numpy数组操作

@TOC

查看维度数量

ndim

a = np.array([1, 2, 3])
print(a.shape)  # (3,)
print(a.dim)  # 1

增加一维

newaxis

将数组增加一维度，类似于pytorch中的unsqueeze

a = np.array([1, 2, 3])
a = a[np.newaxis, :]
print(a.shape)
'''
(1, 3)
'''

拼接

concatenate((a1, a2...), axis=0)

将数组序列进行拼接

a = np.array([[1, 2], [3, 4]])
b = np.array([[5, 6]])
np.concatenate((a, b), axis=0)    # 按行拼接
np.concatenate((a, b.T), axis=1)  # 按列拼接
np.concatenate((a, b), axis=None) # flatten
'''
array([[1, 2],
       [3, 4],
       [5, 6]])
array([[1, 2, 5],
       [3, 4, 6]]) 
array([1, 2, 3, 4, 5, 6])
'''