Arjun Gupta
Arjun Gupta
Hi Zitong! Thank you very much for your prompt reply. I was very impressed by how quickly you responded to my question and updated the library -- Thanks a ton!...
Oh I see now. Sorry, I had missed line 135. Thanks so much, Zitong!
@shaunakjoshi12 were you able to answer this question through experimentation?
Thanks for your response. I used the hyper-parameters in that code (which were very very similar to yours) and observed the same results. Do you have any ideas why this...
No, I did not. The 'Training for Diversity in Image Paragraph Captioning' [code](https://github.com/lukemelas/image-paragraph-captioning) doesn't use block-trigrams for the initial cross-entropy training, it only used it for the self-critical sequence training...
Did you mean turning block-trigrams on only during evaluation and not during training? I set block-trigrams to 1 only during evaluation and not during training and got pretty much the...
Got it, thanks!
I am getting this error while using [dopamine](https://github.com/google/dopamine/blob/master/dopamine/replay_memory/circular_replay_buffer.py#L733). I tried the pip install command above but that didn't help. Any ideas?
Nevermind, fixed by upgrading gin-config: `pip install --upgrade gin-config`
I see, thanks! That was helpful. I was able to set a goal class as the goal state when using `SimpleSetup` as follows: ``` ss = og.SimpleSetup(space) goal = MyArbitraryGoal(si)...