from transformers import ViltProcessor, ViltForImagesAndTextClassification
import requests
from PIL import Image

image = Image.open(requests.get("https://camo.githubusercontent.com/ffcbeada14077b8e6d4b16817c91f78ba50aace210a1e4754418f1413d99797f/687474703a2f2f696d616765732e636f636f646174617365742e6f72672f747261696e323031372f3030303030303038303333362e6a7067", stream=True).raw)
text = "The person is ahead of the cow."

processor = ViltProcessor.from_pretrained("juletxara/vilt-vsr-random")
model = ViltForImagesAndTextClassification.from_pretrained("juletxara/vilt-vsr-random")

# prepare inputs
encoding = processor(image, text, return_tensors="pt")

# forward pass
outputs = model(input_ids=encoding.input_ids, pixel_values=encoding.pixel_values.unsqueeze(0))
logits = outputs.logits
idx = logits.argmax(-1).item()
print("Predicted answer:", model.config.id2label[idx])

Training data

(to do)

Training procedure

Preprocessing

(to do)

Pretraining

(to do)

Evaluation results

(to do)

BibTeX entry and citation info

bibtex

@misc{kim2021vilt,
      title={ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision}, 
      author={Wonjae Kim and Bokyung Son and Ildoo Kim},
      year={2021},
      eprint={2102.03334},
      archivePrefix={arXiv},
      primaryClass={stat.ML}
}

@article{liu2022visual,
  title={Visual Spatial Reasoning},
  author={Liu, Fangyu and Emerson, Guy and Collier, Nigel},
  journal={arXiv preprint arXiv:2205.00363},
  year={2022}
}