From 282aa96731d2a23f93f8f0f26a5215fbe809a1a5 Mon Sep 17 00:00:00 2001
From: Yancey1989 <yancey1989@gmail.com>
Date: Thu, 15 Mar 2018 18:53:39 +0800
Subject: [PATCH] modify the picture size

---
 doc/fluid/design/dist_train/prefetch_parameter.md | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/doc/fluid/design/dist_train/prefetch_parameter.md b/doc/fluid/design/dist_train/prefetch_parameter.md
index 952d2bada92..b3e2210fdff 100644
--- a/doc/fluid/design/dist_train/prefetch_parameter.md
+++ b/doc/fluid/design/dist_train/prefetch_parameter.md
@@ -16,13 +16,13 @@ Prior to reading this design, it would be useful for the reader to make themselv
 
 The execution of `lookup local table` is as follows:
 
-<img src="src/lookup_local_table.png" width="400" />
+<img src="src/lookup_local_table.png" width="700" />
 
 For some cases, the parameter(`weight`) may be very large, such as 10 billion features, the entire
 data could not be stored in one trainer's memory, so we need to partition this parameter and
 pre-fetch it at the beginning of each mini-batch, and we call it `lookup remote table`:
 
-<img src="src/lookup_remote_table.png" width="400">
+<img src="src/lookup_remote_table.png" width="700" />
 
 The processing flow of `lookup remote table` is as follows:
 
-- 
GitLab