-
-
在尝试用albert_large_zh模型跑fine-tuning时出错，按照以下命令执行
```
export BERT_BASE_DIR=./albert_large_zh
export TEXT_DIR=./lcqmc
nohup python3 run_classifier.py --task_name=lcqmc_pair --do_train=true…
-
Hi,
謝謝提供這個很好的代碼,我遇到了以下的問題,想請教你
1.當我把albert_base 換成 albert_xlarge時, 在model.py檔中的forword function再取出hidden states時,從原來的 [64,128,768] 變成 [128,2048],將會導致給self.lstm的input維度錯誤,這個是albert模型本身的問題嗎？
`all_h…
-
用example/task_sentiment_albert.py加载徐亮的tiny版albert ,出现下面错误
ValueError: Layer weight shape (21128, 312) not compatible with provided weight shape (21128, 128)
用keras_bert加载也出现同样的错，这是怎么回事啊？
-
环境 Ubuntu 18.04, Python 3.6
```
In [2]: recognizer = hanlp.load(hanlp.pretrained.ner.MSRA_NER_BERT_BASE_CN)
---------------------------------------------------------------------------
AssertionErr…
-
I want to load ALBERT in Pytorch, so I convert albert_zh_xlarge from Tensorflow to Pytorch and succeeded.
But I don't know how to load it in Pytorch, is there some usage examples of it?
-
will you release an ALBERT model?
it sets the new state of art;
# 🌟New model addition
ALBERT: A LITE BERT FOR SELF-SUPERVISED
LEARNING OF LANGUAGE REPRESENTATIONS
https://arxiv.org/pdf/1909.11…
-
---
name: Bug Report
about: Something isn't working as expected
title: The relative path of the post image is not correct.
labels: Bug
assignees: ''
---
Please follow this Issue template to…
-
文档中：
- embeddings
- chinese_L-12_H-768_A-12/(取谷歌预训练好点的模型,已经压缩上传,
keras-bert还可以加载百度版ernie(需转换，[https://github.com/ArthurRizar/tensorflow_ernie](https://githu…
-
