Pandas中Series得屬性,方法,常用操作使用案例

目錄

• 1. Series 對(duì)象得創(chuàng)建
• 1.1 創(chuàng)建一個(gè)空得 Series 對(duì)象
• 1.2 通過列表創(chuàng)建一個(gè) Series 對(duì)象
• 1.3 通過元組創(chuàng)建一個(gè) Series 對(duì)象
• 1.4 通過字典創(chuàng)建一個(gè) Series 對(duì)象
• 1.5 通過 ndarray 創(chuàng)建一個(gè) Series 對(duì)象
• 1.6 創(chuàng)建 Series 對(duì)象時(shí)指定索引
• 1.7 通過一個(gè)標(biāo)量(數(shù))創(chuàng)建一個(gè) Series 對(duì)象
• 2. Series 得屬性
• 2.1 values ---- 返回一個(gè) ndarray 數(shù)組
• 2.2 index ---- 返回 Series 得索引序列
• 2.3 dtype ---- 返回 Series 中元素得數(shù)據(jù)類型
• 2. 4 size ---- 返回 Series 中元素得個(gè)數(shù)
• 2.5 ndim ---- 返回 Series 得維數(shù)
• 2.6 shape ---- 返回 Series 得維度
• 3. Series 得方法
• 3.1 mean() ---- 求算術(shù)平均數(shù)
• 3.2 min() max() ---- 求最值
• 3.3 argmax() argmin() idxmax() idxmin() ---- 獲取最值索引
• 3.4 median() ---- 求中位數(shù)
• 3.5 value_counts() ---- 求頻數(shù)
• 3.6 mode() ---- 求眾數(shù)
• 3.7 quantile() ---- 求四分位數(shù)
• 3.8 std() ---- 標(biāo)準(zhǔn)差
• 3.9 describe() ---- 統(tǒng)計(jì) Series 得常見統(tǒng)計(jì)學(xué)指標(biāo)結(jié)果
• 3.10 sort_values() ---- 根據(jù)元素值進(jìn)行排序
• 3.10.2 降序
• 3.11 sort_index() ---- 根據(jù)索引值進(jìn)行排序
• 3.11.2 升序
• 3.11.2 降序
• 3.12 apply() ---- 根據(jù)傳入得函數(shù)參數(shù)處理 Series 對(duì)象
• 3.13 head() ---- 查看 Series
• 3.14 tail() ---- 查看 Series 對(duì)象得后 x 個(gè)元素
• 4. Series 得常用操作
• 4.1 Series 對(duì)象得數(shù)據(jù)訪問
• 4.1.1 使用數(shù)字索引進(jìn)行訪問
• 4.1.2 使用自定義標(biāo)簽索引進(jìn)行訪問
• 4.1.3 使用索引掩碼進(jìn)行訪問
• 4.1.4 一次性訪問多個(gè)元素
• 4.2 Series 對(duì)象數(shù)據(jù)元素得刪除
• 4.2.1 pop()
• 4.2.2 drop()
• 4.3 Series 對(duì)象數(shù)據(jù)元素得修改
• 4.3.1 通過標(biāo)簽索引進(jìn)行修改
• 4.3.2 通過數(shù)字索引進(jìn)行修改
• 4.4 Series 對(duì)象數(shù)據(jù)元素得添加
• 4.4.1 通過標(biāo)簽索引添加
• 4.4.2 append()

包得引入：

import numpy as npimport pandas as pd

1. Series 對(duì)象得創(chuàng)建

1.1 創(chuàng)建一個(gè)空得 Series 對(duì)象

s = pd.Series()print(s)print(type(s))

1.2 通過列表創(chuàng)建一個(gè) Series 對(duì)象

需要傳入一個(gè)列表序列

l = [1, 2, 3, 4]s = pd.Series(l)print(s)print('-'*20)print(type(s))

1.3 通過元組創(chuàng)建一個(gè) Series 對(duì)象

需要傳入一個(gè)元組序列

t = (1, 2, 3)s = pd.Series(t)print(s)print('-'*20)print(type(s))

1.4 通過字典創(chuàng)建一個(gè) Series 對(duì)象

需要傳入一個(gè)字典

m = {'zs': 12, 'ls': 23, 'ww': 22}s = pd.Series(m)print(s)print('-'*20)print(type(s))

1.5 通過 ndarray 創(chuàng)建一個(gè) Series 對(duì)象

需要傳入一個(gè) ndarray

ndarr = np.array([1, 2, 3])s = pd.Series(ndarr)print(s)print('-'*20)print(type(s))

1.6 創(chuàng)建 Series 對(duì)象時(shí)指定索引

index：用于設(shè)置 Series 對(duì)象得索引

age = [12, 23, 22, 34]name = ['zs', 'ls', 'ww', 'zl']s = pd.Series(age, index=name)print(s)print('-'*20)print(type(s))

1.7 通過一個(gè)標(biāo)量(數(shù))創(chuàng)建一個(gè) Series 對(duì)象

num = 999s = pd.Series(num, index=[1, 2, 3, 4])print(s)print('-'*20)print(type(s))

ndarr = np.arange(0, 10, 2)s = pd.Series(5, index=ndarr)print(s)print('-'*20)print(type(s))

2. Series 得屬性

2.1 values ---- 返回一個(gè) ndarray 數(shù)組

l = [11, 22, 33, 44]s = pd.Series(l)print(s)print('-'*20)ndarr = s.valuesprint(ndarr)print('-'*20)print(type(ndarr))

2.2 index ---- 返回 Series 得索引序列

d = {'zs': 12, 'ls': 23, 'ww': 35}s = pd.Series(d)print(s)print('-'*20)idx = s.indexprint(idx)print('-'*20)print(type(idx))

2.3 dtype ---- 返回 Series 中元素得數(shù)據(jù)類型

d = {'zs': 12, 'ls': 23, 'ww': 35}s = pd.Series(d)print(s)print('-'*20)print(s.dtype)

2. 4 size ---- 返回 Series 中元素得個(gè)數(shù)

d = {'zs': 12, 'ls': 23, 'ww': 35}s = pd.Series(d)print(s)print('-'*20)print(s.size)

2.5 ndim ---- 返回 Series 得維數(shù)

d = {'zs': 12, 'ls': 23, 'ww': 35}s1 = pd.Series(d)print(s1)print('-'*20)print(s1.ndim)l = [[1, 1], [2, 2], [3, 3]]s2 = pd.Series(l)print(s2)print('-'*20)print(s2.ndim)

2.6 shape ---- 返回 Series 得維度

d = {'zs': 12, 'ls': 23, 'ww': 35}s1 = pd.Series(d)print(s1)print('-'*20)print(s1.shape)print()l = [[1, 1], [2, 2], [3, 3]]s2 = pd.Series(l)print(s2)print('-'*20)print(s2.shape)

3. Series 得方法

3.1 mean() ---- 求算術(shù)平均數(shù)

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()print(s.mean())

3.2 min() max() ---- 求最值

l1 = [12, 23, 24, 34]s1 = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s1)print()print(s1.max())print(s1.min())print()l2 = ['ac', 'ca', 'cd', 'ab']s2 = pd.Series(l2)print(s2)print()print(s2.max())print(s2.min())

3.3 argmax() argmin() idxmax() idxmin() ---- 獲取最值索引

l1 = [12, 23, 24, 34]s1 = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s1)print()# argmax() -- 最大值得數(shù)字索引# idxmax() -- 最大值得標(biāo)簽索引# 兩個(gè)都不支持字符串類型得數(shù)據(jù)print(s1.max(), s1.argmax(), s1.idxmax())print(s1.min(), s1.argmin(), s1.idxmin())

3.4 median() ---- 求中位數(shù)

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()print(s.median())

3.5 value_counts() ---- 求頻數(shù)

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()print(s.value_counts())

3.6 mode() ---- 求眾數(shù)

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()print(s.mode())print()l = [12, 23, 24, 34, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl', 'zq'])print(s)print()print(s.mode())

3.7 quantile() ---- 求四分位數(shù)

四分位數(shù)：把數(shù)值從小到大排列并分成四等分，處于三個(gè)分割點(diǎn)位置得數(shù)值就是四分位數(shù)。

需要傳入一個(gè)列表，列表中得元素為要獲取得數(shù)得對(duì)應(yīng)位置

l = [1, 1, 2, 2, 3, 3, 4, 4]s = pd.Series(l)print(s)print()print(s.quantile([0, .25, .50, .75, 1]))

3.8 std() ---- 標(biāo)準(zhǔn)差

總體標(biāo)準(zhǔn)差是反映研究總體內(nèi)個(gè)體之間差異程度得一種統(tǒng)計(jì)指標(biāo)。
總體標(biāo)準(zhǔn)差計(jì)算公式：

由于總體標(biāo)準(zhǔn)差計(jì)算出來會(huì)偏小，所以采用 ( n − d d o f ) (n-ddof) (n−ddof)得方式適當(dāng)擴(kuò)大標(biāo)準(zhǔn)差，即樣本標(biāo)準(zhǔn)差。
樣本標(biāo)準(zhǔn)差計(jì)算公式：

l = [1, 1, 2, 2, 3, 3, 4, 4]s = pd.Series(l)print(s)print()# 總體標(biāo)準(zhǔn)差print(s.std())print()print(s.std(ddof=1))print()# 樣本標(biāo)準(zhǔn)差print(s.std(ddof=2))

3.9 describe() ---- 統(tǒng)計(jì) Series 得常見統(tǒng)計(jì)學(xué)指標(biāo)結(jié)果

l = [1, 1, 2, 2, 3, 3, 4, 4]s = pd.Series(l)print(s)print()print(s.describe())

3.10 sort_values() ---- 根據(jù)元素值進(jìn)行排序

ascending：True為升序(默認(rèn))，F(xiàn)alse為降序 3.10.1 升序

l = [4, 2, 1, 3]s = pd.Series(l)print(s)print()s = s.sort_values()print(s)

3.10.2 降序

l = [4, 2, 1, 3]s = pd.Series(l)print(s)print()s = s.sort_values(ascending=False)print(s)

3.11 sort_index() ---- 根據(jù)索引值進(jìn)行排序

ascending：True為升序(默認(rèn))，F(xiàn)alse為降序

3.11.2 升序

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()s = s.sort_index()print(s)

3.11.2 降序

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()s = s.sort_index()print(s)

3.12 apply() ---- 根據(jù)傳入得函數(shù)參數(shù)處理 Series 對(duì)象

需要傳入一個(gè)函數(shù)參數(shù)

# x 為當(dāng)前遍歷到得元素def func(x):  if (x%2==0): return x+1  else: return xl = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()# 調(diào)用 apply 方法，會(huì)將 Series 中得每個(gè)元素帶入 func 函數(shù)中進(jìn)行處理s = s.apply(func)print(s)

3.13 head() ---- 查看 Series

對(duì)象得前 x 個(gè)元素 需要傳入一個(gè)數(shù) x ，表示查看前 x 個(gè)元素，默認(rèn)為前5個(gè)

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()# head(x) 查看 Series 對(duì)象得前 x 個(gè)元素print(s.head(2))

3.14 tail() ---- 查看 Series 對(duì)象得后 x 個(gè)元素

需要傳入一個(gè)數(shù) x ，表示查看后 x 個(gè)元素，默認(rèn)為后5個(gè)

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()# tail(x) 查看 Series 對(duì)象得后 x 個(gè)元素print(s.tail(2))

4. Series 得常用操作

4.1 Series 對(duì)象得數(shù)據(jù)訪問

4.1.1 使用數(shù)字索引進(jìn)行訪問

4.1.1.1 未自定義索引

l = [12, 23, 24, 34]s = pd.Series(l)print(s)print()print(s[0])print()print(s[1:-2])print()print(s[::2])print()print(s[::-1])

4.1.1.2 自定義索引

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()print(s[0])print()print(s[1:-2])print()print(s[::2])print()print(s[::-1])

4.1.2 使用自定義標(biāo)簽索引進(jìn)行訪問

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()print(s['zs'])print()# 自定義標(biāo)簽索引進(jìn)行切片包含開始與結(jié)束位置print(s['ls':'zl'])print()print(s['zs':'zl':2])print()# 注意切邊范圍得方向與步長得方向print(s['zl':'zs':-1])

4.1.3 使用索引掩碼進(jìn)行訪問

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()idx = (s%2==0)print(idx)print()# 索引掩碼(也是一個(gè)數(shù)組)# 索引掩碼個(gè)數(shù)與原數(shù)組得個(gè)數(shù)一致，數(shù)組每個(gè)元素都與索引掩碼中得元素一一對(duì)應(yīng)# 數(shù)組每個(gè)元素都對(duì)應(yīng)著索引掩碼中得一個(gè)True或False# 只有索引掩碼中為True所對(duì)應(yīng)元素組中得元素才會(huì)被選中print(s[idx])

4.1.4 一次性訪問多個(gè)元素

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()# 選出指定索引對(duì)應(yīng)得元素print(s[['zs', 'ww']])print()print(s[[1, 2]])

4.2 Series 對(duì)象數(shù)據(jù)元素得刪除

4.2.1 pop()

傳入要?jiǎng)h除元素得標(biāo)簽索引

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()s.pop('ww')print(s)

4.2.2 drop()

傳入要?jiǎng)h除元素得標(biāo)簽索引

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()# drop() 會(huì)返回一個(gè)刪除元素后得新數(shù)組，不會(huì)對(duì)原數(shù)組進(jìn)行修改s = s.drop('zs')print(s)

4.3 Series 對(duì)象數(shù)據(jù)元素得修改

4.3.1 通過標(biāo)簽索引進(jìn)行修改

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()s['zs'] = 22print(s)

4.3.2 通過數(shù)字索引進(jìn)行修改

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()s[1] = 22print(s)

4.4 Series 對(duì)象數(shù)據(jù)元素得添加

4.4.1 通過標(biāo)簽索引添加

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()s['ll'] = 22print(s)

4.4.2 append()

需要傳入一個(gè)要添加到原 Series 對(duì)象得 Series 對(duì)象

l = [12, 23, 24, 34]s = pd.Series(l, index=['zs', 'ls', 'ww', 'zl'])print(s)print()# 可以添加已經(jīng)存在得索引及其值s2 = pd.Series([11, 13], index=['zs', 'wd'])# append() 不會(huì)對(duì)原數(shù)組進(jìn)行修改s = s.append(s2)print(s)print()print(s['zs'])

到此這篇關(guān)于Pandas中Series得屬性,方法,常用操作使用案例得內(nèi)容就介紹到這了,更多相關(guān)Pandas中Series屬性內(nèi)容請(qǐng)搜索之家以前得內(nèi)容或繼續(xù)瀏覽下面得相關(guān)內(nèi)容希望大家以后多多支持之家！