mirror of
https://github.com/lllyasviel/ControlNet.git
synced 2026-01-12 07:28:26 -05:00
40 lines
1.1 KiB
Python
40 lines
1.1 KiB
Python
import json
|
|
import cv2
|
|
import numpy as np
|
|
|
|
from torch.utils.data import Dataset
|
|
|
|
|
|
class MyDataset(Dataset):
|
|
def __init__(self):
|
|
self.data = []
|
|
with open('./training/fill50k/prompt.json', 'rt') as f:
|
|
for line in f:
|
|
self.data.append(json.loads(line))
|
|
|
|
def __len__(self):
|
|
return len(self.data)
|
|
|
|
def __getitem__(self, idx):
|
|
item = self.data[idx]
|
|
|
|
source_filename = item['source']
|
|
target_filename = item['target']
|
|
prompt = item['prompt']
|
|
|
|
source = cv2.imread('./training/fill50k/' + source_filename)
|
|
target = cv2.imread('./training/fill50k/' + target_filename)
|
|
|
|
# Do not forget that OpenCV read images in BGR order.
|
|
source = cv2.cvtColor(source, cv2.COLOR_BGR2RGB)
|
|
target = cv2.cvtColor(target, cv2.COLOR_BGR2RGB)
|
|
|
|
# Normalize source images to [0, 1].
|
|
source = source.astype(np.float32) / 255.0
|
|
|
|
# Normalize target images to [-1, 1].
|
|
target = (target.astype(np.float32) / 127.5) - 1.0
|
|
|
|
return dict(jpg=target, txt=prompt, hint=source)
|
|
|