Recent content on Binal Patel
The code for this post is here. I’m continuously impressed with GPT Vision the more I use it. I’ve recently been working on a side project that involves parsing PDFs with embedded tables/images/flow charts for a RAG (retrieval augmented generation) chatbot, and found the most painful part to be dealing with parsing and structuring the PDFs. There’s a lot of knobs to mess with…
I’ve been fascinated by this paper: Is ChatGPT Good at Search? Investigating Large Language Models as Re-Ranking Agent and have been trying out the ideas with success in a few personal projects. They essentially found that GPT4 is excellent at ranking things. Given a set of items (candidates) and some query it can rank these items very well (as well as or better than the current SOTA models…
The code for this post is here. The second experiment I’ve tried is GPT3 to extract and structure data and have been pretty impressed. The below example took me about an hour to setup, most of it just being iterating on the prompts I’m using as directions to the model. An additional bonus I wasn’t expecting - this also turned out to be a decent recipe generator. If I input in…
The code for this post is here. I’ve been playing around with GPT3 (and ChatGPT) a lot and have been wanting to start posting again, so I decided to start posting little experiments I do. These won’t be polished but hopefully interesting at least. Here’s the first - a rough pass on how to enable GPT3 to retain memory of past encounters/conversations (as opposed to having cram…
2020 Update A lot has changed since I first wrote this post - as of today I’d recommend you use VSCode Remote Containers Extension instead of this. PyCharm has similar functionality as well. What I normally do now - start up a remote container on a development server running JupyerLab (still using Conda for the Python environment), and attach to it. These docs may be useful is you want to do…
Soon after I started as a data scientist at an early stage startup I was tasked with helping productionalize and deploy analytical models as we ramped up more and more clients. The first few deploys were rushed, but relatively easy - just setup the root environment with all the dependencies, and have a simple bash script call the code on a schedule. Then as time went on things started to get…
We’ll walk through how to deal with nested data using Pandas (for example - a JSON string column), transforming that data into a tabular format that’s easier to deal with and analyze.
I am a full stack machine learning engineer and data scientist with several years of experience in doing everything necessary to get machine learning systems deployed and usable in the real world. I’m currently a Data Scientist at Sequoia Capital Global Equities, working on gathering and applying data to make informed investment decisions. My prior roles include applied NLP, CV, and…
Email me at binalkp91@gmail.com if you’d like to reach me. I’m also available at Twitter and LinkedIn, you can find the links to those on the top right of the page.