Visual Document Retrieval
ColPali
Safetensors
English
vidore
vidore-experimental

Unrecognized image processor in vidore/colqwen2.5-v0.2.

#3
by jjovalle99 - opened

Hello @QuentinJG !

Sorry to bother you again. I am getting the following error:

ValueError: Unrecognized image processor in vidore/colqwen2.5-v0.2. Should have a `image_processor_type` key in its preprocessor_config.json of config.json, or one of the following `model_type` keys in its config.json: align, aria, beit, bit, blip, blip-2, bridgetower, chameleon, chinese_clip, clip, clipseg, conditional_detr, convnext, convnextv2, cvt, data2vec-vision, deformable_detr, deit, depth_anything, depth_pro, deta, detr, dinat, dinov2, donut-swin, dpt, efficientformer, efficientnet, flava, focalnet, fuyu, git, glpn, got_ocr2, grounding-dino, groupvit, hiera, idefics, idefics2, idefics3, ijepa, imagegpt, instructblip, instructblipvideo, kosmos-2, layoutlmv2, layoutlmv3, levit, llava, llava_next, llava_next_video, llava_onevision, mask2former, maskformer, mgp-str, mllama, mobilenet_v1, mobilenet_v2, mobilevit, mobilevitv2, nat, nougat, oneformer, owlv2, owlvit, paligemma, perceiver, pix2struct, pixtral, poolformer, pvt, pvt_v2, qwen2_5_vl, qwen2_vl, regnet, resnet, rt_detr, sam, segformer, seggpt, siglip, superglue, swiftformer, swin, swin2sr, swinv2, table-transformer, timesformer, timm_wrapper, tvlt, tvp, udop, upernet, van, videomae, vilt, vipllava, vit, vit_hybrid, vit_mae, vit_msn, vitmatte, xclip, yolos, zoedepth

This is the code:

model_name = "vidore/colqwen2.5-v0.2"
processor = ColQwen2_5_Processor.from_pretrained(
    pretrained_model_name_or_path=model_name
)

Model loading works fine tho, the issue is with the processor.

ILLUIN Vidore org

I'll check what is going on, if you installed transformers 4.49.0 it probably comes from that as we have not tested it with that version yet and everything worked fine before.

ILLUIN Vidore org
edited 2 days ago

It's a known error since the latest release of transformers, i am testing everything and updating the colqwen2_5 branch once I checked the fix didn't break anything

ILLUIN Vidore org

It should be all good now if you pull the latest version of the colqwen2.5 branch and use transformers 4.49.0 (and update your models, i had to change the configs). Don't hesitate to tell me if something is off.

it works now, thanks you @QuentinJG !
I am just getting this warning, not sure if it matters or not:

Using a slow image processor as `use_fast` is unset and a slow processor was saved with this model. `use_fast=True` will be the default behavior in v4.48, even if the model was saved with a slow processor. This will result in minor differences in outputs. You'll still be able to use a slow processor with `use_fast=False`.
ILLUIN Vidore org

Actually not sure too, I leave it like that for now.

QuentinJG changed discussion status to closed

Sign up or log in to comment