ymcui
commited on
Commit
•
250b61e
1
Parent(s):
6fbbdf1
update usage
Browse files
README.md
CHANGED
|
@@ -4,6 +4,8 @@ language:
|
|
| 4 |
license: "apache-2.0"
|
| 5 |
---
|
| 6 |
|
|
|
|
|
|
|
| 7 |
## Chinese ELECTRA
|
| 8 |
Google and Stanford University released a new pre-trained model called ELECTRA, which has a much compact model size and relatively competitive performance compared to BERT and its variants.
|
| 9 |
For further accelerating the research of the Chinese pre-trained model, the Joint Laboratory of HIT and iFLYTEK Research (HFL) has released the Chinese ELECTRA models based on the official code of ELECTRA.
|
|
@@ -40,4 +42,4 @@ If you find our resource or paper is useful, please consider including the follo
|
|
| 40 |
url = "https://www.aclweb.org/anthology/2020.findings-emnlp.58",
|
| 41 |
pages = "657--668",
|
| 42 |
}
|
| 43 |
-
```
|
| 4 |
license: "apache-2.0"
|
| 5 |
---
|
| 6 |
|
| 7 |
+
**Please use `ElectraForPreTraining` for `discriminator` and `ElectraForMaskedLM` for `generator` if you are re-training these models.**
|
| 8 |
+
|
| 9 |
## Chinese ELECTRA
|
| 10 |
Google and Stanford University released a new pre-trained model called ELECTRA, which has a much compact model size and relatively competitive performance compared to BERT and its variants.
|
| 11 |
For further accelerating the research of the Chinese pre-trained model, the Joint Laboratory of HIT and iFLYTEK Research (HFL) has released the Chinese ELECTRA models based on the official code of ELECTRA.
|
| 42 |
url = "https://www.aclweb.org/anthology/2020.findings-emnlp.58",
|
| 43 |
pages = "657--668",
|
| 44 |
}
|
| 45 |
+
```
|