Get value by key at specified index (indices) if idx == None, returns value for key at each output index if idx is an integer, return

r"""Get an item from registry with key 'name' Args: name (string): Key whose value needs to be retrieved. default: If

Print information about the VQA annotation file. :return:

Build a pretrained model from default configuration file, specified by model_type. Args: - model_type (str): model type,

Assume the first argument is the name of the argument.

Args: samples (dict): A dictionary containing the following keys: - image (torch.Tensor): A tensor of shape (batc

Load checkpoint as specified in the config file. If load_finetuned is True, load the finetuned model; otherwise, load the pretrained

Load model and its related preprocessors. List all available models and types in registry: >>> from lavis.models import model_zoo >>

Utility for deleting a directory. Useful for cleaning the storage space that contains various training artifacts like checkpoints, data etc.

Load supported models. To list all available models and types in registry: >>> from lavis.models import model_zoo >>> print(model_zo

Performs all_gather operation on the provided tensors. *** Warning ***: torch.distributed.all_gather has no gradient.

Download a file from a URL and cache it locally. If the file already exists, it is not downloaded again. If distributed, only the main proces

Example >>> processor = load_processor("alpro_video_train", cfg=None)

Args: samples (dict): A dictionary containing the following keys: - image (torch.Tensor): A tensor of shape (batc

Extract features from the model for samples. Keys allowed are "image" and "text_input" in samples. If either key is missing,

r"""Fills the input Tensor with values drawn from a truncated normal distribution. The values are effectively drawn from the normal distributi

打开一个 hdfs 文件, 用 contextmanager. Args: hfdfs_path (str): hdfs文件路径 mode (str): 打开模式，支持 ["r", "w", "wa"]

vis_root (string): Root directory of images (e.g. coco/images/) ann_root (string): directory to store the annotation file spl

Class Method to Add pages to the project Args: title ([str]): The title of page which we are adding to the list of apps

Create by split datasets inheriting torch.utils.data.Datasets. # build() can be dataset-specific. Overwrite to customize.

Download a file from a url and place it in root. Args: url (str): URL to download file from root (str): Directory to place downloa

r"""Register a path to registry with key 'name' Args: name: Key with which the path will be registered. Usage:

Performs all_gather operation on the provided tensors. Graph remains connected for backward grad computation.