#[BERT/TF2] issue raised on NVIDIA github

1 messages · Page 1 of 1 (latest)

shrewd orchid
#

HI Folks
No activity and failed stack when pre-training starts
Below is the Issue link with all the details
https://github.com/NVIDIA/DeepLearningExamples/issues/1262

Please let me know if more details are required .
Any help or suggestion will affect me big , I am just stuck now 😦

GitHub

Related to Model/Framework(s) BERT/TF2 Describe the bug Below is the full trace from start of pre-training and no activity after that root@node001:~/bert_tf2# bash scripts/run_pretraining_lamb.sh 6...

shrewd orchid
#

Anyone who can have a look at the issue

shrewd orchid
#

anyone who can help ...guys

sly trail
#

I blame tensorflow 🤢 does anyone actually still use it

#

Does TF even support Cuda 11.7? I thought the highest it went is 11.2, and unlike PyTorch, TF is super sensitive to these things

muted dust