当前位置:网站首页>Pytoch foundation - (2) mathematical operation of tensor
Pytoch foundation - (2) mathematical operation of tensor
2022-07-06 03:27:00 【Up and down black】
Catalog
Two 、 Maximum 、 minimum value 、 mean value 、 The absolute value 、 Sort
3、 ... and 、tensor The index of
One 、 Basic operation
import torch
x = torch.tensor([2, 2, 2], dtype=torch.float32)
y = torch.tensor([3, 4, 5], dtype=torch.float32)
- Add
out1 = torch.add(x, y)
print(out1)
- Subtraction
out2 = torch.sub(x, y)
print(out2)
- division
out3 = torch.div(x, y)
print(out3)
- Multiplication
out4 = torch.mul(x, y)
print(out4)
- Matrix multiplication
a = torch.rand((2, 5))
b = torch.rand((5, 3))
out5 = torch.mm(a, b)
print(out5)
print(out5.shape)
- Batch matrix multiplication
batch = 16
c1 = 5
c2 = 10
c3 = 20
x1 = torch.rand(batch, c1, c2) # 16*5*10
x2 = torch.rand(batch, c2, c3) # 16*10*20
out6 = torch.bmm(x1, x2) # 16*5*20
print(out6.shape)
- Index
out7 = x.pow(3)
print(out7)
- Matrix index
x = torch.tensor([[1, 1], [1, 1]]) # (2*2) The number of rows and columns must be the same
out8 = x.matrix_power(2)
print(out8) # 2*2
- broadcasting
x1 = torch.rand((1, 3))
x2 = torch.rand((3, 3))
out9 = x1 -x2
print(out9)
Two 、 Maximum 、 minimum value 、 mean value 、 The absolute value 、 Sort
import torch
x = torch.tensor([[-1, 2, 3], [4, 5, 6]], dtype=torch.float32)
y = torch.tensor([[1, 1, 2], [0, 10, 9]], dtype=torch.float32)
- Maximum 、 minimum value 、 mean value
values, indice = torch.max(x, dim=0) # dim=0 The column ,1 Said line
print(values) # Value of maximum value
print(indice) # Index of maximum value
values, indice = torch.min(x, dim=0) # dim=0 The column ,1 Said line
print(values) # The minimum value
print(indice) # Index of minimum value
values, indice = torch.mean(x, dim=0)
print(values)
print(indice)
# Find the maximum 、 Index of minimum value
out1 = torch.argmax(x, dim=0)
out2 = torch.argmin(x, dim=0)
- The absolute value
out = torch.abs(x)
print(out)
- Sort
values, indice = torch.sort(x, dim=1, descending=False) # descending=False Expressing ascending order
print(values)
print(indice)
3、 ... and 、tensor The index of
- Simple operation
x = torch.tensor([1, 4, 5, 6, 0, 8, 6, 1, 4, 5])
print(x[0]) # First element
print(x[1: 6]) # Output No 2 One to the first 6 Elements
x = torch.randn((3, 10)) # size 3x10
print(x[0]) # Output No 1 All numbers of rows , size 1x10
print(x[0, :]) # size 1x10
print(x[:, 0]) # Output No 1 All the numbers in the column , size 10x1
x = torch.randn((4, 10))
rows = [1, 3]
colums = [2, 9]
print(x[rows, colums]) # (2, 3) (4, 10)
- Conditional
x = torch.tensor([1, 4, 5, 6, 0, 8, 6, 1, 4, 5])
print(torch.where(x > 5, x, x/2)) # x>5 Time output x, Otherwise output x/2
- other
print(x.unique()) # Output non repeating elements
print(x.numel()) # Output x The number of elements in
边栏推荐
- IPv6 comprehensive experiment
- 遥感图像超分辨重建综述
- Getting started with applet cloud development - getting user search content
- How to do function test well
- 八道超经典指针面试题(三千字详解)
- Derivation of anti Park transform and anti Clarke transform formulas for motor control
- 暑期刷题-Day3
- 3857墨卡托坐标系转换为4326 (WGS84)经纬度坐标
- resulttype和resultmap的区别和应用场景
- Svg drag point crop image JS effect
猜你喜欢
电机控制反Park变换和反Clarke变换公式推导
Buuctf question brushing notes - [geek challenge 2019] easysql 1
Idea push rejected solution
Suggestions for new engineer team members
【SLAM】lidar-camera外参标定(港大MarsLab)无需二维码标定板
1.16 - 校验码
Tomb. Weekly update of Finance (February 7 - February 13)
Linear regression and logistic regression
Pytorch基础——(1)张量(tensor)的初始化
RT-Thread--Lwip之FTP(2)
随机推荐
The solution of permission denied (750 permissions should be used with caution)
Shell pass parameters
SAP ALV cell level set color
Four logs of MySQL server layer
NR modulation 1
Redo file corruption repair
Record the process of reverse task manager
1、工程新建
Handwriting database client
3.1 detailed explanation of rtthread serial port device (V1)
Differences and application scenarios between resulttype and resultmap
Shell 传递参数
Audio audiorecord binder communication mechanism
Overview of OCR character recognition methods
[risc-v] external interrupt
Pelosi: Congress will soon have legislation against members' stock speculation
Explore pointers and pointer types in depth
2. GPIO related operations
1.16 - check code
Distributed service framework dobbo