update paddlenlp text matching docs (#5041)

* add paddle.models api reference docs

* update docs

* update docs

* update docs

* tmp

* update docs

* update docs

* update docs

* update docs

PaddleNLP/examples/text_classification/pretrained_models/train.py

@@ -118,6 +118,7 @@ def set_seed(args):
     paddle.seed(args.seed)
 
 
+@paddle.no_grad()
 def evaluate(model, criterion, metric, data_loader):
     """
     Given a dataset, it evals model and computes the metric.

PaddleNLP/examples/text_matching/sentence_transformers/README.md

@@ -41,9 +41,8 @@ PaddleNLP提供了丰富的预训练模型，并且可以便捷地获取PaddlePa
 + BERT([Bidirectional Encoder Representations from Transformers](https://arxiv.org/abs/1810.04805))中文模型，简写`bert-base-chinese`， 其由12层Transformer网络组成。
 + ERNIE([Enhanced Representation through Knowledge Integration](https://arxiv.org/pdf/1904.09223))，支持ERNIE 1.0中文模型（简写`ernie-1.0`）和ERNIE Tiny中文模型（简写`ernie-tiny`)。
    其中`ernie`由12层Transformer网络组成，`ernie-tiny`由3层Transformer网络组成。
-+ RoBERTa([A Robustly Optimized BERT Pretraining Approach](https://arxiv.org/abs/1907.11692))，支持24层Transformer网络的`roberta-wwm-ext-large`和12层Transformer网络的`roberta-wwm-ext`。
++ RoBERTa([A Robustly Optimized BERT Pretraining Approach](https://arxiv.org/abs/1907.11692))，支持12层Transformer网络的`roberta-wwm-ext`。
 
-## TODO 增加模型效果
 | 模型  | dev acc | test acc |
 | ---- | ------- | -------- |
 | bert-base-chinese  | 0.86537 | 0.84440 |

PaddleNLP/paddlenlp/__init__.py

@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = '2.0.0a2'
+__version__ = '2.0.0a5'
 
 from . import data
 from . import datasets