|
from opencompass.multimodal.models.minigpt_4 import ( |
|
MiniGPT4COCOCaotionPromptConstructor, |
|
MiniGPT4COCOCaptionPostProcessor, |
|
) |
|
|
|
|
|
val_pipeline = [ |
|
dict(type='mmpretrain.LoadImageFromFile'), |
|
dict(type='mmpretrain.ToPIL', to_rgb=True), |
|
dict(type='mmpretrain.torchvision/Resize', |
|
size=(384, 384), |
|
interpolation=3), |
|
dict(type='mmpretrain.torchvision/ToTensor'), |
|
dict(type='mmpretrain.torchvision/Normalize', |
|
mean=(0.48145466, 0.4578275, 0.40821073), |
|
std=(0.26862954, 0.26130258, 0.27577711)), |
|
dict(type='mmpretrain.PackInputs', algorithm_keys=['image_id']) |
|
] |
|
|
|
dataset = dict(type='mmpretrain.Flickr30kCaption', |
|
data_root='data/flickr30k', |
|
ann_file='annotations/dataset_flickr30k.json', |
|
data_prefix='images', |
|
split='val', |
|
pipeline=val_pipeline) |
|
|
|
minigpt_4_flickr30k_dataloader = dict(batch_size=1, |
|
num_workers=4, |
|
dataset=dataset, |
|
collate_fn=dict(type='pseudo_collate'), |
|
sampler=dict(type='DefaultSampler', |
|
shuffle=False)) |
|
|
|
|
|
minigpt_4_flickr30k_model = dict( |
|
type='minigpt-4', |
|
low_resource=False, |
|
img_size=384, |
|
llama_model='/path/to/vicuna_weights_7b/', |
|
is_caption_task=True, |
|
prompt_constructor=dict(type=MiniGPT4COCOCaotionPromptConstructor, |
|
image_prompt='###Human: <Img><ImageHere></Img>', |
|
reply_prompt='###Assistant:'), |
|
post_processor=dict(type=MiniGPT4COCOCaptionPostProcessor)) |
|
|
|
|
|
minigpt_4_flickr30k_evaluator = [ |
|
dict( |
|
type='mmpretrain.COCOCaption', |
|
ann_file='data/flickr30k/annotations/flickr30k_val_gt.json', |
|
) |
|
] |
|
|
|
minigpt_4_flickr30k_load_from = '/path/to/prerained_minigpt4_7b.pth' |
|
|
