TuragaLab / DECODE

This is the official implementation of our publication "Deep learning enables fast and dense single-molecule localization with high accuracy" (Nature Methods)
GNU General Public License v3.0
89 stars 26 forks source link

Running it on a CPU (since I don't have CUDA-enabled GPU) #139

Closed tsuijenk closed 3 years ago

tsuijenk commented 3 years ago

Hello,

Since I don't have a CUDA-enabled GPU, I have to run it on my CPU for now. I am following this tutorial: https://pytorch.org/tutorials/recipes/recipes/save_load_across_devices.html

I would like some confirmation to see whether or not I am doing it correctly. Any advice and guidance is appreciated! :)

** refers to the newly updated lines for running it on CPU.

In Load Model and Parameters,

ckpt_file = 'sample_data/colab_fit_experimental_rc/ckpt.pt' #@param {type:"string"}

**device = torch.device('cpu')**

ckpt = torch.load(ckpt_file, map_location=**device**)

In Setup Preprocessing Pipeline,


# hardware
#device = 'cuda:0'
**device = 'cpu'**