Breaking News: Grepper is joining You.com. Read the official announcement!

tensor indexing pytorch

ZCaa369 answered on December 23, 2022 Popularity 9/10 Helpfulness 2/10

answer tensor indexing pytorch

related pytorch tensor

tensor indexing pytorch

Comment

tensor = torch.Tensor([[1,2,3],[4,5,6],[7,8,9]])
# normal indexing/slicing works
# multiple dimensions can be put inside a single bracket pair
tensor[1:, ::2]
"""
tensor([[4., 6.],
        [7., 9.]])
"""

Popularity 9/10 Helpfulness 2/10 Language python

Source: Grepper

Tags: python pytorch tensor-indexing

Link to this answer
Share Copy Link

Contributed on Dec 23 2022

ZCaa369

0 Answers Avg Quality 2/10

Closely Related Answers

pytorch tensor

Comment

import torch

# Creating tensor from list
some_tensor = torch.tensor(some_list)
# Converting tensor to list
list_convert = some_tensor.tolist()
# Creating tensor from numpy array
numpy_tensor = torch.from_numpy(some_array)
# Convert tensor to numpy array
array_convert = numpy_tensor.numpy()
# Access tensor element
print(some_tensor[0])
# Change tensor element
some_tensor[1] = 100
# Extract tensor element
print(some_tensor[0].item())
# Slicing tensor
sliced_tensor = numpy_tensor[1:3]
# Data type of tensor and type of tensor
print(some_tensor.dtype)
print(some_tensor.type())
# Create tensor of specified data-type
tensor_a = torch.tensor([1.3,1.4,4.2,7.4], dtype=torch.float64)
# Create tensor of specified type
tensor_b = torch.FloatTensor(some_list)
# Convert type of tensor
new_tensor = some_tensor.type(torch.FloatTensor)
# Number of elements in a tensor
print(new_tensor.size())
# Dimension or rank of a tensor
print(new_tensor.ndimension())
# Number of elements in a multi-dimensional tensor
print(new_tensor.numel())
# Shape of tensor
print(new_tensor.shape)
# Change dimension of tensor (from row to column)
row_tensor = tensor_a
col_tensor = row_tensor.view(-1,1) # -1 means all data, 1 stands for 1 column

# Tensor operations
u = torch.tensor([1,0])
v = torch.tensor([0,1])

z = u + v # Vector addition
u2 = u * 2 # Vector multiplication with scalar
prod_vec = u * v # Hadamard Product of two vectors 
dot_prod = torch.dot(u,v) # Dot Product of two vectors 
print(tensor_a.mean()) # Average of a tensor
print(tensor_a.max()) 
print(tensor_a.min()) # Min of a tensor
# Matrix multiplication of two vectors 
matrix_prod = torch.mm(a,b)
# Mathematical function
x= torch.tensor([x, x/2, 1])
print(torch.sin(x))

Popularity 10/10 Helpfulness 2/10 Language javascript

Source: Grepper

Tags: python py

Link to this answer
Share Copy Link

Contributed on May 29 2023

Innocent Iguana

0 Answers Avg Quality 2/10

tensor indexing pytorch

Contents

More Related Answers

tensor indexing pytorch

Closely Related Answers

pytorch tensor

Grepper

Documentation

Social

Legal

Contact

Oops, You will need to install Grepper and log-in to perform this action.