NVIDIA
NVIDIA
BERT for TensorFlow2
Resource
NVIDIA
NVIDIA
BERT for TensorFlow2

BERT is a method of pre-training language representations which obtains state-of-the-art results on a wide array of NLP tasks.

NameSizeUpdatedActions
.dockerignore
870 BApril 28, 2021 UTC
classifier_data_lib.py
19.1 KBApril 28, 2021 UTC
common_flags.py
3.94 KBApril 28, 2021 UTC
create_finetuning_data.py
6.61 KBApril 28, 2021 UTC
create_pretraining_data.py
22.85 KBApril 28, 2021 UTC
dllogger_class.py
2.6 KBApril 28, 2021 UTC
Dockerfile
1.74 KBApril 28, 2021 UTC
gpu_affinity.py
1.97 KBApril 28, 2021 UTC
input_pipeline.py
8.58 KBApril 28, 2021 UTC
model_saving_utils.py
4.13 KBApril 28, 2021 UTC
README.md
73.03 KBApril 28, 2021 UTC