tcrossprod.HDF5Matrix

tcrossprod.HDF5Matrix

DECOMPOSITIONS

1 Usage

tcrossprod.HDF5Matrix(x, y = NULL, outgroup = NULL, outdataset = NULL, ...)

2 Arguments

Parameter	Description
`x`	An object.
`y`	An object, or (default) to compute .
`outgroup`	Character or . HDF5 group where the result is stored. Default .
`outdataset`	Character or . Dataset name for the result. Default (single matrix) or (two matrices).

3 Value

A new pointing to the result dataset.

4 Details

Computes (or when y = NULL). Uses the dedicated BigDataStatMeth block-wise transposed cross-product algorithm, which is more efficient than explicitly computing x \%*\% t(y).

Performance settings:

This method uses global options set via hdf5matrix_options.

Symmetric optimization:

When y = NULL or y refers to the same dataset as x, the symmetric optimisation is applied automatically, providing significant speedup.

5 Examples

\donttest{
fn <- tempfile(fileext = ".h5")
X <- hdf5_create_matrix(fn, "INPUT/X", data = matrix(rnorm(60), 6, 10))
Y <- hdf5_create_matrix(fn, "INPUT/Y", data = matrix(rnorm(60), 6, 10))

# t(X) %*% X  → stored in OUTPUT/CrossProd_X
C1 <- tcrossprod(X)
dim(C1)

# t(X) %*% Y  → stored in OUTPUT/CrossProd_X_x_Y
C2 <- tcrossprod(X, Y)

# Custom output location
C3 <- tcrossprod(X, outgroup = "RESULTS", outdataset = "my_tcrossprod")

hdf5_close_all()
unlink(fn)
}

6 See Also

hdf5matrix_options for global performance settings

---
title: "tcrossprod.HDF5Matrix"
subtitle: "tcrossprod.HDF5Matrix"
---

<span class="category-badge decompositions">DECOMPOSITIONS</span>

## Usage

```r
tcrossprod.HDF5Matrix(x, y = NULL, outgroup = NULL, outdataset = NULL, ...)
```
## Arguments

::: {.param-table}
| Parameter | Description |
|-----------|-------------|
| `x` | An \code{HDF5Matrix} object. |
| `y` | An \code{HDF5Matrix} object, or \code{NULL} (default) to   compute \code{x \%*\% t(x)}. |
| `outgroup` | Character or \code{NULL}. HDF5 group where the result is   stored. Default \code{"OUTPUT"}. |
| `outdataset` | Character or \code{NULL}. Dataset name for the result.   Default \code{"tCrossProd_x"} (single matrix) or   \code{"tCrossProd_x_x_y"} (two matrices). |
:::

## Value

::: {.return-value}
A new \code{HDF5Matrix} pointing to the result dataset.

:::
## Details

Computes \eqn{x \times t(y)} (or \eqn{x \times t(x)} when `y = NULL`).
Uses the dedicated BigDataStatMeth block-wise transposed cross-product
algorithm, which is more efficient than explicitly computing `x \%*\% t(y)`.

**Performance settings:**

This method uses global options set via [hdf5matrix_options](../hdf5matrix_core/hdf5matrix_options.html).

**Symmetric optimization:**

When `y = NULL` or `y` refers to the same dataset as `x`,
the symmetric optimisation is applied automatically, providing significant speedup.

## Examples
```{r}
#| eval: false
#| warning: false

\donttest{
fn <- tempfile(fileext = ".h5")
X <- hdf5_create_matrix(fn, "INPUT/X", data = matrix(rnorm(60), 6, 10))
Y <- hdf5_create_matrix(fn, "INPUT/Y", data = matrix(rnorm(60), 6, 10))

# t(X) %*% X  → stored in OUTPUT/CrossProd_X
C1 <- tcrossprod(X)
dim(C1)

# t(X) %*% Y  → stored in OUTPUT/CrossProd_X_x_Y
C2 <- tcrossprod(X, Y)

# Custom output location
C3 <- tcrossprod(X, outgroup = "RESULTS", outdataset = "my_tcrossprod")

hdf5_close_all()
unlink(fn)
}

```
## See Also

::: {.see-also}
[hdf5matrix_options](../hdf5matrix_core/hdf5matrix_options.html) for global performance settings

:::