UInt8 conversion error with HyperTransform

[Metalkiler]

Environment Details

Please indicate the following details about the environment in which you found the bug:

• RDT version: 1.0
• Python version: 3.8
• Operating System: Ubuntu

Error Description

I was trying to convert a pandas dataframe using the hypertransformer obtained from the openml website, however this process gave this error
File "/home/luismatos/.conda/envs/luismatos/lib/python3.8/site-packages/rdt/hyper_transformer.py", line 600, in detect_initial_config
self._learn_config(data)
File "/home/luismatos/.conda/envs/luismatos/lib/python3.8/site-packages/rdt/hyper_transformer.py", line 574, in _learn_config
self._set_field_sdtype(data, field)
File "/home/luismatos/.conda/envs/luismatos/lib/python3.8/site-packages/rdt/hyper_transformer.py", line 559, in _set_field_sdtype
self.field_sdtypes[field] = self._DTYPES_TO_SDTYPES[kind]
KeyError: 'u'

however i was expecting something like the example provided in the github page

Steps to reproduce

import openml
import ast
#sys argument is used write the id of a dataset in the openml website and the target column 

print("Running code with dataset:", sys.argv[1])
print("Target column", sys.argv[2])
data = openml.datasets.get_dataset(sys.argv[1])

df = data.get_data()[0] 

print(irisdf.info())

ht = HyperTransformer()

ht.detect_initial_config(df)

ht.fit(df)

df = ht.transform(df)

However using the following process:

tipos = irisdf.convert_dtypes().dtypes
listaTipos = [str(i) for i in tipos]
for i in trange(0, len(listaTipos)):
  if listaTipos[i] == "UInt8":
    cols = tipos.keys()[i]
    df[cols] = df[cols].astype(str).astype(int)

it solves the issue! It only happens with the UInt8 (so far as I know).

Thanks for the attention :)