Daniel Severo

I think the generator version will consume less memory, but will be slower (since the computer itself is simple and generators have some overhead). So that would be interesting to...

Not sure it's that simple. "The best solution" doesn't exist. The example above illustrates this: one consumes less RAM the other CPU time.

@lmxhappy I recently had the same issue and I've started a self-study program directed towards Generative Models in general, but I ended up focusing on GANs for now. You can...

Did you notice it bugs if you change the citation format? It only outputs the first result. ```bash ./scholar.py --phrase "Online Clustering of Bandits" --citations-only --citation bt ``` ```tex @inproceedings{kawale2015efficient,...

I can start on this in about 2 weeks if nobody else gets on it :)

What do you guys think of using [pytorch-lightning](https://github.com/williamFalcon/pytorch-lightning) to refactor the trainer?

I vote for fully dropping the current `Trainer` only on a new major release. How about the following: Currently we have: - `BaseTrainer()` - `Trainer(BaseTrainer)` - `ParallelTrainer(BaseTrainer)` We should move...

Great. Also, this NN is usually used for generating time-series data. As far as I've seen `torchgan.trainer.Trainer` specializes in images. [I had to hack it a bit](https://github.com/dsevero/torchgan/commit/ad489f869a16c6a7aed880d49dc532b652fc6cd9) so that the...

That was my next suggestion actually. GANs did focus on image generation in the beginning, but I think it's safe to say that the scenario has changed. Last thing, what...

I see. What's the difference between this repo. and `dask-glm`? Can't the later be a subset of the former?