Python可视化数据分析05、Pandas数据分析

1
2
3
4
5
6
pip3 config set global.index-url https://repo.huaweicloud.com/repository/pypi/simple
pip3 config list
pip3 install --upgrade pip
pip3 install requests
pip3 install pandas

1
2
3
4
5
6
7
#导入Pandas模块中的Series类
from Pandas import Series  
#最简单的Series
Series对象名 = Series(列表)  
#带有标记的索引
Series对象名 = Series(列表, index=索引的列表) 

1
2
3
4
5
6
7
8
9
from pandas import Series

obj1 = Series([1, 3, 5])  # 最简单的Series对象
obj2 = Series([5, 8, 7], index=["a", "b", "c"])  # 带有对各个数据标记的索引
print(obj1)
print(obj2)
print("Series的值为：", obj1.values)
print("Series的索引对象为：", obj1.index)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
from pandas import Series

obj = Series([8, 6, -5, 2], index=["a", "b", "c", "d"])
print(obj["a"])  # 输出8
print(obj["d"])  # 输出2
print(obj[["a", "c", "d"]])  # 选取索引为"a","c","d"对应的值。
print(obj[obj > 0])  # 获取值大于0的数据。
print(obj * 2)  # 输出Series对象中每个数据乘2之后的结果。
print("a" in obj)  # 判断obj对象中是否存在索引值为"a"的数据。
dict = {"a": 23, "b": 12, "c": 8}  # 定义字典对象
print(Series(dict))  # 将dict对象转化为Series对象
print(obj.isnull())  # 检测缺失数据
obj2 = Series([8, 6, -5], index=["a", "b", "e"])  # 再定义一个Series对象
print(obj + obj2)  # 输出两个Series对象的和
obj.name = "population"  # 给obj对象本身的name属性赋值
obj.index.name = "state"  # obj对象的索引的name属性赋值
print(obj)
obj.index = ["f", "g", "k", "m"]
print(obj)

1
2
3
4
5
6
7
8
9
10
from pandas import Series, DataFrame

data = {"name": ["a1", "a2", "a3"], "marks": [200, 400, 800], "price": [9, 3, 7]}
frame = DataFrame(data)
# 将相等长度列表的字典对象转化为DataFrame对象
print(frame)
# 跟Series类似，DataFrame也能自定义索引
frame2 = DataFrame(data, columns=["name", "price", "marks", "dept"], index=["a", "b", "c"])
print(frame2)

1
2
3
4
5
6
7
8
9
10
11
12
from pandas import Series, DataFrame

data = {"name": ["a1", "a2", "a3"], "marks": [200, 400, 800], "price": [9, 3, 7]}
frame3 = DataFrame(data, columns=["name", "price", "marks", "dept"], index=["a", "b", "c"])
print(frame3.columns)
print(frame3["name"])
frame3["dept"] = 90  # 统一给frame3对象的dept列赋值
print(frame3)
dept = Series([2.2, 3.3], index=["a", "c"])
frame3["dept"] = dept
print(frame3)

1
2
3
4
5
6
7
8
9
from pandas import Series, DataFrame

data1 = {"num1": [200, 400, 800], "num2": [60, 300, 60]}
data2 = {"name": ["a1", "a2", "a3"], "marks": [200, 400, 800], "price": [9, 3, 7]}
frame1 = DataFrame(data1)
frame2 = DataFrame(data2)
print(frame1.values)
print(frame2.values)

1
2
3
4
5
6
7
8
9
from pandas import Series

obj = Series(range(3), index=["a", "b", "c"])
# 获取Series的索引对象
index = obj.index
print(index)
# 获取第二位及之后的元素
print(index[1:])

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
from pandas import Series, DataFrame;
import numpy as np

obj = Series(np.arange(4), index=["a", "b", "c", "d"])
print(obj["b"])  # 获取索引值为b的元素
print(obj[1])  # 获取第二个元素
print(obj[2:4])  # 获取第3个元素和第4个元素
print(obj[["a", "d"]])  # 获取索引值为a和d的元素
obj["b":"c"] = 5  # 设置索引值b到c的值为5
print(obj)
frame = DataFrame(np.arange(9).reshape(3, 3), index=["a", "c", "d"],
                  columns=["Ohio", "Texas", "California"])  # DataFrame进行索引
print(frame)
print(frame["Ohio"])  # 获取索引值Ohio的元素
print(frame[["Texas", "California"]])  # 获取索引值Texas和California的元素
print(frame[:2])  # 获取前2个元素
print(frame[frame["California"] > 3])  # 选取索引为California中值大于3的记录
print(frame < 5)  # 通过布尔型DataFrame进行索引
frame[frame < 5] = 0  # 通过布尔型DataFrame进行索引
print(frame)

1
2
3
4
5
6
7
8
9
10
11
12
from pandas import Series, DataFrame;
import numpy as np

s1 = Series([7.2, -6.5, -3.5, 1.6], index=["a", "c", "d", "e"])
s2 = Series([-2, 3, 4, 5, 8], index=["a", "c", "e", "f", "f"])
print(s1 + s2)  # +算术运算
df1 = DataFrame(np.arange(9).reshape(3, 3), columns=list("bcd"), index=["Ohio", "Texas", "Colorado"])
df2 = DataFrame(np.arange(12).reshape(4, 3), columns=list("bde"), index=["Oregon", "Ohio", "Texas", "Colorado"])
print(df1, df2)
print(df1 + df2)  # +算术运算
print(df1.add(df2, fill_value=0))  # add函数

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
from pandas import Series
import numpy as np
from pandas import DataFrame


# 定义普通函数
def func(arg):
    return arg + 1


print(func(1))
mylambda = lambda arg: arg + 1  # 定义函数（lambda表达式）
print(mylambda(123))
frame = DataFrame(np.arange(9).reshape(3, 3), columns=list("bcd"), index=["Ohio", "Texas", "Colorado"])
f = lambda x: x.max() - x.min()  # 定义函数（lambda表达式）
print(frame.apply(f))
# 定义函数（lambda表达式）
sub = lambda x: x - 1
print(frame.applymap(sub))

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
from pandas import Series
import numpy as np
from pandas import DataFrame


# 定义普通函数
def func(arg):
    return arg + 1


print(func(1))
mylambda = lambda arg: arg + 1  # 定义函数（lambda表达式）
print(mylambda(123))
frame = DataFrame(np.arange(9).reshape(3, 3), columns=list("bcd"), index=["Ohio", "Texas", "Colorado"])
f = lambda x: x.max() - x.min()  # 定义函数（lambda表达式）
print(frame.apply(f))
# 定义函数（lambda表达式）
sub = lambda x: x - 1
print(frame.applymap(sub))

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
from pandas import Series
import numpy as np
from pandas import DataFrame

obj = Series(range(4), index=["d", "a", "b", "c"])
print(obj.sort_index())  # sort_index函数
frame = DataFrame(np.arange(9).reshape(3, 3), columns=[4, 6, 5], index=["c", "a", "d"])
print(frame.sort_index())  # sort_index函数
print(frame.sort_index(axis=1))
print(frame.sort_index(axis=1, ascending=False))
frame = DataFrame({"b": [4, -3, 7, 2], "a": [1, 6, 5, 3]})
print(frame.sort_values(by="b"))  # 对"b"这一列进行升序排列
print(frame.sort_values(by=["a", "b"]))  # 同时对两列进行升序排列
obj = Series([7, -2, 4, 3, 12])
print(obj.rank())  # rank()函数
# 根据值在原始数据中出现的顺序给出排名
print(obj.rank(method="first"))

1
2
3
4
5
6
7
8
9
from pandas import Series, DataFrame
import numpy as np

frame = DataFrame({"a": [1, 6, 5, 3], "b": [4, -3, 7, np.nan]})
# 按列进行求和
print(frame.sum())
# 按行进行求和
print(frame.sum(axis=1))

1
2
3
4
5
6
7
8
9
10
11
12
13
14
from pandas import Series, DataFrame

obj = Series(["a", "c", "a", "c", "b", "a", "d", "d"])
uniques = obj.unique()  # 获取Series中的唯一值数组
print(uniques)
uniques.sort()  # 对Series数组进行排序
print(uniques)
# 计算Series数组各值出现的频率
print(obj.value_counts())
# obj各值是否包含于["b","c"]中
mask = obj.isin(["b", "c"])
print(mask)
print(obj[mask])  # 选取Series中数据的子集

1
2
3
4
5
6
7
8
9
10
11
12
13
from pandas import Series, DataFrame
import numpy as np

data = Series(["a", np.nan, "c", "d"])
print(data.isnull())  # 判断是否为空对象
data = Series([1, np.nan, 3, np.nan, 7])
print(data.dropna())  # 滤掉缺失数据
# 通过布尔值索引滤除数据
print(data[data.notnull()])
data = DataFrame([[1, 6, 5], [2, np.nan, np.nan]])
# 滤除DataFrame中的缺失数据
print(data.dropna())