Uno, Numpy
1.1 Crear
# 类型转换方式创建
a = np.array([1,2,3]) # array([1, 2, 3])
b = np.array([[1,2,3],[4,5,6],[7,8,9]])
'''
array([[1, 2, 3],
[4, 5, 6],
[7, 8, 9]])
'''np.random.seed(666) # 指定随机种子，增加随机生成时的可重复性
# ------arange方法
np.arange(6) # [0 1 2 3 4 5]
np.arange(2,6,1) # [2 3 4 5]
# ------randint方法
np.random.randint(0,10) # 1
np.random.randint(1,3,8) # array([2, 2, 1, 1, 2, 1, 2, 1])
np.random.randint(0,10,size=(3,5))
'''
array([[4, 9, 3, 2, 9],
[3, 7, 5, 9, 7],
[9, 5, 5, 0, 4]])
'''
# ------random方法
np.random.random() # 0.7004371218578347
np.random.random((3,5))
'''
array([[0.84418664, 0.67651434, 0.72785806, 0.95145796, 0.0127032 ],
[0.4135877 , 0.04881279, 0.09992856, 0.50806631, 0.20024754],
[0.74415417, 0.192892 , 0.70084475, 0.29322811, 0.77447945]])
'''
np.random.normal() # 均值为0，方差为1的浮点数
np.random.normal(0,1,(3,5)) # 均值为0方差为1，三行五列的随机矩阵
'''
array([[-0.57577075, -1.68290077, 0.22918525, -1.75662522, 0.84463262],
[ 0.27721986, 0.85290153, 0.1945996 , 1.31063772, 1.5438436 ],
[-0.52904802, -0.6564723 , -0.2015057 , -0.70061583, 0.68713795]])
'''
# ------linspace方法
np.linspace(2,5,4) # 初始值为2，终值为5，元素个数为4个的等差数组 [2. 3. 4. 5.]
# ------zeros方法
np.zeros(5) # [0. 0. 0. 0. 0.]
np.zeros(10, dtype=int) # array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0])
np.zeros(shape=(3, 5),dtype=int )
'''
array([[0, 0, 0, 0, 0],
[0, 0, 0, 0, 0],
[0, 0, 0, 0, 0]])
'''
# ------ones方法
np.ones(10) # array([1., 1., 1., 1., 1., 1., 1., 1., 1., 1.])
np.ones((3, 5 ))
'''
array([[1., 1., 1., 1., 1.],
[1., 1., 1., 1., 1.],
[1., 1., 1., 1., 1.]])
'''
# ------full方法
np.full(shape=(3,5),fill_value=666)
'''
array([[666, 666, 666, 666, 666],
[666, 666, 666, 666, 666],
[666, 666, 666, 666, 666]])
'''1.2 Propiedades
x = np.arange(10)
x # array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])
X = np.arange(15).reshape(3,5)
X
'''
array([[ 0, 1, 2, 3, 4],
[ 5, 6, 7, 8, 9],
[10, 11, 12, 13, 14]])
'''
# ------ndim查询是几维数组-----
x.ndim # 1
X.ndim # 2
# ------shape查询元素形状，返回元组
x.shape # (10,)
X.shape # (3, 5)
# -----size查询元素个数------
x.size # 10
X.size # 151.3 Acceso a datos
# ------索引访问------
x[0] # 0
x[-1] # 9
X[0][0] # 0 不建议使用
X[(2,4)] # 14 访问第2行第4个元素
X[2,4] # 14 同上
# ------切片访问------
x[0:5] # array([0, 1, 2, 3, 4]) 从0开始到5，不包含5
x[:3] # array([0, 1, 2]) 从头开始到3的位置
x[:3] # array([3, 4, 5, 6, 7, 8, 9]) 从3的位置到最后
x[::3] # array([0, 3, 6, 9]) 从头到尾，步长为3
x[::-1] # array([9, 8, 7, 6, 5, 4, 3, 2, 1, 0]) 倒着访问
# ------切片访问------
X[:2,:3] # 访问前两行的前三列
'''
array([[0, 1, 2],
[5, 6, 7]])
'''
X[::-1,::-1]
'''
array([[14, 13, 12, 11, 10],
[ 9, 8, 7, 6, 5],
[ 4, 3, 2, 1, 0]])
'''
X[0] # array([0, 1, 2, 3, 4]) 取一行
X[0,:] # 同上
X[:,1] # array([ 1, 6, 11]) 取一列1.4 Modificación
a = np.array([1,2,3,4,5])
np.append(a,7) # array([1, 2, 3, 4, 5, 7])
np.insert(a,0,0) # array([0, 1, 2, 3, 4, 5])
np.delete(a,3) # array([1, 2, 3, 5])
a[0] = 666
a # array([666, 2, 3, 4, 5])
a[2:4] = [999,888]
a # array([666, 2, 999, 888, 5])x.ndim # 1 维度
x.reshape(10,-1) # 改变成十行，列自动确定
'''
array([[0],
[1],
[2],
[3],
[4],
[5],
[6],
[7],
[8],
[9]])
'''
x1 = x.reshape(-1,10) # array([[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]])
x1.ndim # 2 维度
a = np.array([[1,2,3],[4,5,6]])
'''
array([[1, 2, 3],
[4, 5, 6]])
'''
a.reshape(3,2)
'''
array([[1, 2],
[3, 4],
[5, 6]])
'''
a.reshape(1,-1) # array([[1, 2, 3, 4, 5, 6]])
a.reshape(-1) # array([1, 2, 3, 4, 5, 6])a = np.array([1,2,3,4,5],dtype=np.int64)
a.dtype # dtype('int64')
a = a.astype(np.float32)
a.dtype # dtype('float32')1.5 Fusión
# ------一维------
x = np.array([1,2,3])
y = np.array([4,5,6])
np.concatenate([x,y]) # array([1, 2, 3, 4, 5, 6])
# ------二维------
a = np.array([[1,2,3],[4,5,6]])
'''
array([[1, 2, 3],
[4, 5, 6]])
'''
np.concatenate([a,a])
'''
array([[1, 2, 3],
[4, 5, 6],
[1, 2, 3],
[4, 5, 6]])
'''
np.concatenate([a,a],axis=1)
'''
array([[1, 2, 3, 1, 2, 3],
[4, 5, 6, 4, 5, 6]])
'''
# ------一维、二维拼接------
b = np.array([666,888,999])
np.concatenate([a, b.reshape(1,-1)])
'''
array([[ 1, 2, 3],
[ 4, 5, 6],
[666, 888, 999]])
'''
np.vstack([a,b])
'''
array([[ 1, 2, 3],
[ 4, 5, 6],
[666, 888, 999]])
'''
np.hstack([a,a])
'''
array([[1, 2, 3, 1, 2, 3],
[4, 5, 6, 4, 5, 6]])
'''1.6 dividir
# ------一维
a = np.arange(10)
np.split(a,[3,8]) # [array([0, 1, 2]), array([3, 4, 5, 6, 7]), array([8, 9])]
np.split(a,[6]) # [array([0, 1, 2, 3, 4, 5]), array([6, 7, 8, 9])]
# ------二维
a = np.arange(20)
b = a.reshape((4,5))
# ---行分割
np.split(b,[2])
'''
[array([[0, 1, 2, 3, 4],
[5, 6, 7, 8, 9]]),
array([[10, 11, 12, 13, 14],
[15, 16, 17, 18, 19]])]
'''
# ---列分割
np.split(b,[2],axis=1)
'''
[array([[ 0, 1],
[ 5, 6],
[10, 11],
[15, 16]]),
array([[ 2, 3, 4],
[ 7, 8, 9],
[12, 13, 14],
[17, 18, 19]])]
'''
np.vsplit(b,[3])
'''
[array([[ 0, 1, 2, 3, 4],
[ 5, 6, 7, 8, 9],
[10, 11, 12, 13, 14]]),
array([[15, 16, 17, 18, 19]])]
'''
np.hsplit(b,[2])
'''
[array([[ 0, 1],
[ 5, 6],
[10, 11],
[15, 16]]),
array([[ 2, 3, 4],
[ 7, 8, 9],
[12, 13, 14],
[17, 18, 19]])]
'''
# ------示例------
data = np.arange(16).reshape(4,4)
'''
array([[ 0, 1, 2, 3],
[ 4, 5, 6, 7],
[ 8, 9, 10, 11],
[12, 13, 14, 15]])
'''
X, y = np.hsplit(data, [-1])
'''
array([[ 0, 1, 2],
[ 4, 5, 6],
[ 8, 9, 10],
[12, 13, 14]])
array([[ 3],
[ 7],
[11],
[15]])
'''
y[:,0] # 矩阵转换为向量 array([ 3, 7, 11, 15])1.7 Operaciones con matrices
L =np.arange(10) # array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])
2 * L # array([ 0, 2, 4, 6, 8, 10, 12, 14, 16, 18])
b = np.arange(1,16).reshape(3,5)
'''
array([[ 1, 2, 3, 4, 5],
[ 6, 7, 8, 9, 10],
[11, 12, 13, 14, 15]])
'''
b + 8
'''
array([[ 9, 10, 11, 12, 13],
[14, 15, 16, 17, 18],
[19, 20, 21, 22, 23]])
'''
b * 3
'''
array([[ 3, 6, 9, 12, 15],
[18, 21, 24, 27, 30],
[33, 36, 39, 42, 45]])
'''
b**3
'''
array([[ 1, 8, 27, 64, 125],
[ 216, 343, 512, 729, 1000],
[1331, 1728, 2197, 2744, 3375]], dtype=int32)
'''
np.abs(b) # 绝对值
np.sin(b) # 正弦函数a = np.arange(4).reshape(2,2)
'''
array([[0, 1],
[2, 3]])
'''
b = np.full((2,2),10)
'''
array([[10, 10],
[10, 10]])
'''
a + b
'''
array([[10, 11],
[12, 13]])
'''
a * b
'''
array([[ 0, 10],
[20, 30]])
'''
# 备注：矩阵运算要保证维度相同
# ------矩阵的乘法：第一行乘第一列相加，第一行乘以第二列相加。。。------
a.dot(b)
'''
array([[10, 10],
[50, 50]])
'''
# ------矩阵的转置------
a.T
'''
array([[0, 2],
[1, 3]])
'''1.8 Operaciones sobre matrices y vectores
a
'''
array([[0, 1],
[2, 3]])
'''
b = np.array([1,2])
'''
array([1, 2])
'''
a + b # b和a的每一行的相应元素相加
'''
array([[1, 3],
[3, 5]])
'''
# ------将b堆叠，行堆叠两次，列不变
np.tile(b,(2,1))
'''
array([[1, 2],
[1, 2]])
'''
# ------将b堆叠，乘以a的行数
np.vstack([b]* a.shape[0]) # 同上
# ------b乘以第一列相应元素加和，b乘以第二列相应元素加和
b.dot(a) # array([4, 7])
# ------a第一行乘以b相应元素加和，a第二行乘以b相应元素加和
a.dot(b) # array([2, 8])1.9 Operaciones agregadas
# ------一维------
a = np.random.random(100)
sum(a) # 效率低
np.sum(a) # 效率更高
np.min(a)
np.max(a)
# ------矩阵-----
data = np.arange(16).reshape(4,4)
np.sum(data) # 120
np.sum(data,axis=0) # array([24, 28, 32, 36]) 每一列的和
np.sum(data,axis=1) # array([ 6, 22, 38, 54]) 每一行的和
np.prod(data) # 0 所有数据相乘
np.mean(data) # 7.5
np.var(data)
np.std(data)Dos, pandas
2.1 Objeto de serie
2.1.1 Crear
# 创建
a = pd.Series([1,2,3],index=['a','b','c'])
a
'''
a 1
b 2
c 3
dtype: int64
'''
b = pd.Series([1,2,3])
b
'''
0 1
1 2
2 3
dtype: int64
'''
c = pd.Series({"a":1,"b":2,"c":3})
c
'''
a 1
b 2
c 3
dtype: int64
'''2.1.2 Consulta
print(a['a']) # 1
print(a[2]) # 3
print(a.values) # Series由两个数组组成，数值和索引可作属性访问 [1 2 3]
print(a.index) # Index(['a', 'b', 'c'], dtype='object')
print(a[['a','b']]) # 通过索引查
'''
a 1
b 2
dtype: int64
'''
print(a[[1,2]]) # 通过下标查
'''
b 2
c 3
dtype: int64
'''
print(a[:1]) # 切片查询
'''
a 1
dtype: int64
'''2.1.3 Cruce de elementos
for index , value in a.iteritems():
print(index,value) # index为索引，value为数据值
'''
a 1
b 2
c 3
'''2.1.4 Agregar
print(a.append(c)
'''
a 1
b 2
c 3
a 1
b 2
c 3
dtype: int64
'''2.1.5 Eliminar
print(a.drop('a'))
'''
b 2
c 3
dtype: int64
'''2.2 Marco de datos
2.2.1 Crear
dic = {"a":[1,3],"b":[2,4]} # a,b为列名
print(pd.DataFrame(dic,index=['item1','itme2'])) #index指定行索引
'''
a b
item1 1 2
itme2 3 4
'''
arr = [{"a":1,"b":2},{"a":3,"b":4}] # 每一个字典为一条记录
print(pd.DataFrame(arr))
'''
a b
0 1 2
1 3 4
'''
arr = [[1,2],[3,4]]
print(pd.DataFrame(arr,columns=['a','b'])) # columns指定列名
'''
a b
0 1 2
1 3 4
'''2.2.2 Agregar
df = pd.DataFrame([[1,2],[11,12]],columns=['a','b'])
print(df.append({'a':21,'b':22},ignore_index=True))
'''
a b
0 1 2
1 11 12
2 21 22
'''
print(df.append(df,ignore_index=True))
'''
a b
0 1 2
1 11 12
2 1 2
3 11 12
'''
arr = [[1,2],[11,12]]
df = pd.DataFrame(arr,columns=['a','b'])
df['c'] = [3,13] # 添加新列c
df
'''
a b c
0 1 2 3
1 11 12 13
'''
df.insert(0,'x',[0,10]) # 开始位置插入新列
df
'''
x a b c
0 0 1 2 3
1 10 11 12 13
'''2.2.3 Eliminar
df = pd.DataFrame([[1,2],[11,12]],columns=['a','b']) # 删除第一行
print(df.drop(1))
'''
a b
0 1 2
'''
print(df.drop('a',axis=1)) # 删除a列
'''
b
0 2
1 12
'''2.2.4 objeto de índice
df = pd.DataFrame({"a":[1,3],"b":[2,4]},index=['L1','L2'])
# -----查看索引
print(df.index) # 行索引 Index(['L1', 'L2'], dtype='object')
print(df.columns) # 列索引 Index(['a', 'b'], dtype='object')
# ----修改索引
df.index=(['11','12'])
df
'''
a b
11 1 2
12 3 4
'''
print(df.reindex(columns=['b','a']))
'''
b a
11 2 1
12 4 3
'''2.2.5 Acceso a datos
df = pd.DataFrame([[1,2],[11,12]],columns=['a','b'])
# ----访问列
print(df['a'])
'''
0 1
1 11
'''
print(df[['a','b']])
'''
a b
0 1 2
1 11 12
'''
# ----切片访问行
print(df[:1]) # 返回的是一个DataFrame类型的数据
'''
a b
0 1 2
'''
# ----条件访问
print(df[df['a']==11]) # 筛选a值为11的所有行
print(df[(df['a']>10) & (df['a']<20)]) #多条件筛选
'''
a b
1 11 12
'''2.2.6 Acceso a elementos específicos
df = pd.DataFrame([[1,2],[11,12]],columns=['a','b'])
'''
a b
0 1 2
1 11 12
'''
# 重点：loc和iloc，二者区别在于loc访问数据时使用行索引和列名，iloc方法使用行下标和列下标。
df.iloc[0,0] # 用下标访问数据 1
df.loc[0,'b'] # 访问单个元素 2