PyTorch

PyTorch

一、张量基础

import numpy as np
import torch

# Creating a zero tensor
x = torch.Tensor(3, 4)
print("Type: {}".format(x.type()))
print("Size: {}".format(x.shape))
print("Values: \n{}".format(x))

输出结果：

# Creating a random tensor
x = torch.randn(2, 3) # normal distribution (rand(2,3) -> uniform distribution)
print (x)

输出结果：

# Zero and Ones tensor
x = torch.zeros(2, 3)
print (x)
x = torch.ones(2, 3)
print (x)

输出结果：

# List → Tensor
x = torch.Tensor([[1, 2, 3],[4, 5, 6]])
print("Size: {}".format(x.shape)) 
print("Values: \n{}".format(x))

输出结果：

# NumPy array → Tensor
x = torch.from_numpy(np.random.rand(2, 3))
print("Size: {}".format(x.shape)) 
print("Values: \n{}".format(x))

输出结果：

# Changing tensor type
x = torch.Tensor(3, 4)
print("Type: {}".format(x.type()))
x = x.long()
print("Type: {}".format(x.type()))

输出结果：

二、张量运算

# Addition
x = torch.randn(2, 3)
y = torch.randn(2, 3)
z = x + y
print("Size: {}".format(z.shape)) 
print("Values: \n{}".format(z))

输出结果：

# Dot product
x = torch.randn(2, 3)
y = torch.randn(3, 2)
z = torch.mm(x, y)
print("Size: {}".format(z.shape)) 
print("Values: \n{}".format(z))

输出结果:

# Transpose
x = torch.randn(2, 3)
print("Size: {}".format(x.shape)) 
print("Values: \n{}".format(x))
y = torch.t(x)
print("Size: {}".format(y.shape)) 
print("Values: \n{}".format(y))

输出结果：

# Reshape
z = x.view(3, 2)
print("Size: {}".format(z.shape)) 
print("Values: \n{}".format(z))

输出结果：

# Dangers of reshaping (unintended consequences)
x = torch.tensor([
    [[1,1,1,1], [2,2,2,2], [3,3,3,3]],
    [[10,10,10,10], [20,20,20,20], [30,30,30,30]]
])
print("Size: {}".format(x.shape)) 
print("Values: \n{}\n".format(x))
a = x.view(x.size(1), -1)
print("Size: {}".format(a.shape)) 
print("Values: \n{}\n".format(a))
b = x.transpose(0,1).contiguous()
print("Size: {}".format(b.shape)) 
print("Values: \n{}\n".format(b))
c = b.view(b.size(0), -1)
print("Size: {}".format(c.shape)) 
print("Values: \n{}".format(c))

输出结果：

# Dimensional operations
x = torch.randn(2, 3)
print("Values: \n{}".format(x))
y = torch.sum(x, dim=0) # add each row's value for every column
print("Values: \n{}".format(y))
z = torch.sum(x, dim=1) # add each columns's value for every row
print("Values: \n{}".format(z))

输出结果：

三、索引、切片、合并

x = torch.randn(3, 4)
print("x: \n{}".format(x))
print ("x[:1]: \n{}".format(x[:1]))
print ("x[:1, 1:3]: \n{}".format(x[:1, 1:3]))

输出结果：

# Select with dimensional indicies
x = torch.randn(2, 3)
print("Values: \n{}".format(x))
col_indices = torch.LongTensor([0, 2])
chosen = torch.index_select(x, dim=1, index=col_indices) # values from column 0 & 2
print("Values: \n{}".format(chosen)) 
row_indices = torch.LongTensor([0, 1])
chosen = x[row_indices, col_indices] # values from (0, 0) & (2, 1)
print("Values: \n{}".format(chosen)) 

输出结果：

# Concatenation
x = torch.randn(2, 3)
print("Values: \n{}".format(x))
y = torch.cat([x, x], dim=0) # stack by rows (dim=1 to stack by columns)
print("Values: \n{}".format(y))

输出结果：

四、梯度

# Tensors with gradient bookkeeping
x = torch.rand(3, 4, requires_grad=True)
y = 3*x + 2
z = y.mean()
z.backward() # z has to be scalar
print("Values: \n{}".format(x))
print("x.grad: \n", x.grad)

公式：

输出结果：

五、CUDA

# Is CUDA available?
print (torch.cuda.is_available())

输出结果：
True

# Creating a zero tensor
x = torch.Tensor(3, 4).to("cpu")
print("Type: {}".format(x.type()))

输出结果：

# Creating a zero tensor
x = torch.Tensor(3, 4).to("cuda")
print("Type: {}".format(x.type()))

输出结果：