update docs

PaddlePaddle · Oct 19, 2021 · 79169c2 · 79169c2
1 parent d9565fc
commit 79169c2
Show file tree

Hide file tree

Showing 2 changed files with 28 additions and 8 deletions.
diff --git a/community/junnyu/ckiplab-bert-base-chinese-ner/README.md b/community/junnyu/ckiplab-bert-base-chinese-ner/README.md
@@ -12,7 +12,7 @@
 import paddle
 import paddle.nn.functional as F
 from paddlenlp.transformers import BertForTokenClassification, BertTokenizer
-path = "ckiplab-bert-base-chinese-ner"
+path = "junnyu/ckiplab-bert-base-chinese-ner"
 model = BertForTokenClassification.from_pretrained(path)
 model.eval()
 tokenizer = BertTokenizer.from_pretrained(path)

diff --git a/docs/model_zoo/transformers.rst b/docs/model_zoo/transformers.rst
@@ -9,8 +9,8 @@ PaddleNLP为用户提供了常用的 ``BERT``、``ERNIE``、``ALBERT``、``RoBER
 Transformer预训练模型汇总
 ------------------------------------
 
-下表汇总了介绍了目前PaddleNLP支持的各类预训练模型以及对应预训练权重。我们目前提供了**21**种网络结构， **91** 种预训练的参数权重供用户使用，
-其中包含了 **45** 种中文语言模型的预训练权重。
+下表汇总了介绍了目前PaddleNLP支持的各类预训练模型以及对应预训练权重。我们目前提供了**21**种网络结构， **96** 种预训练的参数权重供用户使用，
+其中包含了 **48** 种中文语言模型的预训练权重。
 
 +--------------------+-----------------------------------------+--------------+-----------------------------------------+
 | Model              | Pretrained Weight                       | Language     | Details of the model                    |
@@ -124,6 +124,31 @@ Transformer预训练模型汇总
 |                    |                                         |              | and Traditional text using              |
 |                    |                                         |              | Whole-Word-Masking with extented data.  |
 |                    +-----------------------------------------+--------------+-----------------------------------------+
+|                    |``junnyu/ckiplab-bert-base-chinese-ner`` | Chinese      | 12-layer, 768-hidden,                   |
+|                    |                                         |              | 12-heads, 102M parameters.              |
+|                    |                                         |              | Finetuned on NER task.                  |
+|                    +-----------------------------------------+--------------+-----------------------------------------+
+|                    |``junnyu/ckiplab-bert-base-chinese-pos`` | Chinese      | 12-layer, 768-hidden,                   |
+|                    |                                         |              | 12-heads, 102M parameters.              |
+|                    |                                         |              | Finetuned on POS task.                  |
+|                    +-----------------------------------------+--------------+-----------------------------------------+
+|                    |``junnyu/ckiplab-bert-base-chinese-ws``  | Chinese      | 12-layer, 768-hidden,                   |
+|                    |                                         |              | 12-heads, 102M parameters.              |
+|                    |                                         |              | Finetuned on WS task.                   |
+|                    +-----------------------------------------+--------------+-----------------------------------------+
+|                    |``junnyu/nlptown-bert-base-``            | Multilingual | 12-layer, 768-hidden,                   |
+|                    |``multilingual-uncased-sentiment``       |              | 12-heads, 167M parameters.              |
+|                    |                                         |              | Finetuned for sentiment analysis on     |
+|                    |                                         |              | product reviews in six languages:       |
+|                    |                                         |              | English, Dutch, German, French,         |
+|                    |                                         |              | Spanish and Italian.                    |
+|                    +-----------------------------------------+--------------+-----------------------------------------+
+|                    |``junnyu/tbs17-MathBERT``                | English      | 12-layer, 768-hidden,                   |
+|                    |                                         |              | 12-heads, 110M parameters.              |
+|                    |                                         |              | Trained on pre-k to graduate math       |
+|                    |                                         |              | language (English) using a masked       |
+|                    |                                         |              | language modeling (MLM) objective.      |
+|                    +-----------------------------------------+--------------+-----------------------------------------+
 |                    |``macbert-base-chinese``                 | Chinese      | 12-layer, 768-hidden,                   |
 |                    |                                         |              | 12-heads, 102M parameters.              |
 |                    |                                         |              | Trained with novel MLM as correction    |
@@ -133,11 +158,6 @@ Transformer预训练模型汇总
 |                    |                                         |              | 16-heads, 326M parameters.              |
 |                    |                                         |              | Trained with novel MLM as correction    |
 |                    |                                         |              | pre-training task.                      |
-|                    +-----------------------------------------+--------------+-----------------------------------------+
-|                    |``simbert-base-chinese``                 | Chinese      | 12-layer, 768-hidden,                   |
-|                    |                                         |              | 12-heads, 108M parameters.              |
-|                    |                                         |              | Trained on 22 million pairs of similar  |
-|                    |                                         |              | sentences crawed from Baidu Know.       |
 +--------------------+-----------------------------------------+--------------+-----------------------------------------+
 |BigBird_            |``bigbird-base-uncased``                 | English      | 12-layer, 768-hidden,                   |
 |                    |                                         |              | 12-heads, _M parameters.                |