Sara Rabhi
Sara Rabhi
**Describe the bug** I tested the end-to-end example from the Transformers4Rec repo ( [link](https://github.com/NVIDIA-Merlin/Transformers4Rec/blob/main/examples/end-to-end-session-based/02-End-to-end-session-based-with-Yoochoose-PyT.ipynb) to the notebook). But I am getting an error from the NVTabular workflow related to the...
Slicing sequence of past interactions can vary between rows based on a specified rule (such as taking 80% of the original sequence, taking interactions from the last hour.) **Describe the...
## Problem: The integration of session-based support in Merlin Models requires the implementation/refractory of different blocks. (Ticket #433 summarizes the motivation and an overview of all the requirements). ## Goal:...
## Problem: Following the integration of session-based support (#451, #452), the missing component would be providing recsys practitioners with all the necessary blocks for advanced use-cases such as multi-task learning...
## Problem: The session-based blocks defined in #451 are implemented as standalone tasks and we need to link them together to provide known session-based models. ## Goal: - Implement the...
# 🚀 Feature request There is an open [issue](https://github.com/NVIDIA-Merlin/Transformers4Rec/issues/28 ) reporting that the NVT PyT data-loader is not correctly supporting `torch.nn.parallel.DistributedDataParallel`. This feature is to expose the missing option in...
### Bug description Calling the method fit after re-loading the model from disk raises the following issue: ``` _ _ _ _ _ _ _ _ _ _ _ _...
### Bug description mypy checking is failing when installing latest HF transformers (4.19.1) ### Steps/Code to reproduce bug - Install latest HF transformers - run the following command: ```mypy transformers4rec...
### Bug description There Is an information leakage when training TF XLNet model using `mlm` task. The returned ranking metrics are close to 1.  ### Steps/Code to reproduce bug...
- The current implementation of ranking metrics computes the top k scores within each metric class using the method [extract_topk ](https://github.com/NVIDIA-Merlin/Transformers4Rec/blob/238fdc8176b2fe43d72e92c5ebb259904f94810d/transformers4rec/torch/ranking_metric.py#L65) which is computationally expensive. - A possible solution is...