Zafir Stojanovski comments

Repositories
Issues
Comments

Results 4 comments of


                                            Zafir Stojanovski

Community contribution: enable dynamic resolution input for more vision models.

@amyeroberts Doesn't `idefics2` already handle this? https://github.com/huggingface/transformers/blob/cf7bed98325a0be9d195cb6b66c6a0bef9fccbc8/src/transformers/models/idefics2/modeling_idefics2.py#L139-L149 For example, the following sample script: ```py import torch import requests from PIL import Image from transformers import Idefics2Processor, Idefics2ForConditionalGeneration device = torch.device("cuda"...

Community contribution: enable dynamic resolution input for more vision models.

Opened a PR (#30722) addressing this issue for the BLIP family of models (BLIP, BLIP2, InstructBLIP).

Blip dynamic input resolution

@amyeroberts good point! the tests now also have a check for the textual content.

Print "higher_is_better" in results table

@haileyschoelkopf I believe this issue should be closed now.