kathrinse / be_great

A novel approach for synthesizing tabular data using pretrained large language models
MIT License
276 stars 46 forks source link

TypeError: '<' not supported between instances of 'list' and 'int' #25

Closed Bezdarnost closed 1 year ago

Bezdarnost commented 1 year ago

image image

how to fix?

unnir commented 1 year ago

Can't reproduce the error. It works great on colab. Have you used the code from the readme page?

from be_great import GReaT
from sklearn.datasets import fetch_california_housing

data = fetch_california_housing(as_frame=True).frame

model = GReaT(llm='distilgpt2', batch_size=32, epochs=50)
model.fit(data)
synthetic_data = model.sample(n_samples=100)

If yes, please try to reinstall the be_great and restart your jupyter notebook. Plus, I suggest deleting files checkpoints.

Hope it helps.

unnir commented 1 year ago

also: https://github.com/kathrinse/be_great/issues/12

Bezdarnost commented 1 year ago

oh, training started, thank you a lot(honestly, I've tried a couple of reinstallations)

unnir commented 1 year ago

happy to read it, if you have more questions, do not hesitate to ask :)