NVIDIA
NVIDIA
BERT for TensorFlow
Resource
NVIDIA
NVIDIA
BERT for TensorFlow

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

NameSizeUpdatedActions
__init__.py
616 BJanuary 6, 2021 UTC
.dockerignore
779 BJanuary 6, 2021 UTC
CONTRIBUTING.md
1.29 KBJanuary 6, 2021 UTC
Dockerfile
1.28 KBJanuary 6, 2021 UTC
extract_features.py
13.68 KBJanuary 6, 2021 UTC
fp16_utils.py
1.47 KBJanuary 6, 2021 UTC
fused_layer_norm.py
5.49 KBJanuary 6, 2021 UTC
gpu_environment.py
1.53 KBJanuary 6, 2021 UTC
LICENSE
11.09 KBJanuary 6, 2021 UTC
modeling.py
37.84 KBJanuary 6, 2021 UTC
NOTICE
158 BJanuary 6, 2021 UTC
README.md
118.07 KBJanuary 6, 2021 UTC