Index contains NaNs when using sample_remaining_columns

[amontanez24]

Environment Details

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

• SDV version: 0.17.0.dev1
• Python version: Any
• Operating System: Any

Error Description

When calling sample_remaining_columns with conditions that have different values, the index in the output contains Nans.

Steps to reproduce

Create any model and fit to the demo data. Then create conditions on the same column with multiple different values. Call sample_remaining_columns with those conditions.

from sdv.sampling import Condition
import pandas as pd 
from sdv.tabular import CopulaGAN
from sdv.demo import load_tabular_demo

data = load_tabular_demo('student_placements')

model = CopulaGAN(
    primary_key='student_id',
    epochs=500,
    batch_size=100,
    generator_dim=(256, 256, 256),
    discriminator_dim=(256, 256, 256)
)
model.fit(data)
conditions = pd.DataFrame({
    'gender': ['M', 'M', 'M', 'F', 'F', 'F']
})
sampled = model.sample_remaining_columns(conditions)