golnazghiasi / LRR

code and models for "Laplacian Pyramid Reconstruction and Refinement for Semantic Segmentation"
59 stars 14 forks source link

error on fine tuning LRR with multi-GPU mode #6

Open Mirsadeghi opened 7 years ago

Mirsadeghi commented 7 years ago

I try to fine tune LRR on my dataset an I set the training to the multi-gpu mode by setting : opts.train.gpus = [1:3] same as fcnTrain.m function which works on multi-gpu mode in this way. But i got error while training process want to use "getBatch" function. How can i resolve this error and use multi-gpu mode?