Claudiu Daniel Hromei
Claudiu Daniel Hromei
Hi, I'm interested in this topic too, but I cannot figure out how to use the code you provided (neither if I can) for fine-tuning kosmos2 on my downstream task....
> Hi, I got knn finally compiled with ninja. I use cmd in anaconda terminal. The problem is in cpp_extension of torch. Thanks! Ying > […](#) > On Mon, May...
Hi I have this same problem, any solution?
Thank you @ANABUR920 , you solved part of my problem, but why does this work?
No news, still waiting for someone to answer. It looks like the `processor` did not return any `labels` and the `Trainer` could not compute the loss...
Thank you for the tip. I think I already figured that out and I'm trying to prepare the labels before Training, but I didn't think about masking the input prompt....
Thank you @ydshieh for your invaluable help! As an example, what I want to do is *Image Grounded Command Interpretation*, i.e. I have an image and a command from a...
Hi @ydshieh I figured out how to change the Kosmos Processor in order to add those extra tokens and I started fine-tuning the model on my downstream task. But there...
> The whole model is finetuned unless you do something particular to freeze the vision encoder part of it. (although, I would say, in your case, it might make more...
@ydshieh I added you to a private repository containing my notebook and a sample of data to run the code. Let me know what you think and how we can...