Many settings affect the outputs in interesting ways, but that's half the fun. These LoRAs are very lightly trained; more training may or may not help. The competitions are also performed using zero-shot text guessing, and if Facebook said it, you can bet that's actually Meta AI saying it, and they are leaders in the field. Source: over 1 year ago

You should look at the entire niche of MAE-related papers, that's quite exciting, and the neuroscience-inspired stream of stuff like Barlow Twins. As well, the official Facebook AI blog is surprisingly good coverage of much of the interesting un/semi-supervised DL research FAIR does, and worth going through. Source: almost 2 years ago

I'm not quite sure if this is the place to ask it, but I'll give it a shot. Several years ago, during my PhD, I used to train small CNNs using NVIDIA DIGITS tool (https://developer.nvidia.com/digits), that is basically a frontend to tasks such as build datasets, configure training parameters, follow real time training data (epochs), test classification and export training for usage. This is a oversimplified... Source: over 1 year ago

Also frameworks which make moving to multiGPU easy, like DIGITS: https://developer.nvidia.com/digits. Source: almost 3 years ago