RandomAffine in PyTorch (1)

Buy Me a Coffee☕

*Memos:

• My post explains RandomAffine() about scale argument.
• My post explains RandomAffine() about shear argument (1).
• My post explains RandomAffine() about shear argument (2).
• My post explains RandomAffine() about shear argument (3).
• My post explains RandomAffine() about shear argument (4).
• My post explains RandomRotation().
• My post explains RandomPerspective().
• My post explains OxfordIIITPet().

RandomAffine() can do random rotation or random affine transformation for an image as shown below. *It's about degrees, translate, fill and center argument:

*Memos:

• The 1st argument for initialization is degrees(Required-Type:int, float or tuple/list(int or float)): *Memos:
  • It can do rotation.
  • It's the range of the degrees [min, max] so it must be min <= max.
  • A degrees value is randomly taken from the range of [min, max].
  • A tuple/list must be the 1D with 2 elements.
  • A single value(int or float) means [-degrees(min), +degrees(max)].
  • A single value(int or float) must be 0 <= x.
• The 2nd argument for initialization is translate(Optional-Default:None-Type:tuple/list(int or float)): *Memos:
  • It's [a, b].
  • It must be 0 <= 1.
  • It must be the 1D with 2 elements.
  • The 1st element is for the horizontal shift randomly taken in the range of -img_width * a < horizontal shift < img_width * a.
  • The 2nd element is for the vertical shift randomly taken in the range of -img_height * b < vertical shift < img_height * b.
• The 3rd argument for initialization is scale(Optional-Default:None-Type:tuple/list(int or float)): *Memos:
  • It's [min, max] so it must be min <= max.
  • It must be 0 < x.
  • It must be the 1D with 2 elements.
  • A scale value is randomly taken from the range of [min, max].
• The 4th argument for initialization is shear(Optional-Default:None-Type:int, float or tuple/list(int or float)): *Memos:
  • It can do affine transformation with x and y.
  • It's [min, max, min, max] so it must be min <= max. *Memos:
  • The 1st two elements are the range of x.
  • The 2nd two elements are the range of y.
  • x value is randomly taken from the range of the 1st two elements.
  • y value is randomly taken from the range of the 2nd two elements.
  • A tuple/list must be the 1D with 2 or 4 elements.
  • The tuple/list of 2 elements means [shear[0](min), shear[1](max), 0.0(min), 0.0(max)].
  • A single value means [-shear(min), +shear(max), 0.0(min), 0.0(max)].
  • A single value must be 0 <= x.
• The 5th argument for initialization is interpolation(Optional-Default:InterpolationMode.NEAREST-Type:InterpolationMode).
• The 6th 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 doing rotation or affine transformation for an image.
  • A tuple/list must be the 1D with 1 or 3 elements.
  • If all values are x <= 0, it's black.
• The 7th argument for initialization is center(Optional-Default:None-Type:tuple/list(int or float)): *Memos:
  • It can change the center position of an image.
  • It must be the 1D with 2 elements.
  • If at least one value is x <= 0, an image isn't shown, only showing the background of the image.
• 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 RandomAffine
from torchvision.transforms.functional import InterpolationMode

ra = RandomAffine(degrees=90)
ra = RandomAffine(degrees=[-90, 90], translate=None, scale=None, 
                  shear=None, interpolation=InterpolationMode.NEAREST,
                  fill=0, center=None)
ra
# RandomAffine(degrees=[-90, 90],
#              interpolation=InterpolationMode.NEAREST,
#              fill=0)

ra.degrees
# [-90.0, 90.0]

print(ra.translate)
# None

print(ra.scale)
# None

print(ra.shear)
# None

ra.interpolation
# <InterpolationMode.NEAREST: 'nearest'>

ra.fill
# 0

print(ra.center)
# None

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

d0origin_data = OxfordIIITPet( # `d` is degrees.
    root="data",
    transform=RandomAffine(degrees=0)
    # transform=RandomAffine(degrees=[0, 0])
)

d180_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=180)
    # transform=RandomAffine(degrees=[-180, 180])
    # transform=RandomAffine(degrees=[-360, 0])
    # transform=RandomAffine(degrees=[0, 360])
)

dn180_0_data = OxfordIIITPet( # `n` is negative.
    root="data",
    transform=RandomAffine(degrees=[-180, 0])
    # transform=RandomAffine(degrees=[180, 360])
)

d0_180_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[0, 180])
    # transform=RandomAffine(degrees=[-360, -180])
)

d15_15_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[15, 15])
    # transform=RandomAffine(degrees=[-345, -345])
)

d30_30_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[30, 30])
    # transform=RandomAffine(degrees=[-330, -330])
)

d45_45_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[45, 45])
    # transform=RandomAffine(degrees=[-315, -315])
)

d60_60_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[60, 60])
    # transform=RandomAffine(degrees=[-300, -300])
)

d75_75_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[75, 75])
    # transform=RandomAffine(degrees=[-285, -285])
)

d90_90_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[90, 90])
    # transform=RandomAffine(degrees=[-270, -270])
)

d105_105_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[105, 105])
    # transform=RandomAffine(degrees=[-255, -255])
)

d120_120_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[120, 120])
    # transform=RandomAffine(degrees=[-240, -240])
)

d135_135_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[135, 135])
    # transform=RandomAffine(degrees=[-225, -225])
)

d150_150_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[150, 150])
    # transform=RandomAffine(degrees=[-210, -210])
)

d165_165_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[165, 165])
    # transform=RandomAffine(degrees=[-195, -195])
)

d180_180_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[180, 180])
    # transform=RandomAffine(degrees=[-180, -180])
)

dn15n15_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[-15, -15])
    # transform=RandomAffine(degrees=[345, 345])
)

dn30n30_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[-30, -30])
    # transform=RandomAffine(degrees=[330, 330])
)

dn45n45_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[-45, -45])
    # transform=RandomAffine(degrees=[315, 315])
)

dn60n60_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[-60, -60])
    # transform=RandomAffine(degrees=[300, 300])
)

dn75n75_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[-75, -75])
    # transform=RandomAffine(degrees=[285, 285])
)

dn90n90_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[-90, -90])
    # transform=RandomAffine(degrees=[270, 270])
)

dn105n105_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[-105, -105])
    # transform=RandomAffine(degrees=[255, 255])
)

dn120n120_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[-120, -120])
    # transform=RandomAffine(degrees=[240, 240])
)

dn135n135_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[-135, -135])
    # transform=RandomAffine(degrees=[225, 225])
)

dn150n150_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[-150, -150])
    # transform=RandomAffine(degrees=[210, 210])
)

dn165n165_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[-165, -165])
    # transform=RandomAffine(degrees=[195, 195])
)

dn180n180_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[-180, -180])
    # transform=RandomAffine(degrees=[180, 180])
)

hrvrtran_data = OxfordIIITPet( # `hr` is horizontal and `vr` is vertical.
    root="data",               # `tran` is translate.
    transform=RandomAffine(degrees=[0, 0], translate=[0.8, 0.5])
)

hrtran_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[0, 0], translate=[0.8, 0])
)

vrtran_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[0, 0], translate=[0, 0.5])
)

dn45n45fgray_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[-45, -45], fill=150)
)

d135_135fpurple_data = OxfordIIITPet(
    root="data",
    transform=RandomAffine(degrees=[135, 135], fill=[160, 32, 240])
)

d180_180c270_200_data = OxfordIIITPet( # `c` is center.
    root="data",
    transform=RandomAffine(degrees=[180, 180], center=[270, 200])
)

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.xticks(ticks=[])
        plt.yticks(ticks=[])
    plt.tight_layout()
    plt.show()

show_images1(data=origin_data, main_title="origin_data")
print()
show_images1(data=d0origin_data, main_title="d0origin_data")
show_images1(data=d180_data, main_title="d180_data")
show_images1(data=dn180_0_data, main_title="dn180_0_data")
show_images1(data=d0_180_data, main_title="d0_180_data")
print()
show_images1(data=d0origin_data, main_title="d0origin_data")
show_images1(data=d15_15_data, main_title="d15_15_data")
show_images1(data=d30_30_data, main_title="d30_30_data")
show_images1(data=d45_45_data, main_title="d45_45_data")
show_images1(data=d60_60_data, main_title="d60_60_data")
show_images1(data=d75_75_data, main_title="d75_75_data")
show_images1(data=d90_90_data, main_title="d90_90_data")
show_images1(data=d105_105_data, main_title="d105_105_data")
show_images1(data=d120_120_data, main_title="d120_120_data")
show_images1(data=d135_135_data, main_title="d135_135_data")
show_images1(data=d150_150_data, main_title="d150_150_data")
show_images1(data=d165_165_data, main_title="d165_165_data")
show_images1(data=d180_180_data, main_title="d180_180_data")
print()
show_images1(data=d0origin_data, main_title="d0origin_data")
show_images1(data=dn15n15_data, main_title="dn15n15_data")
show_images1(data=dn30n30_data, main_title="dn30n30_data")
show_images1(data=dn45n45_data, main_title="dn45n45_data")
show_images1(data=dn60n60_data, main_title="dn60n60_data")
show_images1(data=dn75n75_data, main_title="dn75n75_data")
show_images1(data=dn90n90_data, main_title="dn90n90_data")
show_images1(data=dn105n105_data, main_title="dn105n105_data")
show_images1(data=dn120n120_data, main_title="dn120n120_data")
show_images1(data=dn135n135_data, main_title="dn135n135_data")
show_images1(data=dn150n150_data, main_title="dn150n150_data")
show_images1(data=dn165n165_data, main_title="dn165n165_data")
show_images1(data=dn180n180_data, main_title="dn180n180_data")
print()
show_images1(data=hrvrtran_data, main_title="hrvrtran_data")
show_images1(data=hrtran_data, main_title="hrtran_data")
show_images1(data=vrtran_data, main_title="vrtran_data")
print()
show_images1(data=dn45n45fgray_data, main_title="dn45n45fgray_data")
show_images1(data=d135_135fpurple_data, main_title="d135_135fpurple_data")
show_images1(data=d180_180c270_200_data, main_title="d180_180c270_200_data")

# ↓ ↓ ↓ ↓ ↓ ↓ The code below is identical to the code above. ↓ ↓ ↓ ↓ ↓ ↓ 
def show_images2(data, main_title=None, d=None, t=None, sc=None, sh=None,
                 ip=InterpolationMode.NEAREST, f=0, c=None):
    plt.figure(figsize=[10, 5])
    plt.suptitle(t=main_title, y=0.8, fontsize=14)
    if d != None:
        for i, (im, _) in zip(range(1, 6), data):
            plt.subplot(1, 5, i)
            ra = RandomAffine(degrees=d, translate=t, scale=sc, shear=sh,
                              interpolation=ip, center=c, fill=f)
            plt.imshow(X=ra(im))
            plt.xticks(ticks=[])
            plt.yticks(ticks=[])
    else:
        for i, (im, _) in zip(range(1, 6), data):
            plt.subplot(1, 5, i)
            plt.imshow(X=im)
            plt.xticks(ticks=[])
            plt.yticks(ticks=[])  
    plt.tight_layout()
    plt.show()

show_images2(data=origin_data, main_title="origin_data")
print()
show_images2(data=origin_data, main_title="d0origin_data", d=0)
show_images2(data=origin_data, main_title="d180_data", d=180)
show_images2(data=origin_data, main_title="dn180_0_data", d=[-180, 0])
show_images2(data=origin_data, main_title="d0_180_data", d=[0, 180])
print()
show_images2(data=origin_data, main_title="d0origin_data", d=0)
show_images2(data=origin_data, main_title="d15_15_data", d=[15, 15])
show_images2(data=origin_data, main_title="d30_30_data", d=[30, 30])
show_images2(data=origin_data, main_title="d45_45_data", d=[45, 45])
show_images2(data=origin_data, main_title="d60_60_data", d=[60, 60])
show_images2(data=origin_data, main_title="d75_75_data", d=[75, 75])
show_images2(data=origin_data, main_title="d90_90_data", d=[90, 90])
show_images2(data=origin_data, main_title="d105_105_data", d=[105, 105])
show_images2(data=origin_data, main_title="d120_120_data", d=[120, 120])
show_images2(data=origin_data, main_title="d135_135_data", d=[135, 135])
show_images2(data=origin_data, main_title="d150_150_data", d=[150, 150])
show_images2(data=origin_data, main_title="d165_165_data", d=[165, 165])
show_images2(data=origin_data, main_title="d180_180_data", d=[180, 180])
print()
show_images2(data=origin_data, main_title="d0origin_data", d=0)
show_images2(data=origin_data, main_title="dn15n15_data", d=[-15, -15])
show_images2(data=origin_data, main_title="dn30n30_data", d=[-30, -30])
show_images2(data=origin_data, main_title="dn45n45_data", d=[-45, -45])
show_images2(data=origin_data, main_title="dn60n60_data", d=[-60, -60])
show_images2(data=origin_data, main_title="dn75n75_data", d=[-75, -75])
show_images2(data=origin_data, main_title="dn90n90_data", d=[-90, -90])
show_images2(data=origin_data, main_title="dn105n105_data",
             d=[-105, -105])
show_images2(data=origin_data, main_title="dn120n120_data",
             d=[-120, -120])
show_images2(data=origin_data, main_title="dn135n135_data",
             d=[-135, -135])
show_images2(data=origin_data, main_title="dn150n150_data",
             d=[-150, -150])
show_images2(data=origin_data, main_title="dn165n165_data",
             d=[-165, -165])
show_images2(data=origin_data, main_title="dn180n180_data",
             d=[-180, -180])
print()
show_images2(data=origin_data, main_title="hrvrtran_data", d=[0, 0],
             t=[0.8, 0.5])
show_images2(data=origin_data, main_title="hrtran_data", d=[0, 0],
             t=[0.8, 0])
show_images2(data=origin_data, main_title="vrtran_data", d=[0, 0],
             t=[0, 0.5])
print()
show_images2(data=origin_data, main_title="dn45n45fgray_data",
             d=[-45, -45], f=150)
show_images2(data=origin_data, main_title="d135_135fpurple_data",
             d=[135, 135], f=[160, 32, 240])
show_images2(data=origin_data, main_title="d180_180c270_200_data",
             d=[180, 180], c=[270, 200])

---

---

---

---

---