vaex.from_arrays(s=['a,b']).s.str.replace(r'(\w+)',r'--\g<1>==',regex=True)

when using capture group in str, it fails, while str_pandas.replace() is correct

Name: vaex
Version: 4.6.0
Summary: Out-of-Core DataFrames to visualize and explore big tabular datasets
Home-page:

Add documentation (initially just for the TabularPredictor) explaining what feature engineering AutoGluon does for the user by default:

• what is the universe of types each column can be
• how do we infer types per column
• for each type, what further feature engineering is done (e.g. ngrams and the date parsing I just submitted)
• what are the parameters used for the heuristics
• how does

Feature request

As requested by some, and as @ekamioka started on this PR #244. It might be interesting to get some helper functions to use embeddings as it's not the simplest concept in deep learning.

What is the expected behavior?
Calling a few helper function to get all the correct parameters before using TabNet

Example:
In the image below the word starships should begin on a new line to avoid being split.

Terminal width is provided to determine how many columns to print. The terminal width or the total width of the column headers may be used to wrap the text in the footer.

🐛 Bug

When I train a model I want to use it offline, so I save it, but when I load it from the saved model it still pulls the online model
https://github.com/PyTorchLightning/lightning-flash/blob/a0c97a39f2083b5344a08d248ccab7e5bfa91df4/flash/text/classification/model.py#L90

To Reproduce

https://www.kaggle.com/jirkaborovec/toxic-comments-with-lightning-flash-inference?scriptVersio

Is there a way to stabilise the results of the algorithm spot the diff drift detection?
In each run with same configuration and data the results of diff and p values are different.

Does HyperGBM's make_experiment return the best model?
How does it work on paramter tuning? It's say that, what's its seach space (e.g. in XGboost)???

It would be helpful if the progress bar for model fitting could be disabled. This is particularly relevant when trying to optimize model hyperparameters, when the following occurs:

Passing a disable_pbar or similar flag to `f

🚀 Feature request

The original PyTorch implementation of TabularDropout transformation is available at transformers4rec/torch/tabular/transformations.py