Create adapter from config, load GSViT model Args: resolution: Input resolution (typically 224) chec

Convert the model into training mode while keep layers freezed.

Generate dense clips for train/val/test subsets. Args: base_data_path: Root directory containing {train,val,test}_metadata.csv

Given the start and end frame index, sample num_samples frames between the start and end with equal interval. Args: frames (tenso

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

:param x: tensor of shape [B (batch-size), N (num-patches), D (feature-dim)] :param masks: list of tensors of shape [B, K] containing indices

Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks). This is the same as the DropConnect impl I created for Ef

Perform a spatial short scale jittering on the given images and corresponding boxes. Args: images (tensor): images to perform sca

if return_last_state is True, returns (out, last_state) last_state has shape (batch, dim, dstate). Note that the gradient of the last state and pr

Computes the accuracy over the k top predictions for the specified values of k

Generate bbox and apply lambda correction.

Generate bbox and apply lambda correction.

embed_dim: output dimension for each position pos: a list of positions to be encoded: size (M,) returns: (M, D)

Given the arguemnts, load and initialize the configs. Args: args (argument): arguments includes `shard_id`, `num_shards`,

Load video content using Decord

Load video content using Decord

Load video content using Decord

Load video content using Decord

Load video content using Decord

With 50% prob, negate the value

With 50% prob, negate the value

With 50% prob, negate the value

Load video phase recognition dataset.

x: (batch, dim, seqlen) weight: (dim, width) bias: (dim,) activation: either None or "silu" or "swish" out: (batch, dim, seqlen)

Get video randaug transform. Args: input_size: The size of the input video in tuple. auto_augment: Parameters for randaug. A

Parse the following arguments for a default parser for PySlowFast users. Args: shard_id (int): shard id for the current machine. Star

Simple block wrapping a mixer class with LayerNorm/RMSNorm and residual connection. This Block has a slightly different structure co

Given scale, ratio, height and width, return sampled coordinates of the videos.

Given scale, ratio, height and width, return sampled coordinates of the videos.