hacktmz

Results 13 comments of hacktmz

`import argparse from pathlib import Path import torch from torch.nn import CrossEntropyLoss from torch.utils.data import DataLoader from rotate_captcha_crack.common import device from rotate_captcha_crack.dataset import ImgTsSeqFromPath, RotDataset, from_google_streetview, from_pcgs_pics from rotate_captcha_crack.helper import...

添加了50000张不同钱币图片用于训练,但是loss还是只有0.6降不下去了。传统的特征点匹配的方法已经尝试了,因为每个钱币色泽和新旧磨损不一样,所以并不好用

改成RotNet LOSS就下来了