CodeExplode

I've mostly used the original TI code, though am wondering what Automatic's is missing which that settings file allows? Num_workers was a setting which could be played to bring down...

Yeah I'm still not sure if I can get as good results in this repo, have had a few successes but also some which just seem to never work, so...

Hrm yeah I've been trying to train an embedding for a character all day with Automatic's, and had no luck even undoing all my customizations (except for custom prompts). Then...

When generating an embedding with more than 1 vector, is it possible to delete vectors and see what the difference is? Maybe training with a high vector count would be...

Yeah I've tried quite a few times with 4 vectors, since early on I had one get excellent results with 46 training images but it was over-trained on closeup shots...

Another idea, similar to per_image_tokens or progressive_words, is that it could be interesting to train with multiple vectors, and have different data sets and prompts which individual vectors or sub-ranges...

I suppose in the end I'm just reinventing the original tagged images database concept there, and have realized that it would be nice to be able to train multiple embeddings...

Did you end up deciding on an answer for this blackbar? The original CLIP embedding weights only vary by about 0.12 from the lowest to the highest value per weight,...

I've had these black bars showing up for a while depending on resolution settings, but they never seem to appear if the resolution is square. Sometimes I can push the...

CPUs in general aren't as good as the type of math calculations which machine learning uses as GPUs are, which is why they're all made for CUDA. That being said...