Paul Sharpe

Related to **ELECTRA/TensorFlow2** **Describe the bug** The README file includes info about the _warmup_ steps, but it describes it as a _percentage_ whilst also showing a default value that is...

Related to **ELECTRA/TensorFlow2** **Describe the bug** When running the pretraining as root@biber:/workspace/electra# bash scripts/run_pretraining.sh $(source scripts/configs/pretrain_config.sh && dgxa100_1gpu_amp) there are a lot of Warning messages output to _stderr_. These may...

Related to **ELECTRA/TensorFlow2** **Describe the bug** (1) The README.md includes a link "Getting Started Using NVIDIA GPU Cloud" to https://docs.nvidia.com/ngc/ngc-getting-started-guide/index.html but from the file this displays as simply _403 Forbidden_...

Related to **ELECTRA/TensorFlow2** **Describe the bug** The README in the MultiNode section says BATCHSIZE=176 LR=6e-3 GRAD_ACCUM_STEPS=1 PHASE=1 STEPS=10000 WARMUP=2000 b1=0.878 b2=0.974 decay=0.5 skip_adaptive=yes end_lr=0.0 sbatch **N48** --ntasks-per-node=8 run.sub BATCHSIZE=24 LR=4e-3...

[DLRM/TensorFlow2] The --dummy_model option For Training Does Not Seem To Work

1

Related to **DLRM/TensorFlow2** **Describe the bug** I generated synthetic data as per the README instructions, and tried to perform a training run. This failed when creating a _dlrm_ training pipeline,...

Related to **ELECTRA/TF2** **Describe the bug** As a proof-of-concept of the ELECTRA/TF2 process, I have downloaded the 90Gb wiki data. This contains 20-million pages of data. I have extracted just...

Related to **ELECTRA/TF2** **Is your feature request related to a problem? Please describe.** I have previously run the README command to download the wiki data: /workspace/electra/data/create_datasets_from_start.sh wiki_only It has spent...

Related to **ELECTRA/TF2** **Is your feature request related to a problem? Please describe.** The README shows that the datasets can be created from wiki-only: /workspace/electra/data/create_datasets_from_start.sh wiki_books but when you then...

Related to **DLRM/TensorFlow2** **Describe the bug** A minor bug: running the _transcode.py_ as per the README generates a deprecation warning from python. **To Reproduce** root@biber:/home/psharpe/DeepLearningExamples-master/TensorFlow2/Recommendation/DLRM# export DATASET_PARENT_DIRECTORY=`pwd` root@biber:/home/psharpe/DeepLearningExamples-master/TensorFlow2/Recommendation/DLRM# echo $DATASET_PARENT_DIRECTORY...