lavis/configs/models/blip2/blip2_caption_flant5xl.yaml

 # Copyright (c) 2022, salesforce.com, inc.
 # All rights reserved.
 # SPDX-License-Identifier: BSD-3-Clause
 # For full license text, see the LICENSE file in the repo root or https://opensource.org/licenses/BSD-3-Clause

model:
  arch: caption_coco_flant5xl
  load_finetuned: True

  pretrained: "https://storage.googleapis.com/sfr-vision-language-research/LAVIS/models/BLIP2/blip2_pretrained_flant5xl.pth"
  finetuned: "https://storage.googleapis.com/sfr-vision-language-research/LAVIS/models/BLIP2/blip2_caption_flant5xl.pth"

  # vit encoder
  image_size: 364
  drop_path_rate: 0
  use_grad_checkpoint: False
  vit_precision: "fp32"
  freeze_vit: False

  # Q-Former
  num_query_token: 32

  # T5
  t5_model: "google/flan-t5-xl"

  # generation configs
  prompt: "a photo of"


preprocess:
    vis_processor:
        train:
          name: "blip_image_train"
          image_size: 364
        eval:
          name: "blip_image_eval"
          image_size: 364
    text_processor:
        train:
          name: "blip_caption"
        eval:
          name: "blip_caption"