@Suhail i used an extractor on my cursor, claude code, codex, etc sessions to make a private dataset on hf to train/calibrate my own models on.
Hugging Face CEO Extracts Private Dataset From AI Coding Sessions For Model Training
Some users reacted positively to the Hugging Face CEO's table urging custom AI datasets by seeking elaboration on its value and process, while others dismissed the high effort as delivering only marginal precision gains.
No Digg Deeper questions have been answered for this story yet.
Most Activity

@elliotarledge What is this?

@Suhail i used an extractor on my cursor, claude code, codex, etc sessions to make a private dataset on hf to train/calibrate my own models on.
make your own datasets

@elliotarledge what's the context? can't tell what you're referring to

@elliotarledge @Suhail 3h d'ingestion pour gagner 20% de précision. c'est comme passer 3h au four à pain juste pour avoir une mie un peu moins cauchemardesque

@elliotarledge Cool - can you please elaborate on why this is valuable and process?

@elliotarledge We are. We’re flooding Hugging Face with lungs. Anatomically correct ones. For the greater good. https://huggingface.co/spaces/chrisvoncsefalvay/1000lungs-dataset-viewer

@elliotarledge seeing some of these models names (codex-max) was more of a trip down memory lane. It feels like well over year ago that we had that model

@elliotarledge @Suhail (Ps. You know there’s a specific dataset format for this on HF?)

@elliotarledge هلا