search

lhoyer / improving_segmentation_with_selfsupervised_depth

[CVPR21] Implementation of our work "Three Ways to Improve Semantic Segmentation with Self-Supervised Depth Estimation"
244 stars 30 forks source link

can GPU 3090 run your code? #17

Closed yuheyuan closed 1 year ago

yuheyuan commented 2 years ago

I see your datasets need leftImg8bit_sequence_trainvaltest.zip (324GB) [md5],it's too big. I think my machine may spend long time to train a model. So I want to know your machine use which gpu.

lhoyer commented 2 years ago

We have used a single Nvidia V100 for the training. It should be possible to run the code with an RTX 3090 as well. The size of the dataset does not matter in this regard as the model is trained on mini-batches that are loaded from the disk while training.