--- a +++ b/lavis/configs/models/blip_nlvr.yaml @@ -0,0 +1,39 @@ + # 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: blip_nlvr + model_type: nlvr + load_finetuned: True + + finetuned: "https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_base_nlvr.pth" + pretrained: "https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_base_capfilt_large.pth" + + num_classes: 2 + + # vit encoder + vit_type: "base" + vit_grad_ckpt: False + vit_ckpt_layer: 0 + vit_layer_norm_epsilon: 1e-6 + + image_size: 384 + + # bert config + med_config_path: "configs/models/med_config.json" + +preprocess: + vis_processor: + train: + name: "blip_image_train" + image_size: 384 + eval: + name: "blip_image_eval" + image_size: 384 + text_processor: + train: + name: "blip_caption" + eval: + name: "blip_caption"