From aa14503e6d8d12b5ed10a661f3d8137cd77c8c63 Mon Sep 17 00:00:00 2001
From: Steffy-zxf <48793257+Steffy-zxf@users.noreply.github.com>
Date: Mon, 23 Dec 2019 20:33:11 +0800
Subject: [PATCH] Update README.md

---
 demo/qa_classification/README.md | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/demo/qa_classification/README.md b/demo/qa_classification/README.md
index a4f2a2b8..9b6ed83c 100644
--- a/demo/qa_classification/README.md
+++ b/demo/qa_classification/README.md
@@ -94,6 +94,9 @@ config = hub.RunConfig(use_cuda=True, use_data_parallel=True, use_pyreader=True,
 ```
 
 #### 优化策略
+
+PaddleHub提供了许多优化策略，如`AdamWeightDecayStrategy`、`ULMFiTStrategy`、`DefaultFinetuneStrategy`等，详细信息参见[策略](https://github.com/PaddlePaddle/PaddleHub/wiki/PaddleHub-API:-Strategy)
+
 针对ERNIE与BERT类任务，PaddleHub封装了适合这一任务的迁移学习优化策略`AdamWeightDecayStrategy`
 
 * `learning_rate`: Finetune过程中的最大学习率;
-- 
GitLab