Pelumi
Results
2
issues of
Pelumi
For a mixed precision approach, I observed that the query_embeddings for the retrieval task gets cast to the global precision policy while the candidate_embedding preserves its dtype. It causes the...
**Location of incorrect documentation** In the Training a DLRM Model section, [reference_file](https://github.com/NVIDIA-Merlin/NVTabular/blob/main/examples/01-Getting-started.ipynb), the text states that the prediction task is binary classification even though the code shows a regression task...