2nd commit

zsdonghao · Oct 7, 2016 · 6e42c6e · 6e42c6e
1 parent dfb8cd7
commit 6e42c6e
Showing 1 changed file with 7 additions and 4 deletions.
diff --git a/README.md b/README.md
@@ -3,13 +3,16 @@
 We reimplemented the complicated [Google' Image Captioning](https://github.com/tensorflow/models/tree/master/im2txt) model by simple TensorLayer code.
 
 ### 1. Prepare MSCOCO data and Inception model
-Go to ``data`` folder.
+- Creat a ``data`` folder.
+- Download and Preprocessing MSCOCO Data [click here](https://github.com/zsdonghao/models/tree/master/im2txt#prepare-the-training-data)
+- Download the Inception_V3 CKPT [click here](https://github.com/zsdonghao/models/tree/master/im2txt#download-the-inception-v3-checkpoint)
+
 
 ### 2. Train the model
-via ``train.py``
+- via ``train.py``
 
 ### 3. Evaluate the model
-via ``evaluate.py``
+- via ``evaluate.py``
 
 ### 4. Generate captions by given image and model
-via ``run_inference.py``
+- via ``run_inference.py``