bschifferer
bschifferer
**Describe the bug** I run a multi=GPU NVTabular workflow with LambdaOps. It executes the fit functionality, but when the pipeline does tramsform and to parquet, I get following error. I...
**Is your feature request related to a problem? Please describe.** Can we simplify to use multi-CPU NVTabular code? The criteo example code looks pretty complex for new users ``` #...
**Is your feature request related to a problem? Please describe.** As a user, I want that NVTabular produces the optimal dtype output for each columns. I would like a parameter...
**Describe the bug** I convert a dataset of .csv files to .parquet with ``` dataset.to_parquet(output_path, out_files_per_proc=40) ``` and I get the warning ``` /usr/local/lib/python3.8/dist-packages/merlin/io/dataset.py:862: UserWarning: Only created 1 files did...
**What needs doing** We want to develop and provide new examples for NVTabular: Similar to https://github.com/NVIDIA-Merlin/models/tree/main/examples Examples: - [ ] Getting Started - [ ] Advanced workflow - [ ]...
**Describe the bug** When starting mult-gpu training with PyTorch and Horovod, it runs successfully only if all workers have the same number of batches. If the workers have different number...
**Describe the bug** Numeric Inputs in TensorFlow example for scaling criteo are not used in the next layers
**Describe the bug** Although some columns are float32, NVTabular defines all input datatypes as int32. This causes error when deploying to Triton Inference Server  Run criteo example inlcuding the...
**Describe the bug** If I create a single batch with the NVTabular dataloader (PyTorch or TensorFlow), the GPU memory consumption is pretty high. In the case of PyTorch, a single...
- added container info to getting started and scaling criteo examples We discussed in example meeting to clean up NVTabular examples. I started to remove some of them in this...