Bot to read our local PDF and answer using Openai, Embeddings, Langchain & Streamlit

Chat with PDF  | Chatbot Bio-query use case

In visual code Studio, creat directory usecase3 and creat a file search.py

in termial install the below libraries first

pip install chatpdf

pip install faiss-cpu  

pip install langchain

Steps Load + Split+Store+Get+Generate Bot + use Bot

Down load any PDF say test.pdf from which you want to ask. Copy to your current directory

Try this code first for load and split pdf

Create a file names search.py

#load and split pdf

from langchain.document_loaders import PyPDFLoader

loader = PyPDFLoader("test.pdf")

pages_content = loader.load_and_split()

print(len(pages_content), pages_content)

You will get dictionary output with filename, meta data, page no etc.

Now type as below and Run

#load pdf and split

from langchain.document_loaders import PyPDFLoader

loader = PyPDFLoader("test.pdf")

pages_content = loader.load_and_split()

#print(len(pages_content), pages_content)

# #refer openai api embeddings

from langchain.embeddings.openai import OpenAIEmbeddings

from langchain_openai import OpenAIEmbeddings

from langchain.vectorstores import FAISS

embeddings = OpenAIEmbeddings()

db = FAISS.from_documents(pages_content, embeddings)

db.save_local("faiss_index")  # to save local copy

nwdb = FAISS.load_local("faiss_index", embeddings,

                        allow_dangerous_deserialization=True)

query = "are there any educational qualification"

docs  =  nwdb.similarity_search(query) # to create similarity index

# print(docs)

from langchain.chains import RetrievalQA

from langchain.chat_models import ChatOpenAI

llm = ChatOpenAI()  # creat chatbot

qa_chain = RetrievalQA.from_chain_type(llm, retriever=nwdb.as_retriever())

res =qa_chain({"query" : "are there any educational qualification"})

print(res)

To get the results for your query. Here educational qualification related since i have take my bio-dat as pdf test file.

Now add ask query module in the above code as shown below:

#load pdf and split

from langchain.document_loaders import PyPDFLoader

# loader = PyPDFLoader("test.pdf")

# pages_content = loader.load_and_split()

# print(len(pages_content), pages_content)

# #refer openai api embeddings

from langchain.embeddings.openai import OpenAIEmbeddings

from langchain_openai import OpenAIEmbeddings

from langchain.vectorstores import FAISS

embeddings = OpenAIEmbeddings()

# two below lines are commented to avoid to create vectorbase again

# db = FAISS.from_documents(pages_content, embeddings)

# db.save_local("faiss_index")  # to save local vectorbase copy

# you can see files faiss_index and faiss

nwdb = FAISS.load_local("faiss_index", embeddings,

                        allow_dangerous_deserialization=True)

query = "are there any educational qualification"

docs  =  nwdb.similarity_search(query) # to create similarity index

# print(docs) # Note see faiss_index directory with files

from langchain.chains import RetrievalQA

from langchain.chat_models import ChatOpenAI

llm = ChatOpenAI()  # creat chatbot

qa_chain = RetrievalQA.from_chain_type(llm, retriever=nwdb.as_retriever())

res =qa_chain({"query" : "are there any educational qualification"})

# print(res)

# you can see faiss_index directory under which 2 files index.faiss and index.pkl

#addition for chatbot after app.py

def ask(user_query):

     res = qa_chain({"query": user_query})

     return res["result"]

Now create a front end file named as app.py with code as shown below

import streamlit as st

from search import ask

with st.chat_message("assitant"):

    st.write("Hello! May I helpYou ?")

prompt = st.chat_input("Type question")

if prompt:

    st.chat_message("user").markdown(prompt)

    response = ask(prompt)

    st.chat_message("assistant").markdown(response)

pip install streamlit in the terminal window

Save and Run as streamlit run app.py  to get streamlit.app tab in the browser 

Pl. double click the image to see clearly. 

Now type in the text window Educational qualification in bullet points  

and then press play button. You may get the screen as shown below:

Wow ! we have done a chatbot answering questions from my test.pdf

This is the simplest way with openai, embeddings, langchain and streamlit we could develope our custom made chatbot. This is the beginning. You can change code in app.py by refereinng streamlit documentation  to change your frontent and enjoy!!!😊😊😊😊😊😊😊😊

Please use your PDF and change queries as per your PDF.🎈

AskCodi.ai for Student

What is askCodi.ai?

Intelligent virtual assistant powered by AI technology. Offers personalized learning experiences. Integrates seamlessly into educational environments. Paradigm shift in coding How askCodi.ai Helps Students and Teachers:

Personalized Learning:

Understands individual learning styles. Delivers personalized recommendations and resources. 24/7 Assistance: Provides round-the-clock support for students. Alleviates burden on teachers. Interactive Learning Experiences: Engages students with quizzes, simulations, and games. Motivates active participation and deepens understanding. Efficient Communication: Streamlines communication between teachers and students. Facilitates collaboration and assessment.

Use Case and Tutorial Resources:

Scenario: High school student preparing for Python Lab Exam. Features: Access to practice problems and solutions. Step-by-step explanations for challenging concepts. Interactive quizzes and simulations. Real-time assistance and collaboration. Tutorial Resources: Instructional videos, study guides, and learning materials. A typical Demo Session --------------------------- Hands on ----------------------------------------------------------------------------------- Login to app.askCodi.com using google account In Left side Apps Section, you can see Codi Chat, Codi Workbook, Translate Choose Codi Workbook Create work book by filling the asked details like Project, description etc Click create project Python Lan, Language, Python 3 to get the screen like this. UI is very user friendly as shown @ end. From the above screen type in first window : Binary search Binary Sort and then click Buttons, Generate to generate code Explain to get simple explanation in english Document to document Test for test code generation Now you can see all the sections immediately. So easy for Student as well as teacher. Wow! What a Immersive coding tech. is informative, colloborative, productive!. Please dive in to this askCodi to enjoy every bit of programming! Happy Coding ! 😂😂😂😂

Prompt Engineering Demo using Gemini on 14.03.2024 @CSE, AU

Demo140324 done exclusive for CSE Students. For any clarifications contact author.

A G E N D A

-------------------------------------------------------------------------------------------------------------------------

• Video Release  Avatar Base CSE Promotion by HOD

• Heygen Preperation Demo 

• III. Simple Prompts for Coding+

• IV. PE Parameters

------------------------------------------------------------------------------------------------------------------------

https://gemini.google.com

Prompts (fun way to Learn) | Multimodal!!!

1. Basic Example

Write code that asks the user for their name and say "Hello"

2. Turn Comments Into Code

"""

1. Create a list of tourist spots in chidambaram, tamilnadu, india

2. Create a list of ratings for these spots

3. Combine them to make a json object of 10 spots with their ratings.

"""

3. Complete Functions or Next Line

# function to multiply two numbers and add 75 to it

 

def divide(

4. MySQL Query Generation

"""

Table departments, columns = [DepartmentId, DepartmentName]

Table students, columns = [DepartmentId, StudentId, StudentName]

"""

Create a MySQL query for all students in the Computer Science department

5.  Inserting Data

CREATE TABLE departments (

  DepartmentId INT PRIMARY KEY,

  DepartmentName VARCHAR(50)

);

CREATE TABLE students (

  DepartmentId INT,

  StudentId INT PRIMARY KEY,

  StudentName VARCHAR(50),

  FOREIGN KEY (DepartmentId) REFERENCES departments(DepartmentId)

);

Given the database schema above, generate valid insert statements include 4 rows for each table.

6. Explain Code

SELECT students.StudentId, students.StudentName

FROM students

INNER JOIN departments

ON students.DepartmentId = departments.DepartmentId

WHERE departments.DepartmentName = 'Computer Science';

Explain the above SQL statement.

7. Debugging code

"""

SELECT students.StudentId, students.StudentName

FROM students

INNER JOIN 

ON students.DepartmentId = departments.DepartmentId

WHERE departments.DepartmentName = 'Computer Science';

"""

Debug the above statements 

8. Can you create a stunning profile website with html, css and javscript

9. Teach me De Morgan's laws in simple sentences

10. Can you define AI acting like a kid, grad student, researcher and expert in one statements respectively

11. Act like a cooking Expert. Can you recipe for Biryani

12. Can you prepare quantity of grocery list for the above recipe to serve family of 5 members?

13. Can you order the above items in amazon store

14. Can you list the name of Faculty in the web page @https://annamalaiuniversity.ac.in/E04_factmem.php?dc=E04

15. Summarize @Gmail 

16. "Translate the sentence 'Hello, how are you?' from English to Tamil."

Phrase Translation Prompt:

17. Create a google script create a Google 5 MCQ in the topic "Annamalai University"

19. Prove SIN A/ Cos A = TAN A

20. Can you sing Indian National Anthem?

----------------------------------------------------------------------------------

21.. Create a image puppies with flower basket

22. can you create a love song?

23. Write 10 tamil movies and the words with emojis in table

24. Some fun stuff

Crack a joke based on " Why Tomatoes is red?"

Tell me moral story in 5 lines

Tell me a story about thirsty crow in 5 lines

Act like a Poet. Tell me a story about thirsty crow in 5 lines

List down 5 simple tongue twisters

Give me Current News Digest today?

What do mean by 143

What do you mean by 420?

What do you mean by 100?

What do you mean by 80/20 Paradigm?

What do you mean by LoL?

Zero shot : [Prompt] Write you tech script for my tech review channel

One Shot:  Using this ex1 as refernce, then [prompt] ~ related to IPhone

Few Shot: Using this ex1, ex2, ex3  as refernce, then [prompt]

Write a 5 minute script on the latest Iphone camera specifications for my tech review channel. Start with a 10 sec hook and notate a photo for each main point

Arithmetic reasoning:

What is the diameter of Sun?

~ let us go step by step

What is a simple quadratic formula?

~ let us go step by step

Bonus Prompt for Teachers: 

I am a teacher. I have to set Question Paper for Model Test. The following are the topics in the syllabus """ Definition – Types of random variables – probability distribution function – probability density function – expectation and moments – moment generating functions – joint probability distribution – marginal probability distribution function – joint probability density function – marginal probability density function – conditional probability density function""". Help me set up Question Paper with the following format. Part A contains 5 questions from the above topics each with 2 marks. Part B contains 4 questions from the above topics each with 5 marks. Part C contains 1 question either or type from the above topics with 10 marks. Put the questions in table structure with Q.NO, Question, Marks, Bloom's Taxonomy Level, CO, PO, PSO . Please provide PART B and PART C with Numerical problem oriented questions for under grad students in civil engineering

------------------------------------------------------------------------------------

Sumup Notes: Summarizing, Classifying, Translating, Writing email, resume, blogs, websites, Roadmaps for particular career, Study Schedule, Class Notes, Interview prepeartion, MCQ  testing etc. 

Any time, Any where, Anyhow? Companion

Knowledge begins Perplexity.ai

 ----------------------------------------------------------------------------------

Brief  Discussion on PE Parameters for LLM fine Tuning

Temperature : Temperature controls the randomness of the model's output. Lower value is determinitic

Max Tokens: Max tokens define the maximum length of the generated output. 

Top-p (Nucleus) Sampling: Top-p sampling limits the vocabulary size considered during sampling, ensuring that the generated tokens are from the most probable candidates. (0<p<1)

Nucleus Sampling: Top-p sampling is also known as Nucleus sampling. In this approach, the model selects tokens from the "nucleus" of the probability distribution, where the cumulative probability mass is concentrated.

Watch out this blog posts for Fine Tuning LLM using Parameters...!!!

-------------------------------------------------------------------------------------

Subscribe you tube channel for some more enlightening videos

https://www.youtube.com/channel/UC0GE3BT7-aYU7VBAGS41S7g?sub_confirmation=1

Prompt For CODING Student

Hello Dear CSE friends!

My teacher asked me to write code for binary search in python using iterative and recursive methods. optimize the code, debug, test, improve the complexities. I don't have time. Please help me to prompts for the above tasks!

How i should go!. Google it ! or ChatGPT! or Co Pilot! or GEMINI! or Perplexity,ai! any AI tool?

Let me try free version of Chat GPT 3.5.

Copy the above text and paste in text box and press play button. In hus seconds it give the following response

Of course! Here are the prompts tailored to each task your teacher assigned:

1. **Writing Binary Search Code (Iterative and Recursive):**

   - Prompt: "Begin by implementing the binary search algorithm in Python using both iterative and recursive methods. Write clear and concise code that effectively searches for a target element within a sorted array. Compare and contrast the iterative and recursive approaches, considering factors such as readability, performance, and ease of implementation."

2. **Optimizing the Code:**

   - Prompt: "Explore optimization strategies to enhance the efficiency and performance of your binary search implementations. Consider techniques such as minimizing unnecessary comparisons, reducing memory overhead, and optimizing boundary conditions. Experiment with different optimization approaches and analyze their impact on runtime complexity and resource utilization."

3. **Debugging and Testing:**

   - Prompt: "Debugging is an essential part of the development process. Use systematic debugging techniques to identify and resolve errors in your binary search code. Develop comprehensive test cases to validate the correctness and robustness of your implementations across various input scenarios. Consider edge cases, boundary conditions, and extreme inputs to ensure the reliability of your code."

4. **Improving Complexities:**

   - Prompt: "Challenge yourself to improve the time and space complexities of your binary search algorithms. Explore advanced algorithms and data structures that offer superior performance characteristics, such as the binary search tree or interpolation search. Experiment with algorithmic enhancements and optimizations to achieve optimal complexity levels while maintaining correctness and reliability."

5. **Documentation and Presentation:**

   - Prompt: "Document your coding journey and findings in a comprehensive report or presentation. Provide clear explanations of your binary search implementations, optimization strategies, debugging process, and complexity improvements. Include code snippets, test cases, performance metrics, and insights gained from your exploration. Share your learnings with your teacher and peers to foster collaboration and knowledge exchange."

These prompts will help guide you through each stage of the binary search assignment, ensuring that you effectively manage your time and produce high-quality results. Good luck with your coding!

Let us try all these above prompts one by one

Prompt : Write code fo binary search using iterative and recursive in python

Prompt : Can you optimize the above codes

{Use (right - left) // 2 instead of (left + right) // 2 to prevent potential integer overflow when left and right are large.}

Prompt : Can you optimize the above codes

Prompt : Can you give test case for the above code

Prompt : Can you explain the code in detail

Prompt : Can you document the code.

Prompt : Can you compare the performance metrics for the above codes in short and table format

Prompt: Can you prepare power point 10 slides script for explaining above two codes to 10 year old kids.

{ 

Slide 1: Introduction

• Title: Let's Learn Binary Search!
• Welcome, everyone!
• Today, we're going to learn about a cool way to find things quickly in a list.

Slide 2: What's Binary Search?

• Binary search is like a game of "Guess the Number."
• Imagine you have a list of numbers in order, and you need to find one specific number.

Slide 3: How Does It Work?

• We start in the middle of the list and check if our number is there.
• If it's there, we found it! If not, we decide if it's higher or lower than the number we checked.
• Then we repeat the process with the half where the number might be.

Slide 4: Iterative Binary Search

• We have a list of numbers.
• We check the middle number first.
• If it's not the number we want, we decide if our number is higher or lower and look in the right half or left half.
• We keep doing this until we find our number or run out of numbers to check.

Slide 5: Recursive Binary Search

• Recursive means we call a function within itself.
• In our game, we keep playing "Guess the Number" with smaller and smaller lists until we find our number or run out of numbers to check.

Slide 6: Pros and Cons

• Iterative: Easy to understand, uses less memory.
• Recursive: Can be harder to understand, uses more memory.

Slide 7: Let's Play a Game!

• Imagine we have a list of numbers: [1, 3, 5, 7, 9, 11, 13, 15, 17, 19].
• We want to find the number 11.
• Let's play the binary search game together!

Slide 8: Iterative Binary Search Game

• We start in the middle (9) and see it's too low.
• Next, we check 15. Still too low!
• Then we check 17. Still too low!
• Finally, we check 19. Too high!
• We go back and check 15 again. We found it!

Slide 9: Recursive Binary Search Game

• We start in the middle (9) and see it's too low.
• We call the game again with the right half.
• Now, we start in the middle again (15). Too high!
• We call the game again with the left half.
• Now, we start in the middle (11). We found it!
• 
  
• Slide 10: Conclusion
• • Binary search is like a smart way to find things quickly in a list. We learned how to play the binary search game using both iterative and recursive methods!
  • Keep practicing, and you'll become a binary search master! Feel free to adjust the script according to your preferences and add visuals to make the presentation engaging for the kids!  }
  
  

• Happy coding with ChatGPT, Friends!