fast.ai Course Forums

Setting up GPU for fastai_v3

Part 1 (2019) Advanced (Part 1 v3)

Ralph November 13, 2018, 5:11pm 22

Check this thread

Lesson 3 CAMVID Half-precision issue (.to_fp16()) Advanced (Part 1 v3)

Hello, When I try to run the Big version of images in lesson3-camvid.ipynb notebook in half-precision to avoid memory problems (as I am using 1080ti): learn = Learner.create_unet(data, models.resnet34, metrics=metrics).to_fp16() everything trains fine, and I get a pretty good 0.93 accuracy. But when I call learn.show_results(), I get the following error: RuntimeError: Input type (torch.cuda.FloatTensor) and weight type (torch.cuda.HalfTensor) should be the same Any suggestions on how to fix…

show post in topic

Home
Categories
Guidelines
Terms of Service
Privacy Policy

Powered by Discourse, best viewed with JavaScript enabled