Simplify the creation of an environment with the tools required to conduct an effective data analysis
data_schema = {
'feature': {
## For general data wrangling functions
'relevant': ([boolean] indicating if the feature is eliminated in a first phase),
'clean_col_name': ([string] string with the new defined name for the feature),
'data_type': ([string] data type indicating to format the feature (str, int, float, datetime)),
'value_map': ([dict] dict with keys as original name and value as the substitution),
## For ML
'feature_type': ([string] feature type for feature engineering (categorical, numeric)),
'model_relevant': ([boolean] indicating if the feature will be fed to the model),
'id_feature': ([boolean] boolean to indicate if this is the id feature in the dataset),
'predict_label': ([boolean] boolean to indicate if this is the predict label in the dataset),
}
}