Andrej Baranovskij (@andrejusb) / X

Andrej Baranovskij

5,451 posts

Andrej Baranovskij

@andrejusb

Sparrow Creator: Open-Source AI Doc Extraction 🚀 | ML/Oracle Dev | @katana_ml | Try: sparrow.katanaml.io | github.com/katanaml

Katana ML 👉

Joined March 2010

Pinned
Andrej Baranovskij
@andrejusb
Nov 11, 2021
We launched Katana ML katanaml.io in 2018 and now it is time to update the website, to explain where we are now and what we do with #MachineLearning, #MLOps, and #opensource 🚀🚀🚀
Katana
@katana_ml
Nov 11, 2021
We have a new website - katanaml.io It explains what we do with ML in a simple and straightforward way. It is featuring our open source product Skipper, we are using it to run #MLOps. #MachineLearning #MLOps
Andrej Baranovskij
@andrejusb
Oct 6, 2022
It took me ~1 hour to build this dashboard (data is dummy) layout in @streamlit. Using default Streamlit components. I think it would take 10 times longer with HTML/JS. Now I can focus on functionality 👍, not on div alignment 🤣 Code: github.com/katanaml/sparr…
Andrej Baranovskij
@andrejusb
Apr 2, 2024
Do you need LLM to output structured JSON? Then I recommend: 1. Hermes-2-Pro-Mistral-7B LLM (huggingface.co/NousResearch/H…) 2. @UnstructuredIO for data pre-processing and data preparation for LLM tasks 3. @langchain Pydantic parser to process LLM output (python.langchain.com/docs/modules/m…)
45K
Andrej Baranovskij
@andrejusb
May 6, 2024
You Don't Need RAG to Extract Invoice Data Complete video: youtu.be/watch?v=_GoGdF… Code: github.com/katanaml/sparr… Documents like invoices or receipts can be processed by LLM directly, without RAG. I explain how you can do this locally with @ollama and Instructor by @jxnlco.
00:00
52K
Andrej Baranovskij
@andrejusb
Jan 15, 2024
🚀 FastAPI and LlamaIndex RAG: Creating Efficient APIs 🔥 Complete video: youtu.be/watch?v=vntNI3… Code: github.com/katanaml/sparr… FastAPI works great with LlamaIndex RAG. In this video, I show how to build a POST endpoint to execute inference requests for LlamaIndex. RAG
00:00
94K
Andrej Baranovskij
@andrejusb
Apr 18, 2022
I'm impressed with the OCR quality by Mindee docTR. This is an open-source solution and can be easily integrated into document processing pipelines. Well done @MindeeAPI #Python 🚀 docTR GitHub: github.com/mindee/doctr Full video with OCR tests: youtu.be/3nYPIDCToes
00:00
Andrej Baranovskij
@andrejusb
Nov 30, 2023
Running Starling-7B LLM model on local CPU with @Ollama_ai and getting great results for invoice data extraction, even better than Zephyr, Mistral or Llama2. Prompt: retrieve gross worth value for each invoice item from the table. format response as following {\"gross_worth\":
79K
Andrej Baranovskij
@andrejusb
Sep 25, 2023
LLama 2 LLM for PDF Invoice Data Extraction I show how you can extract data from text PDF invoice using LLama2 LLM model and explain how you can improve data retrieval using carefully crafted prompts. Complete video: youtube.com/watch?v=WGNpdv…
00:00
39K
Andrej Baranovskij
@andrejusb
Jan 8, 2024
Transforming Invoice Data into JSON: Local LLM with LlamaIndex & Pydantic 🚀 Complete video: youtu.be/watch?v=VKeYaI… Code: github.com/katanaml/sparr… I explain how to get structured JSON output with LlamaIndex and dynamic Pydantic class. This helps to implement the use case of
00:00
148K
Andrej Baranovskij
@andrejusb
Oct 22, 2023
🔥 Invoice Data Processing with Llama2 13B LLM RAG on Local CPU 🚀 I explain production LLM RAG setup with Weaviate @philipvollet , Haystack for LLM API, and Llama.cpp to run Llama2 13b. Give it a try 🧑‍💻 Full video: youtu.be/XuvdgCuydsM Code: github.com/katanaml/llm-r…
00:00
47K
Andrej Baranovskij
@andrejusb
Apr 19, 2024
JSON structured output from invoice document with Llama3:8b Instruct. Running in Sparrow RAG with @llama_index, @weaviate_io and @ollama 🚀 @katana_ml
42K
Andrej Baranovskij
@andrejusb
Jun 10, 2024
Effective Table Data Extraction from PDF without LLM Sparrow Parse helps to read tabular data from PDFs, relying on various libraries, such as Unstructured or PyMuPDF4LLM. This allows us to avoid data hallucination errors often produced by LLMs when processing complex data
00:00
28K
Andrej Baranovskij
@andrejusb
Jun 17, 2024
Avoid LLM Hallucinations: Use Sparrow Parse for Tabular PDF Data, Instructor LLM for Forms LLMs tend to hallucinate and produce incorrect results for table data extraction. For this reason in Sparrow we are using Instructor by @jxnlco structured output for LLM to query form data
00:00
21K
Andrej Baranovskij
@andrejusb
Mar 25, 2024
LLM Structured Output for Function Calling with Ollama Complete video: youtu.be/watch?v=_-FrUR… Code: github.com/katanaml/sparr… I explain how function calling works with LLM. This is often confused concept, LLM doesn't call a function - LLM retuns JSON response with values to be
00:00
22K