Error when columns contains only numbers. PR exists.

[gszecsenyi]

Environment Details

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

• RDT version: latest
• Python version: 3.9.6
• Operating System: OSX

Error Description

In the case where the column names contain only numbers, typically when scaling the data, the code doesn't work well because the first value of the variable will be a numeric value, not a string, to which you cannot append a string later.

Steps to reproduce

Create a Pandas DF, which contains only numbers as column names.

synthesizer = SingleTablePreset(
    metadata,
    name='FAST_ML'
)

synthesizer.fit(
    data=test_df
)

synthetic_data = synthesizer.sample(
    num_rows=500
)

synthetic_data.head()

I created a pull request: https://github.com/sdv-dev/RDT/pull/737