[Day 86] Made a youtube video - Chat with your PDF for free in colab using huggingface, mongodb, llama_index, langchain

 Hello :)
Today is Day 86!


A quick summary of today:
  • Coded, planned, recorded and posted a video tutorial making a chat with your pdf rag system for free


Well, after waking up today, I definitely did not expect to plan, execute and upload an almost 1hr tutorial on youtube. 

I was looking around chat with your PDF videos, to see what I can improve in my pdf_rag_from_scratch but I saw that most of the videos require an OpenAI api key, and I did not like that, given the availability of so many free resources and models. 

And I found this great resource from huggingface - Building A RAG System with Gemma, MongoDB and Open Source Model. Instead of a pdf, they were using some dataframe for films, so I decided to improve upon that, and make the code preprocess a pdf, embed it, upload to mongodb, load gemma, create a prompt and chat with the pdf (kind of a combination of the tutorial + my pdf_rag_from_scratch). 

The code itself is not that complicated, but I wanted to write it once/twice to make sure when I write live in the video recording, I do not have problems. So the whole process from idea to published video maybe took me 8 hours, mind that I had to find an app to edit the video (the editing was not much, but the app's video processing time was long because I wanted it in 1080p). 

Anyway ~ below I will provide an overall summary of the code

1. Download libraries

2. Preprocess PDF

2.1 Load PDF with llama-index

2.2 Chunk PDF text using langchain
2.3 Embed chunks
3. Set up mongodb
An important part is to set up an atlas vector search

3.1 Connect to the db

3.2 Delete existing (if any), and insert data
4. Find relevant texts
4.1 Perform vector search in db + get context

5. Load gemma using huggingface


6. Prompt engineering + talk with your PDF

I used similar base_prompt with the pdf_rag_from_scratch

Query: Do you pay or charge interest?

Answer: Yes, the Core Banking Agreement states that interest is paid and charged on a daily basis, and the interest rate applicable to your account(s) is stated in the Product & Services Terms & Conditions or, if no such terms are provided, on the website.


The results are not perfect, but is a good starting point for fine-tuning. 


That is all for today!

See you tomorrow :)

Popular posts from this blog

[Day 198] Transactions Data Streaming Pipeline Porject [v1 completed]

Image
 Hello :) Today is Day 198! A quick summary of today: data streaming pipeline project [v1 done] Here is a link to the project's repo. Well ... I did not know I can do it in a day (~14 hours) after yesterday's issues but here we are. Turns out in order to insert the full (~70 variables with nested/list structure), I need the proper pyspark schema. And yesterday I did not have that and that is why when I was reading data in the kafka producer I was getting NULL in the columns - my schema was wrong. Well today I not only fixed the schema for the 4 variables I had yesterday, but included *all* the variables that come from the Stripe API ~ 70 (for completeness).  When I run docker-compose, the data streams and is input into the postgres db (and is still running). Unfortunately, the free Stripe API for creating realistic transactions has a limit of 25, so every 3 seconds, 25 new transactions are sent to the db. It has been running half the day (since I got that set up) and as I am wr
Read more

[Day 107] Transforming natural language to charts

Image
 Hello :) Today is Day 107! A quick summary of today: created text2chart - transforming natural language to charts [ github ] [ webapp ] I am super sick, but in the around 3 hours of my fever not being high, I managed to study a bit. I was again thinking about creating some nlp project but was stuck on the idea of how to host a model. After looking around I found this article where they create a similar app called chat2vis  and they use codellama. I know codellama is an open source model that I can get from huggingface and I wondered, how did they host it. Turns out - they ask the user to input a huggingface api key, which then allows for querying and generating content. (AWESOME! this is applicable to some of the other PDF chat apps that I have, but I will look to make an app of them later).  My fever eased a bit, and I sat on my chair to do some work ^^ Here is a breakdown of the app. I used streamlit. Also, 2 default datasets during dev: Financial Statements of Major Companies(2009-
Read more

[Day 54] I became a backprop ninja! (woohoo)

Image
 Hello :) Today is Day 54! A quick summary of today: Read through cs231n's neural network lectures [1] [2] [3] Spent 8 hours going over Andrej Karpathy's backprop ninja tutorial After yesterday's confidence boost in doing backprop, I came in today with the intention of becoming of, as Andrej Karpathy calls it, a backprop ninja.  The cs231n NN lectures were the usual about how NN became to be, param initialization, activation functions, loss functions, regularization and optimizers. Nothing out of the ordinary, but still was good to reassure myself of that knowledge. But that was definitely not the main even of today. As I mentioned, I wanted to become (or at least attempt to become) a backprop ninja today. I have watched that particular tutorial twice, the 1st time I had no idea what is going on, the 2nd time I could follow, and now it was the 3rd time - and now, I wanted to participate and actually see the why and the how of the things we have to do, when we do manual back
Read more