Passionate about Data Science and Natural Language Processing, with a soft spot for pets, and cycling bikes. Personal opinions and political views are my own.
It's officially been two weeks since ChatGPT4 Vision started rolling out.
My prediction was right, Multimodal AI is about to disrupt SO many sectors.
Here are 14 of the most shocking examples that prove it:
Stanford XCS224U: Natural Language Understanding, 2023
Excellent lectures on natural language understanding. The course covers a wide range of topics including domain adaptation for supervised sentiment, retrieval augmented in-context learning, NLP techniques/methods, etc...
Lecture videos: https://t.co/llYzdFR5MP
GitHub: https://t.co/5eDDCcMmwv
Website: https://t.co/PsOGr4onX6
We have almost reached 5000 registrations for the free MLOps course!
We'll cover:
🔸 Introduction to MLOps
🔸 Experiment tracking and model management
🔸 Orchestration and ML Pipelines
��� Model Deployment
🔸 Model Monitoring
🔸 Best Practices
👉🏼 https://t.co/U5chrFsu1O
The biggest mistake most data scientists make:
They don't use pipelines.
Pipelines instantly improve your data transformation process.
Three advantages:
• Cleaner code
• More robust
• Easier to deploy in production
Look at the attached code.
IPython-GPT - a Jupyter/IPython interface to ChatGPT
Very promising extension to chat with ChatGPT directly from your Jupyter Notebook or IPython shell.
https://t.co/XgwRdp3lZW
#DTails
⁉️ What it is, and what is not?
ℹ️ How it works?
🆕 Why was it created?
1. DTails is a TOOL.
DTails IS NOT a distribution or fork of any GNU/Linux distribution. I DO NOT distribute any .iso/.img .
You build it. 🤙
2. The Idea came when ...
🧵...
Generative AI is ushering in the era of disposable apps ✨
Say hello to "just-in-time" software development 🪄
What use cases are you most excited about? 🤔
I built a GPT-4 'Warren Buffett' financial analyst to 'chat' with and analyze multiple PDF files (~1000 pages) across @elonmusk's Tesla 10-k annual reports (2020-2022)
#gpt4#openai#investing#stocks#finance
GPT-4 is Powerful as it is now multimodal and is capable of understanding both text and image inputs🔥
Here is how you can access GPT-4 model with just a few lines of code and use it:👇
This might be the most eventful week AI has ever seen:
Monday:
-Stanford Alpaca 7B
Tuesday:
-GPT4
-Anthropic releases Claude
-Google's PaLM API
-AdeptAI raises $350M
-Google adds GenAI to workspaces
Wednesday:
-Pytorch 2.0
-MidjourneyV5
Thursday:
-Microsoft 365 Copilot
It took me under 10 minutes to use ChatGPT-4 to generate code that accesses the GPT-3 API to convert a webpage at a given URL into a BibTex entry I can cite. This is an annoying task in my writing that's bothered me for years. There are existing tools but they don't work well.
🚨 BREAKING
Microsoft just introduced Visual ChatGPT 🔥
It enables you to send, receive, edit images during chatting.
You can also ask questions about any images, wild...
This is a valid Dockerfile for a NodeJS application. It is also a pile of 💩!
We can improve:
- 🔒 Security
- 🏎️ Build speed
- 👁️ Clarity
Follow along as we go from 💩 to 🥇!
(code in alt text)
1/ Pools de liquidez (LPs). Dá pra ganhar dinheiro com elas?
A resposta é... complicada. 📚
Passei 2 anos trabalhando com market making, com experiência tanto em CEXs quanto em DEXs (AMMs). Já vivenciei o problema na prática, para que você não precise. 😎
Segue a thread 🧵