python中range np.linspace np.arange torch.range torch.arange的区别

表示整数：皆可
range, np.arange, torch.arange不包含终点
np.linspace, torch.range包含终点
np.linspace第三个参数表示点的个数，其余第三个参数表示步长

import numpy as np
import torch
# x linspace
x_linspace = np.linspace(1, 17, 17)
print("x_linspace:", x_linspace)
print("type x_linspace:", type(x_linspace))

# x arange
x_arange = np.arange(1, 17, 1)
print("x_arange:", x_arange)
print("type x_arange:", type(x_arange))

# x range
x_range = range(1, 17, 1)
print("x_range:", x_range)
print("type x_range:", type(x_range))

# torch range
torch_range = torch.range(1, 17, 1)
print("torch_range", torch_range)
print("type torch_range", torch_range.dtype)

# torch arange
torch_arange = torch.arange(1, 17, 1)
print("torch_arange", torch_arange)
print("type torch_arange", torch_arange.dtype)

output:

x_linspace: [ 1.  2.  3.  4.  5.  6.  7.  8.  9. 10. 11. 12. 13. 14. 15. 16. 17.]
type x_linspace: 
x_arange: [ 1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16]
type x_arange: 
x_range: range(1, 17)
type x_range: 
test.py:44: UserWarning: torch.range is deprecated and will be removed in a future release because its behavior is inconsistent with Python's range builtin. Instead, use torch.arange, which produces values in [start, end).
  torch_range = torch.range(1, 17, 1)
torch_range tensor([ 1.,  2.,  3.,  4.,  5.,  6.,  7.,  8.,  9., 10., 11., 12., 13., 14.,
        15., 16., 17.])
type torch_range torch.float32
torch_arange tensor([ 1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16])
type torch_arange torch.int64

表示浮点数：np.linspace, np.arange, torch.arange, torch.range
arange(np.arange, torch.arange)不包含终点
np.linspace, torch.range包含终点
np.linspace第三个参数表示点的个数，其余第三个参数表示步长

# x linspace_x
x_linspace = np.linspace(0.1, 1.7, 17)
print("x_linspace:", x_linspace)
print("type x_linspace:", type(x_linspace))

# x arange
x_arange = np.arange(0.1, 1.7, 0.1)
print("x_arange:", x_arange)
print("type x_arange:", type(x_arange))

# torch range
torch_range = torch.range(0.1, 1.7, 0.1)
print("torch_range", torch_range)
print("type torch_range", torch_range.dtype)

# torch arange
torch_arange = torch.arange(0.1, 1.7, 0.1)
print("torch_arange", torch_arange)
print("type torch_arange", torch_arange.dtype)

output:

x_linspace: [0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.  1.1 1.2 1.3 1.4 1.5 1.6 1.7]
type x_linspace: 
x_arange: [0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.  1.1 1.2 1.3 1.4 1.5 1.6]
type x_arange: 
test.py:39: UserWarning: torch.range is deprecated and will be removed in a future release because its behavior is inconsistent with Python's range builtin. Instead, use torch.arange, which produces values in [start, end).
  torch_range = torch.range(0.1, 1.7, 0.1)
torch_range tensor([0.1000, 0.2000, 0.3000, 0.4000, 0.5000, 0.6000, 0.7000, 0.8000, 0.9000,
        1.0000, 1.1000, 1.2000, 1.3000, 1.4000, 1.5000, 1.6000, 1.7000])
type torch_range torch.float32
torch_arange tensor([0.1000, 0.2000, 0.3000, 0.4000, 0.5000, 0.6000, 0.7000, 0.8000, 0.9000,
        1.0000, 1.1000, 1.2000, 1.3000, 1.4000, 1.5000, 1.6000])
type torch_arange torch.float32