From e6645df1912528d355ec71f293e62b72527af141 Mon Sep 17 00:00:00 2001
From: Yaguang <liyaguang0123@gmail.com>
Date: Mon, 1 Oct 2018 09:59:47 -0700
Subject: [PATCH] Update README.

---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index 1351ae3..9b01e33 100644
--- a/README.md
+++ b/README.md
@@ -43,7 +43,7 @@ The generated prediction of DCRNN is in `data/results/dcrnn_predictions_[1-12].h
 ```bash
 python dcrnn_train.py --config_filename=data/model/dcrnn_config.yaml
 ```
-Each epoch takes about 5min with a single GTX 1080 Ti. There is a chance that train/val loss will explode,  gradient explosion, 
+Each epoch takes about 5min with a single GTX 1080 Ti.
 
 ## Graph Construction
  As the currently implementation is based on pre-calculated road network distances between sensors, it currently only