ),
     )
-    inputs = processor(
-        text=class_items,
-        images=[image],
-        return_tensors="pt",  # pytorch tensors
+
+llm = OpenAI(temperature=0.1, model=LLM_VERSION, max_tokens=512)
+embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-small-en-v1.5")
+text_splitter = SentenceSplitter(chunk_size=1024, chunk_overlap=20)
+
+prompt_helper = PromptHelper(
+    context_window=4096,
+    num_output=256,
+    chunk_overlap_ratio=0.1,
+    chunk_size_limit=None,
+)
+
+# Settings configuration
+Settings.llm = llm
+Settings.embed_model = embed_model
+Settings.tokenizer = tiktoken.encoding_for_model(LLM_VERSION).encode
+Settings.text_splitter = text_splitter
+Settings.prompt_helper = prompt_helper
+
+def load_data(data=URLS):
+    """
+    Initialize the Index
+    """
+    reader = SimpleWebPageReader(html_to_text=True)
+    documents = reader.load_data(data)
+
+    logging.info("index creating with `%d` documents", len(documents))
+    chroma_client = chromadb.EphemeralClient()
+    chroma_collection = chroma_client.get_or_create_collection("python-data")
+    vector_store = ChromaVectorStore(chroma_collection=chroma_collection)
+    storage_context = StorageContext.from_defaults(vector_store=vector_store)
+    index = VectorStoreIndex.from_documents(documents, storage_context=storage_context, embed_model=embed_model)
+
+    return index
+
+
+def initialize_query_engine(index):
+    """
+    Initialize Query Engine
+    """
+    # Custom Prompt Template
+    template = SYSTEM_PROMPT
+    qa_template = PromptTemplate(template)
+
+    # build query engine with custom template
+    query_engine = index.as_query_engine(text_qa_template=qa_template, similarity_top_k=3)
+
+    return query_engine
+
+
+def build_chat_engine(index):
+    """
+    Initialize Chat Engine
+    """
+    # Custom Prompt Template
+    template = SYSTEM_PROMPT
+    qa_template = PromptTemplate(template)
+
+    chat_engine = index.as_chat_engine(
+        chat_mode="context",
+        text_qa_template=qa_template,
+        verbose=True,
+        streaming=True
     )
-    outputs = model(**inputs)
-    logits_per_image = outputs.logits_per_image
-    class_likelihoods = logits_per_image.softmax(dim=1).detach().numpy()
-    return class_likelihoods[0]
+    return chat_engine
 
+############
+# Main App #
+############
 
-async def process_inputs(class_names: List[str], image_url: str):
+logger = get_logger(name="code_asst_example")
+
+index = load_data()
+
+# Custom Prompt Template
+template = SYSTEM_PROMPT
+qa_template = PromptTemplate(template)
+
+chat_engine = index.as_chat_engine(
+    chat_mode="context", 
+    text_qa_template=qa_template,
+    verbose = True,
+    streaming=True
+    )
+
+# Getting the API Key
+os.getenv('OPENAI_API_KEY')
+
+async def generate_response(
+        contents: str, 
+        user: str, 
+        instance: pn.chat.ChatInterface
+    ):
     """
-    High level function that takes in the user inputs and returns the
-    classification results as panel objects.
+    Docstring placeholder
     """
-    try:
-        main.disabled = True
-        if not image_url:
-            yield "##### ⚠️ Provide an image URL"
-            return
-    
-        yield "##### ⚙ Fetching image and running model..."
-        try:
-            pil_img = await open_image_url(image_url)
-            img = pn.pane.Image(pil_img, height=400, align="center")
-        except Exception as e:
-            yield f"##### 😔 Something went wrong, please try a different URL!"
-            return
-    
-        class_items = class_names.split(",")
-        class_likelihoods = get_similarity_scores(class_items, pil_img)
-    
-        # build the results column
-        results = pn.Column("##### 🎉 Here are the results!", img)
-    
-        for class_item, class_likelihood in zip(class_items, class_likelihoods):
-            row_label = pn.widgets.StaticText(
-                name=class_item.strip(), value=f"{class_likelihood:.2%}", align="center"
-            )
-            row_bar = pn.indicators.Progress(
-                value=int(class_likelihood * 100),
-                sizing_mode="stretch_width",
-                bar_color="secondary",
-                margin=(0, 10),
-                design=pn.theme.Material,
-            )
-            results.append(pn.Column(row_label, row_bar))
-        yield results
-    finally:
-        main.disabled = False
-
-
-# create widgets
-randomize_url = pn.widgets.Button(name="Randomize URL", align="end")
-
-image_url = pn.widgets.TextInput(
-    name="Image URL to classify",
-    value=pn.bind(random_url, randomize_url),
+    response = await chat_engine.astream_chat(contents)
+    text = ""
+    async for token in response.async_response_gen():
+        text += token
+        yield text
+
+    # extract code from LLM response
+    llm_code = re.findall(r"```python\n(.*)\n```", text, re.DOTALL)[0]
+    code_editor.value = llm_code
+
+
+#######################
+# Panel UI Components #
+#######################
+
+chat_interface = pn.chat.ChatInterface(
+    callback=generate_response,
+    show_send=True,
+    show_rerun=False,
+    show_undo=True,
+    show_clear=True,
+    show_button_name=True,
+    sizing_mode="stretch_both",
+    callback_exception="verbose"
+)
+
+chat_interface.send(
+    SYSTEM_PROMPT, 
+    user="System", 
+    respond=False
 )
-class_names = pn.widgets.TextInput(
-    name="Comma separated class names",
-    placeholder="Enter possible class names, e.g. cat, dog",
-    value="cat, dog, parrot",
+
+code_editor = pn.widgets.CodeEditor(
+    value=DEFAULT_CODE_EXAMPLE,
+    language="python",
+    sizing_mode="stretch_both",
+)
+
+# Create a layout for the widgets
+question_layout = pn.Column(
+    EXAMPLE_QUESTIONS,
+    sizing_mode="stretch_width"
 )
 
-input_widgets = pn.Column(
-    "##### 😊 Click randomize or paste a URL to start classifying!",
-    pn.Row(image_url, randomize_url),
-    class_names,
+# lay them out in tabs
+tabs_layout = pn.Tabs(
+    ("Code", code_editor),
+    ("Example Questions", question_layout),
+    sizing_mode = "stretch_both",
 )
 
-# add interactivity
-interactive_result = pn.panel(
-    pn.bind(process_inputs, image_url=image_url, class_names=class_names),
-    height=600,
+component = pn.Row(
+    chat_interface,
+    tabs_layout, 
+    sizing_mode="stretch_both"
 )
 
-# add footer
-footer_row = pn.Row(pn.Spacer(), align="center")
-for icon, url in ICON_URLS.items():
-    href_button = pn.widgets.Button(icon=icon, width=35, height=35)
-    href_button.js_on_click(code=f"window.open('{url}')")
-    footer_row.append(href_button)
-footer_row.append(pn.Spacer())
-
-# create dashboard
-main = pn.WidgetBox(
-    input_widgets,
-    interactive_result,
-    footer_row,
+# Serve UI Template
+template = pn.template.FastListTemplate(
+    title="ScaleUp Code Assistant 🐍",
+    sidebar=[
+            EXPLANATION,
+            _powered_by(), 
+            ],
+    main=[component],
+    main_layout=None,
+    accent=ACCENT,
 )
 
-title = "Panel Demo - Image Classification"
-pn.template.BootstrapTemplate(
-    title=title,
-    main=main,
-    main_max_width="min(50%, 698px)",
-    header_background="#F08080",
-).servable(title=title)
+template.servable()
+
+##################
+# End of the App #
+##################

requirements.txt

@@ -3,4 +3,12 @@ jupyter
 transformers
 numpy
 torch
-aiohttp
+aiohttp
+sentence-transformers
+tiktoken
+chromadb
+llama-index
+llama-index-llms-openai
+llama-index-readers-web
+llama-index-embeddings-instructor
+llama-index-vector-stores-chroma