tabnet VS autogluon

There's architectures that operate on tabular data from a few years back now that uses CNN's and self attention. .look at tabnet for inspiration: https://github.com/dreamquark-ai/tabnet

I tried some other projects like [tabnet](https://github.com/dreamquark-ai/tabnet) they are also yielding *decent* results, but not more than 0.64.

Are you solving a tabular problem? If it is so, you can use TabNet to plot them. It also supports multiclass and multilabel.

🌐 Composed Image Retrieval 📎 Intro to Multimodal LLama 3.2 🛠️ Multi Agent Concierge 💻 RAG with Langchain Granite, Milvus 🫶 Download content ✅ Transformer Replacement? 🤖 vLLM for runing models 🌐 Amphion 📝 Autogluon 🚙 Notebook LLama like Google's Notebook LLM 🫶 Monocle2ai for tracing GenAI app code LFA&D Project 🤖 Bee Agent Framework ✅ LLama RFP Response ▶️ GenAI Script 👽 Simular AI Agent S 🦾 DrawDB with AI ✨ Ollama with LLama 3.2 Vision!!!! Preview 🚕 Powerful RAG Checker 📊 SQL Generator 💻 Role of LLMs 🐍 Document Extraction 🕶️ Open Source Vector DB Reddit 🍔 The Practical Guide to Self Hosting LLM 🦾 Stagehand Controller 🕶️ Understanding HNSWLIB 🐍 Best practices in RAG 💻 Enigma Agent 📝 Langchain, Ollama, Phi3 for Function Calling 🔋 Compass Judger 📝 Princeton NLP SimPO 🍔 Princeton NLP ProLong 🔋 Princeton NLP HELMET 🧐 Ollama Cheatsheet 🚕 Princeton NLP CopyCat 📊 Princeton NLP Shp 🕶️ Can LLM Solve Hard Github Issues 📝 Enabling Large Language Models to Generate Text with Citations 🔋 Princeton NLP CharXiv 📊 Awesome AI Agents List 🦾 Nomic’s Matryoshka text embedding model

Where to check it out: https://github.com/autogluon/autogluon

This seems not very convincing. There are other popular frameworks that provide AutoML with existing datasets (eg https://github.com/autogluon/autogluon)

https://github.com/awslabs/autogluon works well for image/text/tabular data

Some AutoML is getting pretty good, AutoGluon is very solid for tabular data. That being said you still need to have your data in tabular format and deployment still requires some effort.