File size: 3,722 Bytes
52758c1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 | """
Copyright (c) 2022 Samsung Electronics Co., Ltd.
Author:
Abhijith Punnappurath (abhijith.p@samsung.com)
Licensed under the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) License, (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at https://creativecommons.org/licenses/by-nc/4.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an
"AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and limitations under the License.
For conditions of distribution and use, see the accompanying LICENSE.md file.
"""
# no need to run this code separately
import os
import torch
import torchvision
from torchvision import transforms
import utils.data_generator as dg
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
class DatasetRAW(object):
def __init__(self, root, batch_size, patch_size, stride, wb_illum, on_cuda, which_input):
print('Inside raw data generator')
rawimgs = dg.datagenerator_raw(data_dir=os.path.join(root, which_input),
meta_dir=os.path.join(root, 'metadata_raw'),
wb_illum=wb_illum,
batch_size=batch_size, patch_size=patch_size, stride=stride)
rawimgs = torch.from_numpy(rawimgs)
rawimgs = rawimgs.permute(0, 3, 1, 2)
rawimgs = rawimgs ** (1 / 2.2)
if on_cuda:
rawimgs = rawimgs.to(device)
print('Number of patches '+str(rawimgs.shape[0]))
print('Outside raw data generator \n')
self.rawimgs = rawimgs
print('Inside sRGB data generator')
srgbimgs = dg.datagenerator_sRGB(data_dir=os.path.join(root, 'clean'), batch_size=batch_size, patch_size=patch_size,
stride=stride)
print('Number of patches '+str(srgbimgs.shape[0]))
print('Outside sRGB data generator \n')
srgbimgs = torch.from_numpy(srgbimgs)
srgbimgs = srgbimgs.permute(0, 3, 1, 2)
srgbimgs = srgbimgs / 255.0
if on_cuda:
srgbimgs = srgbimgs.to(device)
self.srgbimgs = srgbimgs
def __getitem__(self, idx):
# load images
img = self.rawimgs[idx]
target = self.srgbimgs[idx]
return img, target
def __len__(self):
return len(self.rawimgs)
if __name__ == '__main__':
data_dir = 'synth_night'
batch_size, patch_size, stride = 4, 480, 480
wb_illum = 'avg'
on_cuda = False
which_input = 'noisy_raw'
image_datasets = {
x: DatasetRAW(os.path.join(data_dir, x), batch_size, patch_size, stride, wb_illum,
on_cuda, which_input)
for x in ['val']}
dataloaders = {x: torch.utils.data.DataLoader(image_datasets[x], batch_size=batch_size,
shuffle=True, num_workers=0)
for x in ['val']}
dataset_sizes = {x: len(image_datasets[x]) for x in ['val']}
# Get a batch of training data
inputs, targets = next(iter(dataloaders['val']))
# Make a grid from batch
targets_grid = torchvision.utils.make_grid(targets)
inputs_grid = torchvision.utils.make_grid(inputs)
targets_grid = transforms.ToPILImage()(targets_grid.cpu())
inputs_grid = transforms.ToPILImage()(inputs_grid.cpu())
targets_grid.save("debug_targets_grid.png")
inputs_grid.save("debug_inputs_grid.png")
print('Done!')
|