Nan is encountered when training imagenet_rnn_gate_rl_50

[Willy0919]

I have met a problem when training imagenet_rnn_gate_rl_50 using offered pretrained sp model:

10-26-18 02:30:Epoch: [4][4010/5004] Time 1.793 (1.839) Data 0.000 (0.003) Loss nan (nan) Total rewards nan (nan) Prec@1 0.391 (75.964) Prec@5 1.953 (91.181) 10-26-18 02:30:total gate rewards = 2.560 10-26-18 02:30:*** Computation Percentage: 97.532 % 10-26-18 02:30:Epoch: [4][4020/5004] Time 1.754 (1.838) Data 0.000 (0.003) Loss nan (nan) Total rewards nan (nan) Prec@1 0.000 (75.775) Prec@5 0.000 (90.954)

I didn't change any default configure and the loss and rewards became "nan", is there any other matters needing attention. Please help

Thanks, Willy

[xinw1012]

Really sorry about the late reply. Not sure if you have already solved the issue... Did you try to reduce the learning rate? If this is still bothering you, I can take a look at the code and re-run the experiments for checking.

Thanks, Xin