update readme with wechat.

README.md

@@ -154,7 +154,7 @@ python preprocess.py
 python train.py
 ```
 训练过程截图：
-![train image](https://github.com/shibing624/pycorrector/blob/master/pycorrector/data/git_image/seq2seq_train.png)
+![train image](./docs/git_image/seq2seq_train.png)
 
 
 #### 预测
@@ -220,7 +220,7 @@ input: 由我起开始做 output: 由我开始做
 
 
 ## 讨论群
-![image](https://user-images.githubusercontent.com/12003487/69050479-17fd9900-0a3d-11ea-8788-7a9898db2380.png)
+![image](./docs/git_image/erweima.png)
 
 微信交流群，感兴趣的同学可以加入沟通NLP文本纠错相关技术，issues上回复不及时也可以在群里面提问。
 

pycorrector/data/bert_models/finetuned_chinese_lm/bert_config.json

@@ -0,0 +1,20 @@
+{
+  "attention_probs_dropout_prob": 0.1,
+  "directionality": "bidi",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pooler_fc_size": 768,
+  "pooler_num_attention_heads": 12,
+  "pooler_num_fc_layers": 3,
+  "pooler_size_per_head": 128,
+  "pooler_type": "first_token_transform",
+  "type_vocab_size": 2,
+  "vocab_size": 21128
+}