Closed wanboyang closed 3 years ago
Here are Curves of CIDEr Scores, Losses and Rewards. Yellow line and blue line indicates no train_beam_size and train beam size, respectively. CIDEr Loss Reward
This is expected. You need to use a lower learning rate.
Even though, the performance is better with random sampling. (I didn't suggest to use it, I just provided this option)
Thanks for replying.
I use scripts :
python train.py --id fc_tdsc --cfg configs/fc_rl.yml --train_sample_method greedy --train_beam_size 5 --max_epochs 30 --learning_rate 5e-6
the RL training stage starts at 350k step, the performance corrupts.
what is learning rate you recommende?
I will try lower learning rates
I use the train_beam_size shell script recommended from https://github.com/ruotianluo/ImageCaptioning.pytorch/blob/master/ADVANCED.md to train updown model. here are my shell scripts:
However, RL train performs bad performances: