Community
Learn
Tools Library
AI Tools
Leisure
search
English
Home > Backend Development > Python Tutorial > RandomCrop in PyTorch (1)

RandomCrop in PyTorch (1)

Barbara Streisand
Release： 2025-01-30 12:12:10
Original
283 people have browsed it

Buy Me a Coffee☕

*Memos:

  • My post explains OxfordIIITPet().

RandomCrop() can crop an image randomly as shown below:

*Memos:

  • The 1st argument for initialization is size(Required-Type:int or tuple/list(int) or size()): *Memos:
    • It's [height, width].
    • It must be 1 <= x.
    • A tuple/list must be the 1D with 1 or 2 elements.
    • A single value(int or tuple/list(int)) means [size, size].
  • The 2nd argument for initialization is padding(Optional-Default:None-Type:int or tuple/list(int)): *Memos:
    • It's [left, top, right, bottom] which can be converted from [left-right, top-bottom] or [left-top-right-bottom].
    • A tuple/list must be the 1D with 1, 2 or 4 elements.
    • A single value(int or tuple/list(int)) means [padding, padding, padding, padding].
    • Double values(tuple/list(int)) means [padding[0], padding[1], padding[0], padding[1]].
  • The 3rd argument for initialization is pad_if_needed(Optional-Default:False-Type:bool):
    • If it's False and size is smaller than an original image or the padded image by padding, there is error.
    • If it's True and size is smaller than an original image or the padded image by padding, there is no error, then the image is randomly padded to become size.
  • The 4th argument for initialization is fill(Optional-Default:0-Type:int, float or tuple/list(int or float)): *Memos:
    • It can change the background of an image. *The background can be seen when an image is positively padded.
    • A tuple/list must be the 1D with 1 or 3 elements.
  • The 5th argument for initialization is padding_mode(Optional-Default:'constant'-Type:str). *'constant', 'edge', 'reflect' or 'symmetric' can be set to it.
  • The 1st argument is img(Required-Type:PIL Image or tensor(int)): *Memos:
    • A tensor must be 2D or 3D.
    • Don't use img=.
  • v2 is recommended to use according to V1 or V2? Which one should I use?. 
from torchvision.datasets import OxfordIIITPet
from torchvision.transforms.v2 import RandomCrop

randomcrop = RandomCrop(size=100)
randomcrop = RandomCrop(size=100,
                        padding=None,
                        pad_if_needed=False, 
                        fill=0,
                        padding_mode='constant')
randomcrop
# RandomCrop(size=(100, 100),
#            pad_if_needed=False,
#            fill=0,
#            padding_mode=constant)

randomcrop.size
# (100, 100)

print(randomcrop.padding)
# None

randomcrop.pad_if_needed
# False

randomcrop.fill
# 0

randomcrop.padding_mode
# 'constant'

origin_data = OxfordIIITPet(
    root="data",
    transform=None
)

s300_data = OxfordIIITPet( # `s` is size.
    root="data",
    transform=RandomCrop(size=300)
    # transform=RandomCrop(size=[300, 300])
)

s200_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=200)
)

s100_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=100)
)

s50_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=50)
)

s10_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=10)
)

s1_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=1)
)

s200_300_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=[200, 300])
)

s300_200_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=[300, 200])
)

s300p100_data = OxfordIIITPet( # `p` is padding.
    root="data",
    transform=RandomCrop(size=300, padding=100)
    # transform=RandomCrop(size=300, padding=[100, 100])
    # transform=RandomCrop(size=300, padding=[100, 100, 100, 100])
)

s300p200_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=300, padding=200)
)

s700_594p100origin_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=[700, 594], padding=100)
)

s300p100_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=300, padding=100)
)

s600_594p100_50origin_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=[600, 594], padding=[100, 50])
)

s300p100_50_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=300, padding=[100, 50])
)

s650_494p25_50_75_100origin_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=[650, 494], padding=[25, 50, 75, 100])
)

s300p25_50_75_100_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=300, padding=[25, 50, 75, 100])
)

s300_194pn100origin_data = OxfordIIITPet( # `n` is negative.
    root="data",
    transform=RandomCrop(size=[300, 194], padding=-100)
)

s150pn100_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=150, padding=-100)
)

s300_294pn50n100origin_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=[300, 294], padding=[-50, -100])
)

s150pn50n100_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=150, padding=[-50, -100])
)

s350_294pn25n50n75n100origin_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=[350, 294], padding=[-25, -50, -75, -100])
)

s150pn25n50n75n100_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=150, padding=[-25, -50, -75, -100])
)

s600_444p25_50origin_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=[600, 444], padding=[25, 50])
)

s200p25_50_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=200, padding=[25, 50])
)

s400_344pn25n50origin_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=[400, 344], padding=[-25, -50])
)

s200pn25n50_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=200, padding=[-25, -50])
)

s400_444p25n50origin_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=[400, 444], padding=[25, -50])
)

s200p25n50_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=200, padding=[25, -50])
)

s600_344pn25_50origin_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=[600, 344], padding=[-25, 50])
)

s200pn25_50_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=200, padding=[-25, 50])
)

s700_594p100fgrayorigin_data = OxfordIIITPet( # `f` is fill.
    root="data",
    transform=RandomCrop(size=[700, 594], padding=100, fill=150)
    # transform=RandomCrop(size=[700, 594], padding=100, fill=[150])
)

s300p100fgray_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=300, padding=100, fill=150)
)

s700_594p100fpurpleorigin_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=[700, 594], padding=100, fill=[160, 32, 240])
)

s300p100fpurple_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=300, padding=100, fill=[160, 32, 240])
)

s700_594p100pmconstorigin_data = OxfordIIITPet( # `pm` is padding_mode.
    root="data",                                # `const` is constant.
    transform=RandomCrop(size=[700, 594], padding=100, padding_mode='constant')
)

s300p100pmconst_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=300, padding=100, padding_mode='constant')
)

s700_594p100pmedgeorigin_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=[700, 594], padding=100, padding_mode='edge')
)

s300p100pmedge_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=300, padding=100, padding_mode='edge')
)

s700_594p100pmrefleorigin_data = OxfordIIITPet( # `refle` is reflect.
    root="data",
    transform=RandomCrop(size=[700, 594], padding=100, padding_mode='reflect')
)

s300p100pmrefle_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=300, padding=100, padding_mode='reflect')
)

s700_594p100pmsymmeorigin_data = OxfordIIITPet( # `symme` is symmetric.
    root="data",
    transform=RandomCrop(size=[700, 594], padding=100, 
                         padding_mode='symmetric')
)

s300p100pmsymme_data = OxfordIIITPet(
    root="data",
    transform=RandomCrop(size=300, padding=100, padding_mode='symmetric')
)

import matplotlib.pyplot as plt

def show_images1(data, main_title=None):
    plt.figure(figsize=(10, 5))
    plt.suptitle(t=main_title, y=0.8, fontsize=14)
    for i in range(1, 6):
        plt.subplot(1, 5, i)
        plt.imshow(X=data[0][0])
    plt.tight_layout()
    plt.show()

plt.figure(figsize=(7, 9))
plt.title(label="s500_394origin_data", fontsize=14)
plt.imshow(X=origin_data[0][0])
show_images1(data=origin_data, main_title="s500_394origin_data")
show_images1(data=s300_data, main_title="s300_data")
show_images1(data=s200_data, main_title="s200_data")
show_images1(data=s100_data, main_title="s100_data")
show_images1(data=s50_data, main_title="s50_data")
show_images1(data=s10_data, main_title="s10_data")
show_images1(data=s1_data, main_title="s1_data")
show_images1(data=s200_300_data, main_title="s200_300_data")
show_images1(data=s300_200_data, main_title="s300_200_data")
print()
show_images1(data=s700_594p100origin_data,
             main_title="s700_594p100origin_data")
show_images1(data=s300p100_data, main_title="s300p100_data")
print()
show_images1(data=s600_594p100_50origin_data,
             main_title="s600_594p100_50origin_data")
show_images1(data=s300p100_50_data, main_title="s300p100_50_data")
print()
show_images1(data=s650_494p25_50_75_100origin_data,
             main_title="s650_494p25_50_75_100origin_data")
show_images1(data=s300p25_50_75_100_data, 
             main_title="s300p25_50_75_100_data")
print()
show_images1(data=s300_194pn100origin_data,
             main_title="s300_194pn100origin_data")
show_images1(data=s150pn100_data, 
             main_title="s150pn100_data")
print()
show_images1(data=s300_294pn50n100origin_data,
             main_title="s300_294pn50n100origin_data")
show_images1(data=s150pn50n100_data, 
             main_title="s150pn50n100_data")
print()
show_images1(data=s350_294pn25n50n75n100origin_data,
             main_title="s350_294pn25n50n75n100origin_data")
show_images1(data=s150pn25n50n75n100_data, 
             main_title="s150pn25n50n75n100_data")
print()
show_images1(data=s600_444p25_50origin_data,
             main_title="s600_444p25_50origin_data")
show_images1(data=s200p25_50_data, 
             main_title="s200p25_50_data")
print()
show_images1(data=s400_344pn25n50origin_data,
             main_title="s400_344pn25n50origin_data")
show_images1(data=s200pn25n50_data, 
             main_title="s200pn25n50_data")
print()
show_images1(data=s400_444p25n50origin_data,
             main_title="s400_444p25n50origin_data")
show_images1(data=s200p25n50_data, 
             main_title="s200p25n50_data")
print()
show_images1(data=s600_344pn25_50origin_data,
             main_title="s600_344pn25_50origin_data")
show_images1(data=s200pn25_50_data, 
             main_title="s200pn25_50_data")
print()
show_images1(data=s700_594p100fgrayorigin_data,
             main_title="s700_594p100fgrayorigin_data")
show_images1(data=s300p100fgray_data, main_title="s300p100fgray_data")
print()
show_images1(data=s700_594p100fpurpleorigin_data,
             main_title="s700_594p100fpurpleorigin_data")
show_images1(data=s300p100fpurple_data, main_title="s300p100fpurple_data")
print()
show_images1(data=s700_594p100pmconstorigin_data,
             main_title="s700_594p100pmconstorigin_data")
show_images1(data=s300p100pmconst_data, main_title="s300p100pmconst_data")
print()
show_images1(data=s700_594p100pmedgeorigin_data,
             main_title="s700_594p100pmedgeorigin_data")
show_images1(data=s300p100pmedge_data, main_title="s300p100pmedge_data")
print()
show_images1(data=s700_594p100pmrefleorigin_data,
             main_title="s700_594p100pmrefleorigin_data")
show_images1(data=s300p100pmrefle_data, main_title="s300p100pmrefle_data")
print()
show_images1(data=s700_594p100pmsymmeorigin_data,
             main_title="s700_594p100pmsymmeorigin_data")
show_images1(data=s300p100pmsymme_data, main_title="s300p100pmsymme_data")

# ↓ ↓ ↓ ↓ ↓ ↓ The code below is identical to the code above. ↓ ↓ ↓ ↓ ↓ ↓
def show_images2(data, main_title=None, s=None, p=None,
                 pin=False, f=0, pm='constant'):
    plt.figure(figsize=(10, 5))
    plt.suptitle(t=main_title, y=0.8, fontsize=14)
    temp_s = s
    im = data[0][0]
    for i in range(1, 6):
        plt.subplot(1, 5, i)
        if not temp_s:
            s = [im.size[1], im.size[0]]
        rc = RandomCrop(size=s, padding=p, # Here
                        pad_if_needed=pin, fill=f, padding_mode=pm)
        plt.imshow(X=rc(im)) # Here
    plt.tight_layout()
    plt.show()

plt.figure(figsize=(7, 9))
plt.title(label="s500_394origin_data", fontsize=14)
plt.imshow(X=origin_data[0][0])
show_images2(data=origin_data, main_title="s500_394origin_data")
show_images2(data=origin_data, main_title="s300_data", s=300)
show_images2(data=origin_data, main_title="s200_data", s=200)
show_images2(data=origin_data, main_title="s100_data", s=100)
show_images2(data=origin_data, main_title="s50_data", s=50)
show_images2(data=origin_data, main_title="s10_data", s=10)
show_images2(data=origin_data, main_title="s1_data", s=1)
show_images2(data=origin_data, main_title="s200_300_data", s=[200, 300])
show_images2(data=origin_data, main_title="s300_200_data", s=[300, 200])
print()
show_images2(data=origin_data, main_title="s700_594p100origin_data",
             s=[700, 594], p=100)
show_images2(data=origin_data, main_title="s300p100_data", s=300, p=100)
print()
show_images2(data=origin_data, main_title="s600_594p100_50origin_data",
             s=[600, 594], p=[100, 50])
show_images2(data=origin_data, main_title="s300p100_50_data", s=300,
             p=[100, 50])
print()
show_images2(data=origin_data, main_title="s650_494p25_50_75_100origin_data",
             s=[650, 494], p=[25, 50, 75, 100])
show_images2(data=origin_data, main_title="s300p25_50_75_100_data", s=300, 
             p=[25, 50, 75, 100])
print()
show_images2(data=origin_data, main_title="s300_194pn100origin_data",
             s=[300, 194], p=-100)
show_images2(data=origin_data, main_title="s150pn100_data", s=150, p=-100)
print()
show_images2(data=origin_data, main_title="s300_294pn50n100origin_data",
             s=[300, 294], p=[-50, -100])
show_images2(data=origin_data, main_title="s150pn50n100_data", s=150,
             p=[-50, -100])
print()
show_images2(data=origin_data, main_title="s350_294pn25n50n75n100origin_data",
             s=[350, 294], p=[-25, -50, -75, -100])
show_images2(data=origin_data, main_title="s150pn25n50n75n100_data", s=150,
             p=[-25, -50, -75, -100])
print()
show_images2(data=origin_data, main_title="s600_444p25_50origin_data",
             s=[600, 444], p=[25, 50])
show_images2(data=origin_data, main_title="s200p25_50_data", s=200,
             p=[25, 50])
print()
show_images2(data=origin_data, main_title="s400_344pn25n50origin_data",
             s=[400, 344], p=[-25, -50])
show_images2(data=origin_data, main_title="s200pn25n50_data", s=200,
             p=[-25, -50])
print()
show_images2(data=origin_data, main_title="s400_444p25n50origin_data",
             s=[400, 444], p=[25, -50])
show_images2(data=origin_data, main_title="s200p25n50_data", s=200,
             p=[25, -50])
print()
show_images2(data=origin_data, main_title="s600_344pn25_50origin_data",
             s=[600, 344], p=[-25, 50])
show_images2(data=origin_data, main_title="s200pn25_50_data", s=200,
             p=[-25, 50])
print()
show_images2(data=origin_data, main_title="s700_594p100fgrayorigin_data", 
             s=[700, 594], p=100, f=150)
show_images2(data=origin_data, main_title="s300p100fgray_data", s=300,
             p=100, f=150)
print()
show_images2(data=origin_data, main_title="s700_594p100fpurpleorigin_data",
             s=[700, 594], p=100, f=[160, 32, 240])
show_images2(data=origin_data, main_title="s300p100fpurple_data", s=300,
             p=100, f=[160, 32, 240])
print()
show_images2(data=origin_data, main_title="s700_594p100pmconstorigin_data",
             s=[700, 594], p=100, pm='constant')
show_images2(data=origin_data, main_title="s300p100pmconst_data", s=300, 
             p=100, pm='constant')
print()
show_images2(data=origin_data, main_title="s700_594p100pmedgeorigin_data",
             s=[700, 594], p=100, pm='edge')
show_images2(data=origin_data, main_title="s300p100pmedge_data", s=300, 
             p=100, pm='edge')
print()
show_images2(data=origin_data, main_title="s700_594p100pmrefleorigin_data",
             s=[700, 594], p=100, pm='reflect')
show_images2(data=origin_data, main_title="s300p100pmrefle_data", s=300, 
             p=100, pm='reflect')
print()
show_images2(data=origin_data, main_title="s700_594p100pmsymmeorigin_data",
             s=[700, 594], p=100, pm='symmetric')
show_images2(data=origin_data, main_title="s300p100pmsymme_data", s=300, 
             p=100, pm='symmetric')
Copy after login

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

Image description

The above is the detailed content of RandomCrop in PyTorch (1). For more information, please follow other related articles on the PHP Chinese website!

Previous article：Hello DEV Community! Introducing PydanticRPC: Build gRPC & Connect RPC Services Without Manually Writing Protobuf Files Next article：None
Statement of this Website
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn
Latest Articles by Author
Latest Issues
function_exists() cannot determine the custom function Function test () {return true;} if (function_exists ('test')) {echo "test is function...
From 2024-04-29 11:01:01
0
3
2672
How to display the mobile version of Google Chrome Hello teacher, how can I change Google Chrome into a mobile version?
From 2024-04-23 00:22:19
0
11
2804
The child window operates the parent window, but the output does not respond. The first two sentences are executable, but the last sentence cannot be implemented.
From 2024-04-19 15:37:47
0
1
2366
There is no output in the parent window document.onclick = function(){ window.opener.document.write('I am the output of the child ...
From 2024-04-18 23:52:34
0
1
2228
Where is the courseware about CSS mind mapping? Courseware
From 2024-04-16 10:10:18
0
0
2330
Related Topics
More>
Popular Recommendations
Popular Tutorials
More>
Related Tutorials
Popular Recommendations
Latest courses
Latest Downloads
More>
Web Effects
Website Source Code
Website Materials
Front End Template