search

Inha-CVAI / MADGNet

CVPR2024 Accepted Paper
52 stars 2 forks source link

Request for Training Code #6

Open fruit2rbj opened 2 weeks ago

fruit2rbj commented 2 weeks ago

Thank you for sharing the code, but could you please provide the train.py

skawngus1111 commented 2 weeks ago

Thank you for the question! Our training code (train.py) is based on pranet.

https://github.com/DengPingFan/PraNet

waner1998 commented 1 week ago

Thank you for the question! Our training code (train.py) is based on pranet.

https://github.com/DengPingFan/PraNet

Thank you for your wonderful work. I saw the following code in the PraNet code:

size_rates = [0.75, 1, 1.25] for rate in size_rates: trainsize = int(round(opt.trainsizerate/32)32) if rate != 1: images = F.upsample(images, size=(trainsize, trainsize), mode='bilinear', align_corners=True) gts = F.upsample(gts, size=(trainsize, trainsize), mode='bilinear', align_corners=True)

May I ask if you also used the training strategy of scaling down or scaling up the training images in this work? Looking forward to your reply. Sincerely, Liu.

skawngus1111 commented 6 days ago

@waner1998 Thank you for reply. Yes! We used same multi-scale training strategy with PraNet.