File size: 1,499 Bytes
4b532c0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 |
# 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: img2prompt_vqa
model_type: base
image_question_matching_model:
arch: blip_image_text_matching
load_finetuned: True
finetuned: "https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_large_retrieval_coco_train2014.pth"
# vit encoder
vit_type: "large"
vit_grad_ckpt: False
vit_ckpt_layer: 0
image_size: 384
# bert config
med_config_path: "configs/models/med_large_config.json"
embed_dim: 256
image_captioning_model:
arch: blip_caption
load_finetuned: True
finetuned: "https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_large_caption_coco_train2014.pth"
vit_type: "large"
vit_grad_ckpt: True
vit_ckpt_layer: 5
image_size: 384
# bert config
med_config_path: "configs/models/med_large_config.json"
# generation configs
prompt: "a picture of "
question_generation_moodel:
pretrained: "https://storage.googleapis.com/sfr-vision-language-research/LAVIS/projects/img2prompt/T5_large_QG.pth"
preprocess:
vis_processor:
eval:
name: "blip_image_eval"
image_size: 384
text_processor:
eval:
name: "blip_caption"
|