Closed inkfish2016 closed 8 years ago
No matter the fixed lr or step lr, the loss varies little, and it is always between 1000 and 3000. Are there some other parameters affect the net a lot?
The train loss reported by caffe will be very noisy, since it is only computed on a small number of images. If you compute val loss periodically for a fairly large number of iterations (e.g. 500), this should be much more stable and be seen to decrease steadily over time.
FYI, losses between 3000 and 1000 are normal: the loss is not normalized by the number of pixels.
Thanks for your help, i get similar results in SALICON validation set as your model
Hi, I am trying to implement your paper in pytorch. I would like to know what You mean by 25000 iterations. Is it 25000 epochs or 5 epochs (since there are 10000 images in Salicon training set and ur batch size if 2 so 5000 iterations/ epoch).
Hi @Goutam-Kelam. Yes, that would be 5 epochs
Thankyou for your prompt reply.
Hi @kevinmcguinness can you help me out with the new issue which I have created relating to DeepNet in pytorch.
net:"train_deep_sal.prototxt" test_iter: 500 test_interval: 1000 display: 10 average_loss: 20 lr_policy: "step" gamma: 0.5 stepsize: 100 base_lr: 0.00000013 momentum: 0.9 iter_size: 1 max_iter: 24000 weight_decay: 0.0005 snapshot: 5000 snapshot_prefix: "train_deep_SalNet" solver_mode: GPU test_initialization: false
Is it right?