Gabriel Oliveira

icon indicating a website link https://gabrielsantosrv.github.io/

icon company University of Campinas icon location PhD Student at the Artificial Intelligence Lab., Recod.ai, Institute of Computing, University of Campinas, Brazil. (He/Him)

Results 5 comments of Gabriel Oliveira

Visualization of attention regions

> Hello everyone, > same question here: how do I do it during the decoding phase? During decoding, we have multiple attention weights, since there are several heads and also...

I can't train model using 2 GPUs

@PepZhu I couldn't solve it yet...

Any plan to share the pre-trained models?

> Same question. I'm looking for a model that can perform retrieval as in CLIP or ALIGN and also caption images so would be good if the pretrained weights were...

empirical_sinkhorn_divergence doesn't have a grad_fn

Since I am using a GPU to run my experiments, do you think it is better use `ot.sinkhorn2` with parameter `method="sinkhorn_log"` to compute the sinkhorn divergence?

empirical_sinkhorn_divergence doesn't have a grad_fn

That's ok, thank you :D