search

nv-tlabs / datasetGAN_release

MIT License
340 stars 43 forks source link

Training deeplab on higher resolution images #39

Closed AbdouMechraoui closed 2 years ago

AbdouMechraoui commented 2 years ago

Hi!

Thanks once again for your responsiveness :)

May I ask about the procedure used to train deeplab-v3 on 1024 images, I noticed the provided code in the repo samples train_interpreter.py, and trains train_deeplab.py on images with 512 x 512 resolution for face_34 task.

When doing the same for 1024 x 1024 images, the cross_validation script results in this:

trained_deeplab_1024_500

Could you provide more information on the procedure for training deeplab on 1024 x 1024 images? How large was the training set that you used? The number of epochs, and batch size? I ask the latter, as I also had some issues with OOM cuda error, when training on 32G NVIDIA Tesla V100.

arieling commented 2 years ago

Hi. In datasetGAN paper, we used 512 x 512 resolution. We didn't scale it up to 1024* 1024.

AbdouMechraoui commented 2 years ago

I am closing this, I was able to train deeplab using ~500 images/labels (1024, 1024), 30 epochs, and batch_size=4. Thanks anyways :)