[ALBERT] Checkpoints for pretrained models

google-research / albert

ALBERT: A Lite BERT for Self-supervised Learning of Language Representations

Apache License 2.0

3.23k stars 569 forks source link

[ALBERT] Checkpoints for pretrained models #103

Closed SahanaRamnath closed 4 years ago

SahanaRamnath commented 4 years ago

Hello! Thank you for releasing the code for Albert! Could you upload the pre-trained checkpoints for the 4 Albert models? I would like to run run_squad_sp.py directly for finetuning on SQuAD.

Could you also release instructions on how to run SQuAD using tensorflow hub directly? (Similar to run_classifier_with_tfhub.py?

Thanks in advance!

kamalkraj commented 4 years ago

https://github.com/kamalkraj/ALBERT-TF2.0 [WIP]

maziyarpanahi commented 4 years ago

Hi @kamalkraj, Is it possible to load them in TensorFlow 1.13.x or any way to convert them to TF 1.x?

0x0539 commented 4 years ago

@SahanaRamnath I believe everything should be clear from the README now. Please reopen if you still have questions.