github/ColossalAI
1
0
Fork 0
mirror of https://github.com/hpcaitech/ColossalAI.git synced 2025-09-02 17:46:42 +00:00
Code Issues Packages Projects Releases Wiki Activity

[misc] update pre-commit and run all files (#4752)

Browse Source 
* [misc] update pre-commit

* [misc] run pre-commit

* [misc] remove useless configuration files

* [misc] ignore cuda for clang-format
This commit is contained in:
Hongxin Liu
2023-09-19 14:20:26 +08:00
committed by GitHub
parent 3c6b831c26
commit 079bf3cb26
1268 changed files with 50037 additions and 38444 deletions
Download Patch File Download Diff File Expand all files Collapse all files

209
examples/images/diffusion/ldm/modules/image_degradation/bsrgan.py
View File

@@ -10,33 +10,32 @@
# --------------------------------------------
"""
import numpy as np
import cv2
import torch
from functools import partial
import random
from scipy import ndimage
from functools import partial
import albumentations
import cv2
import ldm.modules.image_degradation.utils_image as util
import numpy as np
import scipy
import scipy.stats as ss
import torch
from scipy import ndimage
from scipy.interpolate import interp2d
from scipy.linalg import orth
import albumentations
import ldm.modules.image_degradation.utils_image as util
def modcrop_np(img, sf):
'''
"""
Args:
img: numpy image, WxH or WxHxC
sf: scale factor
Return:
cropped image
'''
"""
w, h = img.shape[:2]
im = np.copy(img)
return im[:w - w % sf, :h - h % sf, ...]
return im[: w - w % sf, : h - h % sf, ...]
"""
@@ -54,7 +53,7 @@ def analytic_kernel(k):
# Loop over the small kernel to fill the big one
for r in range(k_size):
for c in range(k_size):
big_k[2 * r:2 * r + k_size, 2 * c:2 * c + k_size] += k[r, c] * k
big_k[2 * r : 2 * r + k_size, 2 * c : 2 * c + k_size] += k[r, c] * k
# Crop the edges of the big kernel to ignore very small values and increase run time of SR
crop = k_size // 2
cropped_big_k = big_k[crop:-crop, crop:-crop]
@@ -63,7 +62,7 @@ def analytic_kernel(k):
def anisotropic_Gaussian(ksize=15, theta=np.pi, l1=6, l2=6):
""" generate an anisotropic Gaussian kernel
"""generate an anisotropic Gaussian kernel
Args:
ksize : e.g., 15, kernel size
theta : [0, pi], rotation angle range
@@ -74,7 +73,7 @@ def anisotropic_Gaussian(ksize=15, theta=np.pi, l1=6, l2=6):
k : kernel
"""
v = np.dot(np.array([[np.cos(theta), -np.sin(theta)], [np.sin(theta), np.cos(theta)]]), np.array([1., 0.]))
v = np.dot(np.array([[np.cos(theta), -np.sin(theta)], [np.sin(theta), np.cos(theta)]]), np.array([1.0, 0.0]))
V = np.array([[v[0], v[1]], [v[1], -v[0]]])
D = np.array([[l1, 0], [0, l2]])
Sigma = np.dot(np.dot(V, D), np.linalg.inv(V))
@@ -126,13 +125,13 @@ def shift_pixel(x, sf, upper_left=True):
def blur(x, k):
'''
"""
x: image, NxcxHxW
k: kernel, Nx1xhxw
'''
"""
n, c = x.shape[:2]
p1, p2 = (k.shape[-2] - 1) // 2, (k.shape[-1] - 1) // 2
x = torch.nn.functional.pad(x, pad=(p1, p2, p1, p2), mode='replicate')
x = torch.nn.functional.pad(x, pad=(p1, p2, p1, p2), mode="replicate")
k = k.repeat(1, c, 1, 1)
k = k.view(-1, 1, k.shape[2], k.shape[3])
x = x.view(1, -1, x.shape[2], x.shape[3])
@@ -142,8 +141,8 @@ def blur(x, k):
return x
def gen_kernel(k_size=np.array([15, 15]), scale_factor=np.array([4, 4]), min_var=0.6, max_var=10., noise_level=0):
""""
def gen_kernel(k_size=np.array([15, 15]), scale_factor=np.array([4, 4]), min_var=0.6, max_var=10.0, noise_level=0):
""" "
# modified version of https://github.com/assafshocher/BlindSR_dataset_generator
# Kai Zhang
# min_var = 0.175 * sf # variance of the gaussian kernel will be sampled between min_var and max_var
@@ -157,8 +156,7 @@ def gen_kernel(k_size=np.array([15, 15]), scale_factor=np.array([4, 4]), min_var
# Set COV matrix using Lambdas and Theta
LAMBDA = np.diag([lambda_1, lambda_2])
Q = np.array([[np.cos(theta), -np.sin(theta)],
[np.sin(theta), np.cos(theta)]])
Q = np.array([[np.cos(theta), -np.sin(theta)], [np.sin(theta), np.cos(theta)]])
SIGMA = Q @ LAMBDA @ Q.T
INV_SIGMA = np.linalg.inv(SIGMA)[None, None, :, :]
@@ -208,13 +206,13 @@ def fspecial_laplacian(alpha):
def fspecial(filter_type, *args, **kwargs):
'''
"""
python code from:
https://github.com/ronaldosena/imagens-medicas-2/blob/40171a6c259edec7827a6693a93955de2bd39e76/Aulas/aula_2_-_uniform_filter/matlab_fspecial.py
'''
if filter_type == 'gaussian':
"""
if filter_type == "gaussian":
return fspecial_gaussian(*args, **kwargs)
if filter_type == 'laplacian':
if filter_type == "laplacian":
return fspecial_laplacian(*args, **kwargs)
@@ -226,19 +224,19 @@ def fspecial(filter_type, *args, **kwargs):
def bicubic_degradation(x, sf=3):
'''
"""
Args:
x: HxWxC image, [0, 1]
sf: down-scale factor
Return:
bicubicly downsampled LR image
'''
"""
x = util.imresize_np(x, scale=1 / sf)
return x
def srmd_degradation(x, k, sf=3):
''' blur + bicubic downsampling
"""blur + bicubic downsampling
Args:
x: HxWxC image, [0, 1]
k: hxw, double
@@ -253,14 +251,14 @@ def srmd_degradation(x, k, sf=3):
pages={3262--3271},
year={2018}
}
'''
x = ndimage.filters.convolve(x, np.expand_dims(k, axis=2), mode='wrap') # 'nearest' | 'mirror'
"""
x = ndimage.filters.convolve(x, np.expand_dims(k, axis=2), mode="wrap") # 'nearest' | 'mirror'
x = bicubic_degradation(x, sf=sf)
return x
def dpsr_degradation(x, k, sf=3):
''' bicubic downsampling + blur
"""bicubic downsampling + blur
Args:
x: HxWxC image, [0, 1]
k: hxw, double
@@ -275,22 +273,22 @@ def dpsr_degradation(x, k, sf=3):
pages={1671--1681},
year={2019}
}
'''
"""
x = bicubic_degradation(x, sf=sf)
x = ndimage.filters.convolve(x, np.expand_dims(k, axis=2), mode='wrap')
x = ndimage.filters.convolve(x, np.expand_dims(k, axis=2), mode="wrap")
return x
def classical_degradation(x, k, sf=3):
''' blur + downsampling
"""blur + downsampling
Args:
x: HxWxC image, [0, 1]/[0, 255]
k: hxw, double
sf: down-scale factor
Return:
downsampled LR image
'''
x = ndimage.filters.convolve(x, np.expand_dims(k, axis=2), mode='wrap')
"""
x = ndimage.filters.convolve(x, np.expand_dims(k, axis=2), mode="wrap")
# x = filters.correlate(x, np.expand_dims(np.flip(k), axis=2))
st = 0
return x[st::sf, st::sf, ...]
@@ -314,7 +312,7 @@ def add_sharpening(img, weight=0.5, radius=50, threshold=10):
blur = cv2.GaussianBlur(img, (radius, radius), 0)
residual = img - blur
mask = np.abs(residual) * 255 > threshold
mask = mask.astype('float32')
mask = mask.astype("float32")
soft_mask = cv2.GaussianBlur(mask, (radius, radius), 0)
K = img + weight * residual
@@ -330,8 +328,8 @@ def add_blur(img, sf=4):
l2 = wd2 * random.random()
k = anisotropic_Gaussian(ksize=2 * random.randint(2, 11) + 3, theta=random.random() * np.pi, l1=l1, l2=l2)
else:
k = fspecial('gaussian', 2 * random.randint(2, 11) + 3, wd * random.random())
img = ndimage.filters.convolve(img, np.expand_dims(k, axis=2), mode='mirror')
k = fspecial("gaussian", 2 * random.randint(2, 11) + 3, wd * random.random())
img = ndimage.filters.convolve(img, np.expand_dims(k, axis=2), mode="mirror")
return img
@@ -366,6 +364,7 @@ def add_resize(img, sf=4):
# img = np.clip(img, 0.0, 1.0)
# return img
def add_Gaussian_noise(img, noise_level1=2, noise_level2=25):
noise_level = random.randint(noise_level1, noise_level2)
rnum = np.random.rand()
@@ -374,11 +373,11 @@ def add_Gaussian_noise(img, noise_level1=2, noise_level2=25):
elif rnum < 0.4: # add grayscale Gaussian noise
img = img + np.random.normal(0, noise_level / 255.0, (*img.shape[:2], 1)).astype(np.float32)
else: # add noise
L = noise_level2 / 255.
L = noise_level2 / 255.0
D = np.diag(np.random.rand(3))
U = orth(np.random.rand(3, 3))
conv = np.dot(np.dot(np.transpose(U), D), U)
img = img + np.random.multivariate_normal([0, 0, 0], np.abs(L ** 2 * conv), img.shape[:2]).astype(np.float32)
img = img + np.random.multivariate_normal([0, 0, 0], np.abs(L**2 * conv), img.shape[:2]).astype(np.float32)
img = np.clip(img, 0.0, 1.0)
return img
@@ -392,23 +391,23 @@ def add_speckle_noise(img, noise_level1=2, noise_level2=25):
elif rnum < 0.4:
img += img * np.random.normal(0, noise_level / 255.0, (*img.shape[:2], 1)).astype(np.float32)
else:
L = noise_level2 / 255.
L = noise_level2 / 255.0
D = np.diag(np.random.rand(3))
U = orth(np.random.rand(3, 3))
conv = np.dot(np.dot(np.transpose(U), D), U)
img += img * np.random.multivariate_normal([0, 0, 0], np.abs(L ** 2 * conv), img.shape[:2]).astype(np.float32)
img += img * np.random.multivariate_normal([0, 0, 0], np.abs(L**2 * conv), img.shape[:2]).astype(np.float32)
img = np.clip(img, 0.0, 1.0)
return img
def add_Poisson_noise(img):
img = np.clip((img * 255.0).round(), 0, 255) / 255.
img = np.clip((img * 255.0).round(), 0, 255) / 255.0
vals = 10 ** (2 * random.random() + 2.0) # [2, 4]
if random.random() < 0.5:
img = np.random.poisson(img * vals).astype(np.float32) / vals
else:
img_gray = np.dot(img[..., :3], [0.299, 0.587, 0.114])
img_gray = np.clip((img_gray * 255.0).round(), 0, 255) / 255.
img_gray = np.clip((img_gray * 255.0).round(), 0, 255) / 255.0
noise_gray = np.random.poisson(img_gray * vals).astype(np.float32) / vals - img_gray
img += noise_gray[:, :, np.newaxis]
img = np.clip(img, 0.0, 1.0)
@@ -418,7 +417,7 @@ def add_Poisson_noise(img):
def add_JPEG_noise(img):
quality_factor = random.randint(30, 95)
img = cv2.cvtColor(util.single2uint(img), cv2.COLOR_RGB2BGR)
result, encimg = cv2.imencode('.jpg', img, [int(cv2.IMWRITE_JPEG_QUALITY), quality_factor])
result, encimg = cv2.imencode(".jpg", img, [int(cv2.IMWRITE_JPEG_QUALITY), quality_factor])
img = cv2.imdecode(encimg, 1)
img = cv2.cvtColor(util.uint2single(img), cv2.COLOR_BGR2RGB)
return img
@@ -428,10 +427,10 @@ def random_crop(lq, hq, sf=4, lq_patchsize=64):
h, w = lq.shape[:2]
rnd_h = random.randint(0, h - lq_patchsize)
rnd_w = random.randint(0, w - lq_patchsize)
lq = lq[rnd_h:rnd_h + lq_patchsize, rnd_w:rnd_w + lq_patchsize, :]
lq = lq[rnd_h : rnd_h + lq_patchsize, rnd_w : rnd_w + lq_patchsize, :]
rnd_h_H, rnd_w_H = int(rnd_h * sf), int(rnd_w * sf)
hq = hq[rnd_h_H:rnd_h_H + lq_patchsize * sf, rnd_w_H:rnd_w_H + lq_patchsize * sf, :]
hq = hq[rnd_h_H : rnd_h_H + lq_patchsize * sf, rnd_w_H : rnd_w_H + lq_patchsize * sf, :]
return lq, hq
@@ -452,18 +451,19 @@ def degradation_bsrgan(img, sf=4, lq_patchsize=72, isp_model=None):
sf_ori = sf
h1, w1 = img.shape[:2]
img = img.copy()[:w1 - w1 % sf, :h1 - h1 % sf, ...] # mod crop
img = img.copy()[: w1 - w1 % sf, : h1 - h1 % sf, ...] # mod crop
h, w = img.shape[:2]
if h < lq_patchsize * sf or w < lq_patchsize * sf:
raise ValueError(f'img size ({h1}X{w1}) is too small!')
raise ValueError(f"img size ({h1}X{w1}) is too small!")
hq = img.copy()
if sf == 4 and random.random() < scale2_prob: # downsample1
if np.random.rand() < 0.5:
img = cv2.resize(img, (int(1 / 2 * img.shape[1]), int(1 / 2 * img.shape[0])),
interpolation=random.choice([1, 2, 3]))
img = cv2.resize(
img, (int(1 / 2 * img.shape[1]), int(1 / 2 * img.shape[0])), interpolation=random.choice([1, 2, 3])
)
else:
img = util.imresize_np(img, 1 / 2, True)
img = np.clip(img, 0.0, 1.0)
@@ -475,7 +475,6 @@ def degradation_bsrgan(img, sf=4, lq_patchsize=72, isp_model=None):
shuffle_order[idx1], shuffle_order[idx2] = shuffle_order[idx2], shuffle_order[idx1]
for i in shuffle_order:
if i == 0:
img = add_blur(img, sf=sf)
@@ -487,13 +486,16 @@ def degradation_bsrgan(img, sf=4, lq_patchsize=72, isp_model=None):
# downsample2
if random.random() < 0.75:
sf1 = random.uniform(1, 2 * sf)
img = cv2.resize(img, (int(1 / sf1 * img.shape[1]), int(1 / sf1 * img.shape[0])),
interpolation=random.choice([1, 2, 3]))
img = cv2.resize(
img,
(int(1 / sf1 * img.shape[1]), int(1 / sf1 * img.shape[0])),
interpolation=random.choice([1, 2, 3]),
)
else:
k = fspecial('gaussian', 25, random.uniform(0.1, 0.6 * sf))
k = fspecial("gaussian", 25, random.uniform(0.1, 0.6 * sf))
k_shifted = shift_pixel(k, sf)
k_shifted = k_shifted / k_shifted.sum() # blur with shifted kernel
img = ndimage.filters.convolve(img, np.expand_dims(k_shifted, axis=2), mode='mirror')
img = ndimage.filters.convolve(img, np.expand_dims(k_shifted, axis=2), mode="mirror")
img = img[0::sf, 0::sf, ...] # nearest downsampling
img = np.clip(img, 0.0, 1.0)
@@ -541,18 +543,20 @@ def degradation_bsrgan_variant(image, sf=4, isp_model=None):
"""
image = util.uint2single(image)
isp_prob, jpeg_prob, scale2_prob = 0.25, 0.9, 0.25
sf_ori = sf
h1, w1 = image.shape[:2]
image = image.copy()[:w1 - w1 % sf, :h1 - h1 % sf, ...] # mod crop
image = image.copy()[: w1 - w1 % sf, : h1 - h1 % sf, ...] # mod crop
h, w = image.shape[:2]
hq = image.copy()
image.copy()
if sf == 4 and random.random() < scale2_prob: # downsample1
if np.random.rand() < 0.5:
image = cv2.resize(image, (int(1 / 2 * image.shape[1]), int(1 / 2 * image.shape[0])),
interpolation=random.choice([1, 2, 3]))
image = cv2.resize(
image,
(int(1 / 2 * image.shape[1]), int(1 / 2 * image.shape[0])),
interpolation=random.choice([1, 2, 3]),
)
else:
image = util.imresize_np(image, 1 / 2, True)
image = np.clip(image, 0.0, 1.0)
@@ -564,7 +568,6 @@ def degradation_bsrgan_variant(image, sf=4, isp_model=None):
shuffle_order[idx1], shuffle_order[idx2] = shuffle_order[idx2], shuffle_order[idx1]
for i in shuffle_order:
if i == 0:
image = add_blur(image, sf=sf)
@@ -576,13 +579,16 @@ def degradation_bsrgan_variant(image, sf=4, isp_model=None):
# downsample2
if random.random() < 0.75:
sf1 = random.uniform(1, 2 * sf)
image = cv2.resize(image, (int(1 / sf1 * image.shape[1]), int(1 / sf1 * image.shape[0])),
interpolation=random.choice([1, 2, 3]))
image = cv2.resize(
image,
(int(1 / sf1 * image.shape[1]), int(1 / sf1 * image.shape[0])),
interpolation=random.choice([1, 2, 3]),
)
else:
k = fspecial('gaussian', 25, random.uniform(0.1, 0.6 * sf))
k = fspecial("gaussian", 25, random.uniform(0.1, 0.6 * sf))
k_shifted = shift_pixel(k, sf)
k_shifted = k_shifted / k_shifted.sum() # blur with shifted kernel
image = ndimage.filters.convolve(image, np.expand_dims(k_shifted, axis=2), mode='mirror')
image = ndimage.filters.convolve(image, np.expand_dims(k_shifted, axis=2), mode="mirror")
image = image[0::sf, 0::sf, ...] # nearest downsampling
image = np.clip(image, 0.0, 1.0)
@@ -609,7 +615,7 @@ def degradation_bsrgan_variant(image, sf=4, isp_model=None):
# add final JPEG compression noise
image = add_JPEG_noise(image)
image = util.single2uint(image)
example = {"image":image}
example = {"image": image}
return example
@@ -630,11 +636,11 @@ def degradation_bsrgan_plus(img, sf=4, shuffle_prob=0.5, use_sharp=True, lq_patc
"""
h1, w1 = img.shape[:2]
img = img.copy()[:w1 - w1 % sf, :h1 - h1 % sf, ...] # mod crop
img = img.copy()[: w1 - w1 % sf, : h1 - h1 % sf, ...] # mod crop
h, w = img.shape[:2]
if h < lq_patchsize * sf or w < lq_patchsize * sf:
raise ValueError(f'img size ({h1}X{w1}) is too small!')
raise ValueError(f"img size ({h1}X{w1}) is too small!")
if use_sharp:
img = add_sharpening(img)
@@ -686,11 +692,12 @@ def degradation_bsrgan_plus(img, sf=4, shuffle_prob=0.5, use_sharp=True, lq_patc
with torch.no_grad():
img, hq = isp_model.forward(img.copy(), hq)
else:
print('check the shuffle!')
print("check the shuffle!")
# resize to desired size
img = cv2.resize(img, (int(1 / sf * hq.shape[1]), int(1 / sf * hq.shape[0])),
interpolation=random.choice([1, 2, 3]))
img = cv2.resize(
img, (int(1 / sf * hq.shape[1]), int(1 / sf * hq.shape[0])), interpolation=random.choice([1, 2, 3])
)
# add final JPEG compression noise
img = add_JPEG_noise(img)
@@ -701,30 +708,30 @@ def degradation_bsrgan_plus(img, sf=4, shuffle_prob=0.5, use_sharp=True, lq_patc
return img, hq
if __name__ == '__main__':
print("hey")
img = util.imread_uint('utils/test.png', 3)
print(img)
img = util.uint2single(img)
print(img)
img = img[:448, :448]
h = img.shape[0] // 4
print("resizing to", h)
sf = 4
deg_fn = partial(degradation_bsrgan_variant, sf=sf)
for i in range(20):
print(i)
img_lq = deg_fn(img)
print(img_lq)
img_lq_bicubic = albumentations.SmallestMaxSize(max_size=h, interpolation=cv2.INTER_CUBIC)(image=img)["image"]
print(img_lq.shape)
print("bicubic", img_lq_bicubic.shape)
print(img_hq.shape)
lq_nearest = cv2.resize(util.single2uint(img_lq), (int(sf * img_lq.shape[1]), int(sf * img_lq.shape[0])),
interpolation=0)
lq_bicubic_nearest = cv2.resize(util.single2uint(img_lq_bicubic), (int(sf * img_lq.shape[1]), int(sf * img_lq.shape[0])),
interpolation=0)
img_concat = np.concatenate([lq_bicubic_nearest, lq_nearest, util.single2uint(img_hq)], axis=1)
util.imsave(img_concat, str(i) + '.png')
if __name__ == "__main__":
print("hey")
img = util.imread_uint("utils/test.png", 3)
print(img)
img = util.uint2single(img)
print(img)
img = img[:448, :448]
h = img.shape[0] // 4
print("resizing to", h)
sf = 4
deg_fn = partial(degradation_bsrgan_variant, sf=sf)
for i in range(20):
print(i)
img_lq = deg_fn(img)
print(img_lq)
img_lq_bicubic = albumentations.SmallestMaxSize(max_size=h, interpolation=cv2.INTER_CUBIC)(image=img)["image"]
print(img_lq.shape)
print("bicubic", img_lq_bicubic.shape)
print(img_hq.shape)
lq_nearest = cv2.resize(
util.single2uint(img_lq), (int(sf * img_lq.shape[1]), int(sf * img_lq.shape[0])), interpolation=0
)
lq_bicubic_nearest = cv2.resize(
util.single2uint(img_lq_bicubic), (int(sf * img_lq.shape[1]), int(sf * img_lq.shape[0])), interpolation=0
)
img_concat = np.concatenate([lq_bicubic_nearest, lq_nearest, util.single2uint(img_hq)], axis=1)
util.imsave(img_concat, str(i) + ".png")