diff --git a/paddle2.0_docs/predict_gold_price/predict_gold_price.ipynb b/paddle2.0_docs/predict_gold_price/predict_gold_price.ipynb
new file mode 100644
index 00000000..63c45293
--- /dev/null
+++ b/paddle2.0_docs/predict_gold_price/predict_gold_price.ipynb
@@ -0,0 +1,1253 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "collapsed": false
+ },
+ "source": [
+ "## 标题：**使用PaddlePaddle进行黄金预测**\n",
+ "\n",
+ "# 前言\n",
+ "\n",
+ "能预测黄金价格走势曲线。\n",
+ "\n",
+ "国内黄金价格和国际黄金价格曲线基本一致。\n",
+ "\n",
+ "# 本模型效果图（亮点）\n",
+ "\n",
+ "\n",
+ "# 目标\n",
+ "只需预测未来黄金价格曲线图，因为只需知道价格是上升还是下降，就有很大意义。\n",
+ "不用将价格预测准确，只需要将趋势图预测准确就可以。\n",
+ "\n",
+ "# 个人方案\n",
+ "请大家把更新后的代码导出后，（点击右上角：文件-》导出为ipynb）\n",
+ "pull request到我github的这个文件:\n",
+ "[https://github.com/guojiahuiEmily/predictgoldprice/blob/main/predictgold.ipynb](https://github.com/guojiahuiEmily/predictgoldprice/blob/main/predictgold.ipynb)\n",
+ "\n",
+ "\n",
+ "# 分享成绩\n",
+ "并把自己的预测效果及截图分享到这里，看看谁的效果好：\n",
+ "\n",
+ "[https://github.com/guojiahuiEmily/predictgoldprice/blob/main/rank](https://github.com/guojiahuiEmily/predictgoldprice/blob/main/rank)\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "collapsed": false
+ },
+ "source": [
+ "# 一、灵感来源\n",
+ " 黄金价格对很多事件都有指导价值。\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "collapsed": false
+ },
+ "source": [
+ "# 二、模型应用\n",
+ "\n",
+ "# 时间卷积网络（TCN，Temporal Convolutional Networks）\n",
+ "时间序列是指按照时间先后顺序排列而成的序列，例如每日发电量、每小时营业额等组成的序列。通过分析时间序列中的发展过程、方向和趋势，我们可以预测下一段时间可能出现的情况。在本例中，我们使用时间卷积网络TCN进行建模，将学习到的特征接入全连接层完成预测。TCN的网络如下所示：\n",
+ "\n",
+ "![](\"https://ai-studio-static-online.cdn.bcebos.com/214011346cb24535ad57c50b936a2e4a207da6f7ce3b4e4babbe2571003f1aee\")
\n",
+ "
图1：TCN示意图\n",
+ "\n",
+ "图中是一个filters number=3, dilated rate=1的时间卷积网络，它能够学习前T个时序的数据特征。关于TCN更详细的资料请参考论文：[An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling](https://arxiv.org/pdf/1803.01271.pdf)。\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "collapsed": false
+ },
+ "source": [
+ "# 三、代码展示\n",
+ "\n",
+ "## 1、准备环境\n",
+ "\n",
+ "我们首先需要导入必要的包。\n",
+ "\n",
+ "这里我们使用`paddlenlp.seq2vec`中内置好的模型，关于`seq2vec`的详细介绍可参考这个项目：[seq2vec是什么? ](https://aistudio.baidu.com/aistudio/projectdetail/1283423)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [],
+ "source": [
+ "!pip install paddlenlp>=2.0.0b -i https://pypi.org/simple"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "collapsed": false
+ },
+ "source": [
+ "如果上面运行失败可执行下面"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [],
+ "source": [
+ "# !pip install --upgrade setuptools && python -m pip install --upgrade pip\n",
+ "# !pip install paddlenlp==2.0.0rc"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [],
+ "source": [
+ "\n",
+ "\n",
+ "import os\n",
+ "import sys\n",
+ "\n",
+ "import paddle\n",
+ "import paddle.nn as nn\n",
+ "import numpy as np\n",
+ "\n",
+ "import pandas as pd\n",
+ "import seaborn as sns\n",
+ "from pylab import rcParams\n",
+ "import matplotlib.pyplot as plt\n",
+ "from matplotlib import rc\n",
+ "from sklearn.preprocessing import MinMaxScaler\n",
+ "from pandas.plotting import register_matplotlib_converters\n",
+ "\n",
+ "sys.path.append(os.path.abspath(os.path.join(os.getcwd(), \"../..\")))\n",
+ "from paddlenlp.seq2vec import TCNEncoder"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "collapsed": false
+ },
+ "source": [
+ "为了更好地展示数据结果，我们在这里配置画图功能。"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [],
+ "source": [
+ "# config matplotlib\n",
+ "%matplotlib inline\n",
+ "%config InlineBackend.figure_format='retina'\n",
+ "sns.set(style='whitegrid', palette='muted', font_scale=1.2)\n",
+ "HAPPY_COLORS_PALETTE = [\"#01BEFE\", \"#FFDD00\", \"#FF7D00\", \"#FF006D\", \"#93D30C\", \"#8F00FF\"]\n",
+ "sns.set_palette(sns.color_palette(HAPPY_COLORS_PALETTE))\n",
+ "rcParams['figure.figsize'] = 14, 10\n",
+ "register_matplotlib_converters()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "collapsed": false
+ },
+ "source": [
+ "## 2、数据\n",
+ "\n",
+ "csv文件：https://github.com/guojiahuiEmily/predictgoldprice/blob/main/train.csv"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "collapsed": false
+ },
+ "source": [
+ "## 3、数据预览\n",
+ "\n",
+ "数据集中包含了国际每日黄金价格，单位：美元。"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " | \n",
+ " days | \n",
+ " price | \n",
+ "
\n",
+ " \n",
+ " \n",
+ " \n",
+ " | 0 | \n",
+ " 1978/12/29 | \n",
+ " 226 | \n",
+ "
\n",
+ " \n",
+ " | 1 | \n",
+ " 1979/1/1 | \n",
+ " 226 | \n",
+ "
\n",
+ " \n",
+ " | 2 | \n",
+ " 1979/1/2 | \n",
+ " 227 | \n",
+ "
\n",
+ " \n",
+ " | 3 | \n",
+ " 1979/1/3 | \n",
+ " 219 | \n",
+ "
\n",
+ " \n",
+ " | 4 | \n",
+ " 1979/1/4 | \n",
+ " 223 | \n",
+ "
\n",
+ " \n",
+ "
\n",
+ "
\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " | \n",
+ " price | \n",
+ "
\n",
+ " \n",
+ " | days | \n",
+ " | \n",
+ "
\n",
+ " \n",
+ " \n",
+ " \n",
+ " | 1978-12-29 | \n",
+ " 226 | \n",
+ "
\n",
+ " \n",
+ " | 1979-01-01 | \n",
+ " 226 | \n",
+ "
\n",
+ " \n",
+ " | 1979-01-02 | \n",
+ " 227 | \n",
+ "
\n",
+ " \n",
+ " | 1979-01-03 | \n",
+ " 219 | \n",
+ "
\n",
+ " \n",
+ " | 1979-01-04 | \n",
+ " 223 | \n",
+ "
\n",
+ " \n",
+ "
\n",
+ "
"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "plt.figure(figsize=(6,6))\n",
+ "plt.plot(daily_cases)\n",
+ "plt.title(\"daily\");"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "collapsed": false
+ },
+ "source": [
+ "## 4、数据预处理\n",
+ "\n",
+ "首先划分数据集为训练集与验证集，我们取最后30天的数据作为测试集，其余作为训练集。"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "The number of the samples in train set is : 10951\n"
+ ]
+ }
+ ],
+ "source": [
+ "TEST_DATA_SIZE = 30\n",
+ "\n",
+ "train_data = daily_cases[:-TEST_DATA_SIZE].astype('float32')\n",
+ "test_data = daily_cases[-TEST_DATA_SIZE:].astype('float32')\n",
+ "\n",
+ "# train_data = train_data.astype('float32')\n",
+ "\n",
+ "print(\"The number of the samples in train set is : %i\"%train_data.shape[0])\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "collapsed": false
+ },
+ "source": [
+ "为了提升模型收敛速度与性能，我们使用scikit-learn进行数据归一化。"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "The number of the samples in train set is : 10951\n"
+ ]
+ },
+ {
+ "data": {
+ "text/html": [
+ "\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " | \n",
+ " price | \n",
+ "
\n",
+ " \n",
+ " | days | \n",
+ " | \n",
+ "
\n",
+ " \n",
+ " \n",
+ " \n",
+ " | 1978-12-29 | \n",
+ " 226.0 | \n",
+ "
\n",
+ " \n",
+ " | 1979-01-01 | \n",
+ " 226.0 | \n",
+ "
\n",
+ " \n",
+ " | 1979-01-02 | \n",
+ " 227.0 | \n",
+ "
\n",
+ " \n",
+ " | 1979-01-03 | \n",
+ " 219.0 | \n",
+ "
\n",
+ " \n",
+ " | 1979-01-04 | \n",
+ " 223.0 | \n",
+ "
\n",
+ " \n",
+ "
\n",
+ "
\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " | \n",
+ " price | \n",
+ "
\n",
+ " \n",
+ " | days | \n",
+ " | \n",
+ "
\n",
+ " \n",
+ " \n",
+ " \n",
+ " | 2020-12-21 | \n",
+ " 1880.0 | \n",
+ "
\n",
+ " \n",
+ " | 2020-12-22 | \n",
+ " 1877.0 | \n",
+ "
\n",
+ " \n",
+ " | 2020-12-23 | \n",
+ " 1875.0 | \n",
+ "
\n",
+ " \n",
+ " | 2020-12-24 | \n",
+ " 1875.0 | \n",
+ "
\n",
+ " \n",
+ " | 2020-12-25 | \n",
+ " 1875.0 | \n",
+ "
\n",
+ " \n",
+ "
\n",
+ "