RandomResizedCrop in PyTorch (1)

Buy Me a Coffee☕

*Memos:

• My post explains RandomResizedCrop() with ratio argument.
• My post explains OxfordIIITPet().

RandomResizedCrop() can crop a random part of an image, then resize it to a given size 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 scale(Optional-Type:tuple/list(int or float)): *Memos:
  • It's [min, max] so it must min <= max.
  • It must be 0 <= x.
  • A tuple/list must be the 1D with 2 elements.
  • A double of 0 or 1 <= x gets the same result.
• The 3rd argument for initialization is ratio(Optional-Type:tuple/list(int or float)): *Memos:
  • It's [min, max] so it must min <= max.
  • It must be 0 < x.
  • A tuple/list must be the 1D with 2 elements.
• The 4th argument for initialization is interpolation(Optional-Default:InterpolationMode.BILINEAR-Type:InterpolationMode).
• The 5th argument for initialization is antialias(Optional-Default:True-Type:bool). *Even if setting False to it, it's always True if interpolation is InterpolationMode.BILINEAR or InterpolationMode.BICUBIC.
• The 1st argument is img(Required-Type:PIL Image or tensor(int)): *Memos:
  • A tensor must be 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 RandomResizedCrop
from torchvision.transforms.functional import InterpolationMode

rrc = RandomResizedCrop(size=100)
rrc = RandomResizedCrop(size=100,
                        scale=(0.08, 1.0),
                        ratio=(0.75, 1.3333333333333333),
                        interpolation=InterpolationMode.BILINEAR,
                        antialias=True)
rrc
# RandomResizedCrop(size=(100, 100),
#                   scale=(0.08, 1.0),
#                   ratio=(0.75, 1.3333333333333333), 
#                   interpolation=InterpolationMode.BILINEAR,
#                   antialias=True)

rrc.size
# (100, 100)

rrc.scale
# (0.08, 1.0)

rrc.ratio
# (0.75, 1.3333333333333333)

rrc.interpolationa
# <InterpolationMode.BILINEAR: 'bilinear'>

rrc.antialias
# True

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

s1000_data = OxfordIIITPet( # `s` is size.
    root="data",
    transform=RandomResizedCrop(size=1000)
    # transform=RandomResizedCrop(size=[1000])
    # transform=RandomResizedCrop(size=[1000, 1000])
)

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

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

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

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

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

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

s100sc0_1_data = OxfordIIITPet( # `sc` is scale.
    root="data",
    transform=RandomResizedCrop(size=100, scale=[0, 1])
)

s100sc0_05_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[0, 0.5])
)

s100sc05_1_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[0.5, 1])
)

s100sc0_0_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[0, 0])
)

s100sc0001_0001_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[0.001, 0.001])
)

s100sc001_001_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[0.01, 0.01])
)

s100sc01_01_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[0.1, 0.1])
)

s100sc02_02_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[0.2, 0.2])
)

s100sc03_03_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[0.3, 0.3])
)

s100sc04_04_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[0.4, 0.4])
)

s100sc05_05_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[0.5, 0.5])
)

s100sc06_06_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[0.6, 0.6])
)

s100sc07_07_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[0.7, 0.7])
)

s100sc08_08_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[0.8, 0.8])
)

s100sc09_09_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[0.9, 0.9])
)

s100sc1_1_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[1, 1])
)

s100sc10_10_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[10, 10])
)

s100sc100_100_data = OxfordIIITPet(
    root="data",
    transform=RandomResizedCrop(size=100, scale=[100, 100])
)

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, (im, _) in zip(range(1, 6), data):
        plt.subplot(1, 5, i)
        plt.imshow(X=im)
    plt.tight_layout()
    plt.show()

show_images1(data=origin_data, main_title="origin_data")
show_images1(data=s1000_data, main_title="s1000_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")
print()
show_images2(data=origin_data, main_title="origin_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=s100sc0_1_data, main_title="s100sc0_1_data")
show_images1(data=s100sc0_05_data, main_title="s100sc0_05_data")
show_images1(data=s100sc05_1_data, main_title="s100sc05_1_data")
print()
show_images1(data=s100sc0_0_data, main_title="s100sc0_0_data")
show_images1(data=s100sc0001_0001_data, main_title="s100sc0001_0001_data")
show_images1(data=s100sc001_001_data, main_title="s100sc001_001_data")
show_images1(data=s100sc01_01_data, main_title="s100sc01_01_data")
show_images1(data=s100sc02_02_data, main_title="s100sc02_02_data")
show_images1(data=s100sc03_03_data, main_title="s100sc03_03_data")
show_images1(data=s100sc04_04_data, main_title="s100sc04_04_data")
show_images1(data=s100sc05_05_data, main_title="s100sc05_05_data")
show_images1(data=s100sc06_06_data, main_title="s100sc06_06_data")
show_images1(data=s100sc07_07_data, main_title="s100sc07_07_data")
show_images1(data=s100sc08_08_data, main_title="s100sc08_08_data")
show_images1(data=s100sc09_09_data, main_title="s100sc09_09_data")
show_images1(data=s100sc1_1_data, main_title="s100sc1_1_data")
show_images1(data=s100sc10_10_data, main_title="s100sc10_10_data")
show_images1(data=s100sc100_100_data, main_title="s100sc100_100_data")

# ↓ ↓ ↓ ↓ ↓ ↓ The code below is identical to the code above. ↓ ↓ ↓ ↓ ↓ ↓ 
def show_images2(data, main_title=None, s=None, sc=(0.08, 1.0),
                 r=(0.75, 1.3333333333333333),
                 ip=InterpolationMode.BILINEAR, a=True):
    plt.figure(figsize=[10, 5])
    plt.suptitle(t=main_title, y=0.8, fontsize=14)
    for i, (im, _) in zip(range(1, 6), data):
        plt.subplot(1, 5, i)
        if s:
            rrc = RandomResizedCrop(size=s, scale=sc, # Here
                                    ratio=r, interpolation=ip,
                                    antialias=a)
            plt.imshow(X=rrc(im)) # Here
        else:
            plt.imshow(X=im)
    plt.tight_layout()
    plt.show()

show_images2(data=origin_data, main_title="origin_data")
show_images2(data=origin_data, main_title="s1000_data", s=1000)
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)
print()
show_images2(data=origin_data, main_title="origin_data")
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="s100sc0_1_data", s=100, sc=[0, 1])
show_images2(data=origin_data, main_title="s100sc0_05_data", s=100,
             sc=[0, 0.5])
show_images2(data=origin_data, main_title="s100sc05_1_data", s=100,
             sc=[0.5, 1])
print()
show_images2(data=origin_data, main_title="s100sc0_0_data", s=100, sc=[0, 0])
show_images2(data=origin_data, main_title="s100sc0001_0001_data", s=100,
             sc=[0.001, 0.001])
show_images2(data=origin_data, main_title="s100sc001_001_data", s=100,
             sc=[0.01, 0.01])
show_images2(data=origin_data, main_title="s100sc01_01_data", s=100,
             sc=[0.1, 0.1])
show_images2(data=origin_data, main_title="s100sc02_02_data", s=100,
             sc=[0.2, 0.2])
show_images2(data=origin_data, main_title="s100sc03_03_data", s=100,
             sc=[0.3, 0.3])
show_images2(data=origin_data, main_title="s100sc04_04_data", s=100,
             sc=[0.4, 0.4])
show_images2(data=origin_data, main_title="s100sc05_05_data", s=100,
             sc=[0.5, 0.5])
show_images2(data=origin_data, main_title="s100sc06_06_data", s=100,
             sc=[0.6, 0.6])
show_images2(data=origin_data, main_title="s100sc07_07_data", s=100,
             sc=[0.7, 0.7])
show_images2(data=origin_data, main_title="s100sc08_08_data", s=100,
             sc=[0.8, 0.8])
show_images2(data=origin_data, main_title="s100sc09_09_data", s=100,
             sc=[0.9, 0.9])
show_images2(data=origin_data, main_title="s100sc1_1_data", s=100, sc=[1, 1])
show_images2(data=origin_data, main_title="s100sc10_10_data", s=100,
             sc=[10, 10])
show_images2(data=origin_data, main_title="s100sc100_100_data", s=100,
             sc=[100, 100])

---

---

---