Cannot Convert float NaN to Int

[epi-cvu]

Environment Details

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

• SDV version: Latest
• Python version: 3.10.11
• Operating System: Windows 11

Error Description

I was trying the HMA Synthesizer tutorial using the documentation. And after the fitting process I tried sampling on scale 1 and 0.1 and it failed and gave me an error "cannot convert float NaN to Int"

Steps to reproduce

To reproduce this, we will have to download the CSV from MIMIC IV (physionet) Data set and specially "patient" and "admissions" and put them in the same folder. I followed all the steps of multi-table synthetic data with HMASynthesizer and waited for the fit to finish. After the fit step, I tried sampling with a scale of 1 and it gives me "Cannot convert float NaN to int". I went ahead and tried modifying the columbs anchor_year and anchor_age with computer representation Int32, but still gives me the same message.

Paste the command(s) you ran and the output.
If there was a crash, please include the traceback here.
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Cell In[82], [line 1](vscode-notebook-cell:?execution_count=82&line=1)
----> [1](vscode-notebook-cell:?execution_count=82&line=1) synthesizer.sample(scale=0.1)

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\sdv\multi_table\base.py:393, in BaseMultiTableSynthesizer.sample(self, scale)
    [389](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/multi_table/base.py:389)     raise SynthesizerInputError(
    [390](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/multi_table/base.py:390)         f"Invalid parameter for 'scale' ({scale}). Please provide a number that is >0.0.")
    [392](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/multi_table/base.py:392) with self._set_temp_numpy_seed():
--> [393](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/multi_table/base.py:393)     sampled_data = self._sample(scale=scale)
    [395](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/multi_table/base.py:395) return sampled_data

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\sdv\sampling\hierarchical_sampler.py:222, in BaseHierarchicalSampler._sample(self, scale)
    [220](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:220)     LOGGER.info(f'Sampling {num_rows} rows from table {table}')
    [221](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:221)     sampled_data[table] = self._sample_rows(synthesizer, num_rows)
--> [222](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:222)     self._sample_children(table_name=table, sampled_data=sampled_data)
    [224](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:224) added_relationships = set()
    [225](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:225) for relationship in self.metadata.relationships:

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\sdv\sampling\hierarchical_sampler.py:142, in BaseHierarchicalSampler._sample_children(self, table_name, sampled_data)
    [140](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:140) if child_name not in sampled_data:  # Sample based on only 1 parent
    [141](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:141)     for _, row in sampled_data[table_name].iterrows():
--> [142](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:142)         self._add_child_rows(
    [143](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:143)             child_name=child_name,
    [144](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:144)             parent_name=table_name,
    [145](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:145)             parent_row=row,
    [146](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:146)             sampled_data=sampled_data
    [147](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:147)         )
    [149](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:149)     if child_name not in sampled_data:  # No child rows sampled, force row creation
    [150](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:150)         foreign_key = self.metadata._get_foreign_keys(table_name, child_name)[0]

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\sdv\sampling\hierarchical_sampler.py:108, in BaseHierarchicalSampler._add_child_rows(self, child_name, parent_name, parent_row, sampled_data, num_rows)
    [105](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:105)     num_rows = self._get_num_rows_from_parent(parent_row, child_name, foreign_key)
    [106](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:106) child_synthesizer = self._recreate_child_synthesizer(child_name, parent_name, parent_row)
--> [108](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:108) sampled_rows = self._sample_rows(child_synthesizer, num_rows)
    [110](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:110) if len(sampled_rows):
    [111](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:111)     parent_key = self.metadata.tables[parent_name].primary_key

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\sdv\sampling\hierarchical_sampler.py:71, in BaseHierarchicalSampler._sample_rows(self, synthesizer, num_rows)
     [69](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:69) if num_rows is None:
     [70](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:70)     num_rows = synthesizer._num_rows
---> [71](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/sampling/hierarchical_sampler.py:71) return synthesizer._sample_batch(int(num_rows), keep_extra_columns=True)

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\sdv\single_table\base.py:602, in BaseSingleTableSynthesizer._sample_batch(self, batch_size, max_tries, conditions, transformed_conditions, float_rtol, progress_bar, output_file_path, keep_extra_columns)
    [600](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:600) while num_valid < batch_size and counter < max_tries:
    [601](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:601)     prev_num_valid = num_valid
--> [602](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:602)     sampled, num_valid = self._sample_rows(
    [603](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:603)         num_rows_to_sample,
    [604](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:604)         conditions,
    [605](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:605)         transformed_conditions,
    [606](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:606)         float_rtol,
    [607](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:607)         sampled,
    [608](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:608)         keep_extra_columns
    [609](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:609)     )
    [611](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:611)     num_new_valid_rows = num_valid - prev_num_valid
    [612](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:612)     num_increase = min(num_new_valid_rows, remaining)

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\sdv\single_table\base.py:519, in BaseSingleTableSynthesizer._sample_rows(self, num_rows, conditions, transformed_conditions, float_rtol, previous_rows, keep_extra_columns)
    [516](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:516)     except NotImplementedError:
    [517](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:517)         raw_sampled = self._sample(num_rows)
--> [519](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:519) sampled = self._data_processor.reverse_transform(raw_sampled)
    [520](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:520) if keep_extra_columns:
    [521](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/single_table/base.py:521)     input_columns = self._data_processor._hyper_transformer._input_columns

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\sdv\data_processing\data_processor.py:851, in DataProcessor.reverse_transform(self, data, reset_keys)
    [849](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/data_processing/data_processor.py:849) try:
    [850](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/data_processing/data_processor.py:850)     if not data.empty:
--> [851](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/data_processing/data_processor.py:851)         reversed_data = self._hyper_transformer.reverse_transform_subset(
    [852](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/data_processing/data_processor.py:852)             data[reversible_columns]
    [853](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/data_processing/data_processor.py:853)         )
    [854](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/data_processing/data_processor.py:854) except rdt.errors.NotFittedError:
    [855](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/sdv/data_processing/data_processor.py:855)     LOGGER.info(f'HyperTransformer has not been fitted for table {self.table_name}')

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\rdt\hyper_transformer.py:887, in HyperTransformer.reverse_transform_subset(self, data)
    [876](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/hyper_transformer.py:876) def reverse_transform_subset(self, data):
    [877](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/hyper_transformer.py:877)     """Revert the transformations for a subset of the fitted columns.
    [878](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/hyper_transformer.py:878) 
    [879](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/hyper_transformer.py:879)     Args:
   (...)
    [885](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/hyper_transformer.py:885)             Reversed subset.
    [886](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/hyper_transformer.py:886)     """
--> [887](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/hyper_transformer.py:887)     return self._reverse_transform(data, prevent_subset=False)

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\rdt\hyper_transformer.py:870, in HyperTransformer._reverse_transform(self, data, prevent_subset)
    [868](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/hyper_transformer.py:868)         output_columns = transformer.get_output_columns()
    [869](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/hyper_transformer.py:869)         if output_columns and set(output_columns).issubset(data.columns):
--> [870](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/hyper_transformer.py:870)             data = transformer.reverse_transform(data)
    [872](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/hyper_transformer.py:872) reversed_columns = self._subset(self._input_columns, data.columns)
    [874](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/hyper_transformer.py:874) return data.reindex(columns=reversed_columns)

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\rdt\transformers\base.py:55, in random_state.<locals>.wrapper(self, *args, **kwargs)
     [53](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/transformers/base.py:53) method_name = function.__name__
     [54](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/transformers/base.py:54) with set_random_states(self.random_states, method_name, self.set_random_state):
---> [55](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/transformers/base.py:55)     return function(self, *args, **kwargs)

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\rdt\transformers\base.py:477, in BaseTransformer.reverse_transform(self, data)
    [475](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/transformers/base.py:475) data = data.copy()
    [476](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/transformers/base.py:476) columns_data = self._get_columns_data(data, self.output_columns)
--> [477](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/transformers/base.py:477) reversed_data = self._reverse_transform(columns_data)
    [478](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/transformers/base.py:478) data = data.drop(self.output_columns, axis=1)
    [479](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/transformers/base.py:479) data = self._add_columns_to_data(data, reversed_data, self.columns)

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\rdt\transformers\categorical.py:195, in UniformEncoder._reverse_transform(self, data)
    [192](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/transformers/categorical.py:192)         labels.append(key)
    [194](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/transformers/categorical.py:194) result = pd.cut(data, bins=bins, labels=labels, include_lowest=True)
--> [195](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/rdt/transformers/categorical.py:195) return result.replace(nan_name, np.nan).astype(self.dtype)

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\pandas\core\generic.py:6534, in NDFrame.astype(self, dtype, copy, errors)
   [6530](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/generic.py:6530)     results = [ser.astype(dtype, copy=copy) for _, ser in self.items()]
   [6532](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/generic.py:6532) else:
   [6533](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/generic.py:6533)     # else, only a single dtype is given
-> [6534](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/generic.py:6534)     new_data = self._mgr.astype(dtype=dtype, copy=copy, errors=errors)
   [6535](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/generic.py:6535)     res = self._constructor_from_mgr(new_data, axes=new_data.axes)
   [6536](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/generic.py:6536)     return res.__finalize__(self, method="astype")

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\pandas\core\internals\managers.py:414, in BaseBlockManager.astype(self, dtype, copy, errors)
    [411](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/managers.py:411) elif using_copy_on_write():
    [412](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/managers.py:412)     copy = False
--> [414](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/managers.py:414) return self.apply(
    [415](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/managers.py:415)     "astype",
    [416](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/managers.py:416)     dtype=dtype,
    [417](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/managers.py:417)     copy=copy,
    [418](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/managers.py:418)     errors=errors,
    [419](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/managers.py:419)     using_cow=using_copy_on_write(),
    [420](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/managers.py:420) )

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\pandas\core\internals\managers.py:354, in BaseBlockManager.apply(self, f, align_keys, **kwargs)
    [352](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/managers.py:352)         applied = b.apply(f, **kwargs)
    [353](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/managers.py:353)     else:
--> [354](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/managers.py:354)         applied = getattr(b, f)(**kwargs)
    [355](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/managers.py:355)     result_blocks = extend_blocks(applied, result_blocks)
    [357](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/managers.py:357) out = type(self).from_blocks(result_blocks, self.axes)

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\pandas\core\internals\blocks.py:616, in Block.astype(self, dtype, copy, errors, using_cow)
    [596](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/blocks.py:596) """
    [597](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/blocks.py:597) Coerce to the new dtype.
    [598](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/blocks.py:598) 
   (...)
    [612](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/blocks.py:612) Block
    [613](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/blocks.py:613) """
    [614](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/blocks.py:614) values = self.values
--> [616](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/blocks.py:616) new_values = astype_array_safe(values, dtype, copy=copy, errors=errors)
    [618](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/blocks.py:618) new_values = maybe_coerce_values(new_values)
    [620](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/internals/blocks.py:620) refs = None

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\pandas\core\dtypes\astype.py:238, in astype_array_safe(values, dtype, copy, errors)
    [235](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/dtypes/astype.py:235)     dtype = dtype.numpy_dtype
    [237](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/dtypes/astype.py:237) try:
--> [238](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/dtypes/astype.py:238)     new_values = astype_array(values, dtype, copy=copy)
    [239](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/dtypes/astype.py:239) except (ValueError, TypeError):
    [240](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/dtypes/astype.py:240)     # e.g. _astype_nansafe can fail on object-dtype of strings
    [241](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/dtypes/astype.py:241)     #  trying to convert to float
    [242](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/dtypes/astype.py:242)     if errors == "ignore":

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\pandas\core\dtypes\astype.py:180, in astype_array(values, dtype, copy)
    [176](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/dtypes/astype.py:176)     return values
    [178](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/dtypes/astype.py:178) if not isinstance(values, np.ndarray):
    [179](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/dtypes/astype.py:179)     # i.e. ExtensionArray
--> [180](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/dtypes/astype.py:180)     values = values.astype(dtype, copy=copy)
    [182](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/dtypes/astype.py:182) else:
    [183](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/dtypes/astype.py:183)     values = _astype_nansafe(values, dtype, copy=copy)

File [c:\Users\Charles](file:///C:/Users/Charles) VU\Documents\GitHub\Stage\.venv\lib\site-packages\pandas\core\arrays\categorical.py:550, in Categorical.astype(self, dtype, copy)
    [547](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/arrays/categorical.py:547)     return super().astype(dtype, copy=copy)
    [549](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/arrays/categorical.py:549) elif dtype.kind in "iu" and self.isna().any():
--> [550](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/arrays/categorical.py:550)     raise ValueError("Cannot convert float NaN to integer")
    [552](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/arrays/categorical.py:552) elif len(self.codes) == 0 or len(self.categories) == 0:
    [553](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/arrays/categorical.py:553)     result = np.array(
    [554](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/arrays/categorical.py:554)         self,
    [555](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/arrays/categorical.py:555)         dtype=dtype,
    [556](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/arrays/categorical.py:556)         copy=copy,
    [557](file:///C:/Users/Charles%20VU/Documents/GitHub/Stage/.venv/lib/site-packages/pandas/core/arrays/categorical.py:557)     )

ValueError: Cannot convert float NaN to integer

[npatki]

Seems like this issue was resolved by modifying the default distribution to 'norm'. The root cause is likely #1691. The good news is that this issue has already been resolved and the fix should be available in the upcoming SDV release (0.10.0). After this release, you can leave the default distribution as-is and it should work without failing.