Official PyTorch Implementation code for realizing the technical part of Causal Unsupervised Semantic sEgmentation (CAUSE) to improve performance of unsupervised semantic segmentation. (Under Review)
8
stars
1
forks
source link
Not Fully Unsupervised as the Labels are used to compute the total loss #12
You can find the following line in `train_front_door_tr.py` [[Link](https://github.com/ByungKwanLee/Causal-Unsupervised-Segmentation/blob/48e6459d4f9f57ba732084d230a13c0ada70c6df/train_front_door_tr.py#L210C1-L213C67)]
# Bank and EMA
cluster.bank_init()
ema_init(segment.head, segment.head_ema)
ema_init(segment.projection_head, segment.projection_head_ema
Here, the funcion of ema_init is defined in modules/segment_module.py [Link]
def ema_init(x, x_ema):
for param, param_ema in zip(x.parameters(), x_ema.parameters()): param_ema.data = param.data; param_ema.requires_grad = False
Here, the funcion of
ema_init
is defined inmodules/segment_module.py
[Link]Originally posted by @ByungKwanLee in https://github.com/ByungKwanLee/Causal-Unsupervised-Segmentation/issues/5#issuecomment-1786983385