GraphLab Create & SFrame

发表于 | 分类于

1 启动 GraphLab Create

The first time you use GraphLab create, you must enter a product key to license the software for non-commerical academic use. To register for a free one-year academic license and obtain your key, go to dato.com.

1
2
3
4
5
6
7
8
9
import graphlab
# Set product key on this computer. After running this cell, you will not need to re-enter your product key. 
graphlab.product_key.set_product_key('your product key here')

# Limit number of worker processes. This preserves system memory, which prevents hosted notebooks from crashing.
graphlab.set_runtime_config('GRAPHLAB_DEFAULT_NUM_PYLAMBDA_WORKERS', 4)

# Output active product key.
graphlab.product_key.get_product_key()

2 SFrame

  • to obtain the full path of the home directory of IPython notebook. This path is where your files should go.
    1
    2
    import os
    print os.getcwd()
  • Place any files (notebooks and datasets) under the home directory. You may organize your files using sub-folders.
    将所有文件（包括文本文件和数据集）放在家目录下。可以使用子文件夹
  • 用 SFrame 读取表格数据集（use SFrame to load a tabular data set）
    1
    sf = graphlab.SFrame('people-example.csv')

2.1 SFrame basics

  • 显示表格开头几行
    1
    sf # we can view first few lines of table
    或者
    1
    sf.head() # we can view first few lines of table
  • 显示表格最后几行
    1
    sf.tail() # view end of the table

2.2 GraphLab Canvas

将GraphLab Create 中任何一个数据结构可视化

1
2
# 这将会在浏览器中打开一个网页来显示可视化数据
sf.show()

在 IPython Notebook 中打开（而不是在浏览器中打开）（没有成功）

1
2
3
4
5
# .show() visualizes any data structure in GraphLab Create
# If you want Canvas visualization to show up on this notebook, 
# add this line:
graphlab.canvas.set_target('ipynb')
sf['age'].show(view='Categorical') # “view='Categorical'”表示分类排序图

2.3 查看数据集的列

1
2
# 查看“Country”列
sf['Country']

运行结果：

1
2
3
dtype: str
Rows: 7
['United States', 'Canada', 'England', 'USA', 'Poland', 'United States', 'Switzerland']
1
2
# 查看“age”列
sf['age']

运行结果：

1
2
3
dtype: int
Rows: 7
[24L, 23L, 22L, 23L, 23L, 22L, 25L]

一些简单的运算：

  • 平均值
    1
    sf['age'].mean()
    运行结果：
    1
    23.142857142857146
  • 最大值
    1
    sf['age'].max()
    运行结果：
    1
    25L

2.4 建立新的列

在机器学习中，经常会把一些列进行转换，然后成为新的一列，这被叫做特征工程
GraphLabCreate&SFrame_1

GraphLabCreate&SFrame_2

将“age”列都加 2：

1
sf['age']+2

运行结果：

1
2
3
dtype: int
Rows: 7
[26L, 25L, 24L, 25L, 25L, 24L, 27L]

得到“age”列的平方：

1
sf['age'] * sf['age']

运行结果：

1
2
3
dtype: int
Rows: 7
[576L, 529L, 484L, 529L, 529L, 484L, 625L]

2.5 使用函数 apply() 进行数据转换

在前面的表格中，我们发现“Country”列有的值为“United States”，有的为“USA”，这会影响到计数等操作，所以我们希望将所有的“USA”都转换为“United States”

查看“Country”列
GraphLabCreate&SFrame_3
编写函数：
GraphLabCreate&SFrame_4
测试：
GraphLabCreate&SFrame_5
将函数作用在“Country”列的每一行，并将值赋给在“Country”列，查看效果
GraphLabCreate&SFrame_6