diff --git a/fastNLP/__init__.py b/fastNLP/__init__.py
index 330ad9c2..54d96959 100644
--- a/fastNLP/__init__.py
+++ b/fastNLP/__init__.py
@@ -10,7 +10,7 @@ fastNLP 由 :mod:`~fastNLP.core` 、 :mod:`~fastNLP.io` 、:mod:`~fastNLP.module
 
 fastNLP 中最常用的组件可以直接从 fastNLP 包中 import ，他们的文档如下：
 """
-__all__ = ["Instance", "FieldArray", "Batch", "Vocabulary", "DataSet",
+__all__ = ["Instance", "FieldArray", "Batch", "Vocabulary", "DataSet", "Const",
            "Trainer", "Tester", "Callback",
            "Padder", "AutoPadder", "EngChar2DPadder",
            "AccuracyMetric", "BMESF1PreRecMetric", "SpanFPreRecMetric", "SQuADMetric",
diff --git a/fastNLP/core/__init__.py b/fastNLP/core/__init__.py
index 4e5f8527..97afa364 100644
--- a/fastNLP/core/__init__.py
+++ b/fastNLP/core/__init__.py
@@ -13,6 +13,8 @@ core 模块里实现了 fastNLP 的核心框架，常用的组件都可以从 fa
 
 """
 from .batch import Batch
+from .callback import Callback
+from .const import Const
 from .dataset import DataSet
 from .field import FieldArray, Padder, AutoPadder, EngChar2DPadder
 from .instance import Instance
@@ -22,6 +24,5 @@ from .optimizer import Optimizer, SGD, Adam
 from .sampler import SequentialSampler, BucketSampler, RandomSampler, Sampler
 from .tester import Tester
 from .trainer import Trainer
+from .utils import cache_results
 from .vocabulary import Vocabulary
-from .callback import Callback
-from .utils import cache_results
\ No newline at end of file
diff --git a/fastNLP/models/__init__.py b/fastNLP/models/__init__.py
index e8818492..59200773 100644
--- a/fastNLP/models/__init__.py
+++ b/fastNLP/models/__init__.py
@@ -4,7 +4,10 @@ TODO 详细介绍的表格，与主页相对应
 
 """
 from .base_model import BaseModel
+from .bert import BertForMultipleChoice, BertForQuestionAnswering, BertForSequenceClassification, \
+    BertForTokenClassification
 from .biaffine_parser import BiaffineParser, GraphParser
 from .cnn_text_classification import CNNText
 from .sequence_modeling import SeqLabeling, AdvSeqLabel
 from .snli import ESIM
+from .star_transformer import STSeqCls, STNLICls, STSeqLabel
diff --git a/fastNLP/modules/__init__.py b/fastNLP/modules/__init__.py
index f1e4c9e6..4022de9d 100644
--- a/fastNLP/modules/__init__.py
+++ b/fastNLP/modules/__init__.py
@@ -16,4 +16,3 @@ from .dropout import TimestepDropout
 from .encoder import *
 from .utils import get_embeddings
 
-__version__ = '0.0.0'
diff --git a/fastNLP/modules/aggregator/__init__.py b/fastNLP/modules/aggregator/__init__.py
index c1e6cd17..bfb7579b 100644
--- a/fastNLP/modules/aggregator/__init__.py
+++ b/fastNLP/modules/aggregator/__init__.py
@@ -1,7 +1,7 @@
-__all__ = ["MaxPool", "MaxPoolWithMask", "AvgPool", "MultiHeadAttention"]
+__all__ = ["MaxPool", "MaxPoolWithMask", "AvgPool", "MultiHeadAttention", "BiAttention"]
 from .pooling import MaxPool
 from .pooling import MaxPoolWithMask
 from .pooling import AvgPool
 from .pooling import MeanPoolWithMask
 
-from .attention import MultiHeadAttention
+from .attention import MultiHeadAttention, BiAttention
diff --git a/fastNLP/modules/decoder/__init__.py b/fastNLP/modules/decoder/__init__.py
index ec864a14..dc05fce2 100644
--- a/fastNLP/modules/decoder/__init__.py
+++ b/fastNLP/modules/decoder/__init__.py
@@ -1,4 +1,4 @@
-__all__ = ["MLP", "ConditionalRandomField","viterbi_decode"]
+__all__ = ["MLP", "ConditionalRandomField", "viterbi_decode"]
 from .CRF import ConditionalRandomField
 from .MLP import MLP
-from .utils import viterbi_decode
\ No newline at end of file
+from .utils import viterbi_decode