TabFM: A zero-shot foundation model for tabular data

Discussion (6 Comments)Read Original on HackerNews

woadwarrior01•about 1 hour ago

In hindsight, the Prior Labs exit to SAP couldn't have been timed better.

actusual•about 2 hours ago

150,000 rows of data, where will I store it all?!

nojito•about 1 hour ago

The biggest misconception that people have when modeling using tabular data is that more data = better model.

hodgehog11•about 3 hours ago

On the one hand, this is impressive. TabPFN was already state of the art and is seriously shaking up Bayesian prediction for tabular data (which is almost everything).

On the other hand, perhaps it is just me, but I do not feel that this is an acceptable form of benchmark reporting in this domain. TabArena actually has multiple metrics, since ELO does not properly quantify the degree of improvement. The fact that these are not displayed here should give pause. Also the results section in the GitHub is a dumpster fire.

Eridrus•about 2 hours ago

GitHub Repo: Please see the results folder

Results folder: Here's some undocumented parquet files

Definitely feels like they're hiding the ball lol.

If they had good benchmarks they'd talk about them.

Not comparing to tuned xgboost is also a warning sign.

kingjimmy•about 3 hours ago

interesting to see this from Google after the SAP acquisition of Prior Labs.

TabFM: A zero-shot foundation model for tabular data

⚡ Community Insights

Discussion (6 Comments)Read Original on HackerNews