Sai
Sai
I was reading through the paper, and found interesting examples for MT and POS. I am little unclear on whether the current library can be used for it. If so,...
1. I would like the ability to global filter away the metadata. This can be easily achieved by nesting them under a tag. 2. It would be nice if there...
I have a view with a map, when a marker on map is clicked I want to push a detail view which has small version of map in top, I...
I tried starting training for my data in the downloaded OpenNmt with the following flags, python train.py -gpuid 2 -save_model training/models -data training/data/data -layers 6 \ -rnn_size 512 -word_vec_size 512...
Hi, Great job porting dreamer2 to torch while matching the tf code. I think the Plan2Explore has to be ported over. https://github.com/jsikyoon/dreamer-torch/blob/e42d504ea362ad605c7bbe2a0d89df1d2b3e07f2/exploration.py#L60
Can extracted patterns from error -> corrected pairs be applied to existing sentences reliably? I am interested in GEC task. Other than that can you elaborate on what these patterns...
DreamerV3 is a SOTA MbRL algorithm. It's already [written in jax](https://github.com/danijar/dreamerv3). I want to gather if anyone here is potentially interested to make it's environment interactions also into jax. ie...
It would be helpful to have wallclock comparions for the benchmarks you had posted. I think Danijar's jax implementation uses jax scan heavily to make the imagination/rollouts loops efficient.
In dreamerv2 the flag `mode=train` was passed to the posterior computation in `observe` to use the modes of the stochastic state. I notice now that we always sample. Is it...
Since `embodied/` is moved outside dreamerv3, but now it makes pip installation not work. Can we either move it back into dreamerv3 sub directory? Or did you intend to have...