jingyanwangms
jingyanwangms
- There're two sets of notebooks and it looks confusing: https://github.com/microsoft/AzureML-BERT/tree/master/finetune/PyTorch/notebooks  - There're two run_classifier_azureml.py. Also confusing https://github.com/microsoft/AzureML-BERT/blob/master/finetune/PyTorch/run_classifier_azureml.py https://github.com/microsoft/AzureML-BERT/blob/master/finetune/run_classifier_azureml.py - Need to add better instruction for downloading glue data....
### Description Log ORTModule errors in loglevel
# What does this PR do? Fixes [# 1705](https://github.com/huggingface/optimum/issues/1705)
### System Info ```shell optimum commit 4e987762cdd33c237a63538c181b1ccfa4d7648a Author: fxmarty Date: Thu Feb 15 13:59:11 2024 +0100 Probably caused by breaking change introduced in https://github.com/huggingface/transformers/commit/5f06053dd821c91f7bd697309109abaa3396b605 ``` ### Who can help? @JingyaHuang...
### Description Add GemmaRotaryEmbeddingGrad kernel which is the gradient kernel of https://github.com/microsoft/onnxruntime/pull/20267 ### Motivation and Context
## Description ## Environment **TensorRT Version**: 10.4.0.26-1+cuda12.6 (upgrading from 10.3) **NVIDIA GPU**: V100 **NVIDIA Driver Version**: **CUDA Version**: Cuda compilation tools, release 12.5, V12.5.82 **CUDNN Version**: 9 Operating System: Python...