Find an optimal number of split for `n`, such that `get_required_memory(math.ceil(n / n_split))` fits in target device's RAM. get_required_memory

Compute maximum similarity (or minimum distance) of each vector in `a` with all of the vectors in `b` Parameters: a: torc

Compute maximum similarity (or minimum distance) of each vector in `a` with all of the vectors in `b` Parameters: a: to

Returns available RAM on target device args: device: str or torch.device

Combination of fit() and predict() methods. This is faster than calling fit() and predict() seperately. Parameters: X: to

Combination of fit() and predict() methods. This is faster than calling fit() and predict() seperately. Parameters: X:

Returns True if an array of given byte size fits in target device. if self.safe_mode = False, this function simply compares the given byte size w

Pick k points at random in data (one row = one observation). Parameters ---------- data : ndarray Expect a rank 1 or 2 array. Ran

Picks k points in the data based on the kmeans++ method. Parameters ---------- data : torch.Tensor Expect a rank 1 or 2 array. R

Returns k samples of a random variable whose parameters depend on data. More precisely, it returns k observations sampled from a Gaussian random

Compute cosine similarity of 2 sets of vectors Parameters: a: torch.Tensor, shape: [m, n_features] b: torch.Tensor, shape

Compute cosine similarity of 2 sets of vectors Parameters: a: torch.Tensor, shape: [m, n_features] b: torch.Tensor, s

Compute euclidean similarity of 2 sets of vectors Parameters: a: torch.Tensor, shape: [m, n_features] b: torch.Tensor, sh

Compute euclidean similarity of 2 sets of vectors Parameters: a: torch.Tensor, shape: [m, n_features] b: torch.Tensor

Perform kmeans clustering Parameters: X: torch.Tensor, shape: [n_kmeans, n_samples, n_features]

Perform kmeans clustering Parameters: X: torch.Tensor, shape: [n_samples, n_features]

Predict the closest cluster each sample in X belongs to Parameters: X: torch.Tensor, shape: [n_kmeans, n_samples, n_features]

Predict the closest cluster each sample in X belongs to Parameters: X: torch.Tensor, shape: [n_samples, n_features] R