search

limbee / NTIRE2017

Torch implementation of "Enhanced Deep Residual Networks for Single Image Super-Resolution"
651 stars 146 forks source link

why does the inference so slow #5

Closed realbns2008 closed 7 years ago

realbns2008 commented 7 years ago

hi , i'm using the test.lua script in demo to inference a image. the model i used is EDSR_x2.t7 the gpu i used is Tesla K80 the image i used is 180x100

i type the following command to run the script: th test.lua -type test -model EDSR_x2.t7 -dataset img_input -gpuid 3

it tooks about 30 seconds to inference why is it so slow, is there any problem ?

limbee commented 7 years ago

I think it's normal since EDSR_x2 is quite big. When we used the latest Titan X, it also took a lot of time. You can test it faster if you use multiple GPUs with -nGPU option.

realbns2008 commented 7 years ago

Got it . Thanks , I will try multiple gpus. :)