diff --git a/seq2seq/basic_nmt/README.md b/nmt_without_attention/README.md
similarity index 99%
rename from seq2seq/basic_nmt/README.md
rename to nmt_without_attention/README.md
index 9895e0b0d228a4294659ace62a6f1b58edeecbc4..38361bbfbc3e029de872eba967a17453c5e7dac1 100644
--- a/seq2seq/basic_nmt/README.md
+++ b/nmt_without_attention/README.md
@@ -6,7 +6,7 @@
 ## 模型概览
 基于 RNN 的神经网络机器翻译模型遵循编码器－解码器结构，其中的编码器和解码器均是一个循环神经网络。将构成编码器和解码器的两个 RNN 沿时间步展开，得到如下的模型结构图：
 
-<p align="center"><img src="images/Encoder-Decoder.png" width = "90%" align="center"/><br/>图 1. 编码器－解码器框架 </p>
+<p align="center"><img src="images/encoder-decoder.png" width = "90%" align="center"/><br/>图 1. 编码器－解码器框架 </p>
 
 神经机器翻译模型的输入输出可以是字符，也可以是词或者短语。不失一般性，本例以基于词的模型为例说明编码器／解码器的工作机制：
 
diff --git a/seq2seq/basic_nmt/images/bidirectional-encoder.png b/nmt_without_attention/images/bidirectional-encoder.png
similarity index 100%
rename from seq2seq/basic_nmt/images/bidirectional-encoder.png
rename to nmt_without_attention/images/bidirectional-encoder.png
diff --git a/seq2seq/basic_nmt/images/Encoder-Decoder.png b/nmt_without_attention/images/encoder-decoder.png
similarity index 100%
rename from seq2seq/basic_nmt/images/Encoder-Decoder.png
rename to nmt_without_attention/images/encoder-decoder.png
diff --git a/seq2seq/basic_nmt/images/gru.png b/nmt_without_attention/images/gru.png
similarity index 100%
rename from seq2seq/basic_nmt/images/gru.png
rename to nmt_without_attention/images/gru.png
diff --git a/seq2seq/basic_nmt/nmt_without_attention_v2.py b/nmt_without_attention/nmt_without_attention.py
similarity index 100%
rename from seq2seq/basic_nmt/nmt_without_attention_v2.py
rename to nmt_without_attention/nmt_without_attention.py
diff --git a/seq2seq/README.md b/seq2seq/README.md
deleted file mode 100644
index a0990367ef8b03c70c29d285e22ef85907e1d0b7..0000000000000000000000000000000000000000
--- a/seq2seq/README.md
+++ /dev/null
@@ -1 +0,0 @@
-TBD