paddle 基础函数 cos_sim, dot，dist

cos_sim 官方文档：https://www.paddlepaddle.org.cn/documentation/docs/zh/api_cn/layers_cn/cos_sim_cn.html

dot 官方文档：https://www.paddlepaddle.org.cn/documentation/docs/zh/api_cn/layers_cn/dot_cn.html

dist 官方文档：https://www.paddlepaddle.org.cn/documentation/docs/zh/api_cn/layers_cn/dist_cn.html

 

cos_sim 计算两个向量的余弦。

dot 计算两个向量的内积。

dist 计算两个矩阵之间的距离。

 

cos_sim 示例：

import paddle.fluid as fluid
import numpy as np

x = fluid.data(name='x', shape=[3, 7], dtype='float32')
y = fluid.data(name='y', shape=[1, 7], dtype='float32')
out = fluid.layers.cos_sim(x, y)

place = fluid.CPUPlace()
exe = fluid.Executor(place)
exe.run(fluid.default_startup_program())

np_x = np.random.random(size=(3, 7)).astype('float32')
np_y = np.random.random(size=(1, 7)).astype('float32')
output = exe.run(feed={"x": np_x, "y": np_y}, fetch_list = [out])
print(output)

结果：

[array([[0.87142015],
       [0.7826345 ],
       [0.88534766]], dtype=float32)]

 

dot 示例：

import paddle
import paddle.fluid as fluid
import numpy as np

with fluid.dygraph.guard():
    x = fluid.dygraph.to_variable(np.random.uniform(0.1, 1, [10]).astype(np.float32))
    y = fluid.dygraph.to_variable(np.random.uniform(1, 3, [10]).astype(np.float32))
    z = fluid.layers.dot(x, y)
    
    print(x.numpy())
    print(y.numpy())
    print(z.numpy())

结果：

[0.23269959 0.468843   0.6351422  0.47057495 0.11478688 0.31793362
 0.4687848  0.2103034  0.46470684 0.6318518 ]

[2.6248667 1.5989798 1.2326385 2.1651013 2.523104  1.115678  2.720436
 2.923536  2.7323349 2.0036182]

[8.232403]

 

dist 示例：

import paddle
import paddle.fluid as fluid
import numpy as np

with fluid.dygraph.guard():
    x = fluid.dygraph.to_variable(np.array([[3, 3],[3, 3], [2, 2], [3, 0]]).astype(np.float32))
    y = fluid.dygraph.to_variable(np.array([[3, 3],[3, 1]]).astype(np.float32))
    
    print("x numpy data: \n", x.numpy())
    print()
    print("y numpy data: \n", y.numpy())
    print()

    out = fluid.layers.dist(x, y, 0) # 广播后的 y = [[3, 3], [3, 1], [3, 3], [3, 1]]
    print(out.numpy()) # out = [1.]
    
    out = fluid.layers.dist(x, y, 2)
    print(out.numpy()) # out = [2.]
    
    out = fluid.layers.dist(x, y, float("inf"))
    print(out.numpy()) # out = [2.]
    
    out = fluid.layers.dist(x, y, float("-inf"))
    print(out.numpy()) # out = [0.]

结果：

x numpy data: 
 [[3. 3.]
 [3. 3.]
 [2. 2.]
 [3. 0.]]

y numpy data: 
 [[3. 3.]
 [3. 1.]]

[4.]
[2.6457512]
[2.]
[0.]