Bug description

Currently in the document https://nvidia-merlin.github.io/Transformers4Rec/main/multi_gpu_train.html, we have the following script for users to repartiton the dataset. But this script is not working and is not repartitioning the dataset.

df.to_parquet("filename.parquet", row_group_size=10000)

This script should be modified in a way that the dataset is repartitioned properly.