metadata

tags:
  - image-to-text
  - image-captioning
license: apache-2.0
metrics:
  - rouge
datasets:
  - Mozilla/flickr30k-transformed-captions
widget:
  - src: >-
      https://huggingface.co/datasets/mishig/sample_images/resolve/main/savanna.jpg
    example_title: Savanna
  - src: >-
      https://huggingface.co/datasets/mishig/sample_images/resolve/main/football-match.jpg
    example_title: Football Match
  - src: >-
      https://huggingface.co/datasets/mishig/sample_images/resolve/main/airport.jpg
    example_title: Airport
base_model:
  - google/vit-base-patch16-224-in21k
model-index:
  - name: mozilla/distilvit
    results:
      - task:
          type: image-to-text
          name: Image To Text
        dataset:
          name: Mozilla/flickr30k-transformed-captions
          type: Mozilla/flickr30k-transformed-captions
        metrics:
          - name: ROUGE-1
            type: rouge
            value: 43.006
            verified: true
          - name: ROUGE-2
            type: rouge
            value: 16.9939
            verified: true
          - name: ROUGE-L
            type: rouge
            value: 38.8923
            verified: true
          - name: ROUGE-LSUM
            type: rouge
            value: 38.8877
            verified: true
          - name: loss
            type: loss
            value: 0.19939416646957397
          - name: gen_len
            type: gen_len
            value: 11.327256736227712
            verified: true

distilvit

This model is a work in progress. Fine-tuned version of those base models:

a VIT model for the image encoder: https://huggingface.co/google/vit-base-patch16-224-in21k
a Distilled GPT-2 model for the text decoder: https://huggingface.co/distilbert/distilgpt2

This model was trained on:

Flickr30k debiased
DocOrNot
Alt Text Validation
A debiased version of COCO 2017: https://cocodataset.org

You can find the code used to create the model here: https://github.com/mozilla/distilvit

training results

eval/gen_len 14.99729
eval/loss 0.17093
eval/meteor 0.51479
eval/rouge1 57.8066
eval/rouge2 35.0888
eval/rougeL 52.9138
eval/rougeLsum 52.9101
eval/runtime 760.2135
eval/samples_per_second 11.18
eval/steps_per_second 0.112
train/epoch 8.0
train/global_step 11752
train/learning_rate 0.0
train/loss 0.1034
train/total_flos 1.518634875573869e+20
train/train_loss 0.14875
train/train_runtime 91405.9053
train/train_samples_per_second 12.855
train/train_steps_per_second 0.129