Extract a range of elements

Symbolic extraction range of elements for LazyTensor objects.

Usage

extract(x, m, d)

Arguments

x

a LazyTensor or a ComplexLazyTensor.

m

an integer corresponding to the starting index (starting from 0) of the range of elements from x that will be extracted.

d

an integer corresponding to the output dimension, i.e. the number of consecutive elements from x that will be extracted.

Value

a LazyTensor.

Details

extract(x_i, m, d) encodes, symbolically, the extraction of a range of values x[m:m+d] in the LazyTensor x; (m is the starting index, and d is the dimension of the extracted sub-vector).

IMPORTANT: IN THIS CASE, INDICES START AT ZERO, therefore, m should be in [0, n), where n is the inner dimension of x. And d should be in [0, n-m].

Note: See the examples for a more concrete explanation regarding the use of extract().

Author

Chloe Serre-Combe, Amelie Vernay

Examples

if (FALSE) {
# Two very rudimentary examples
# -----------------------------

# Let's say that you have a matrix `g` looking like this:
#      [,1] [,2] [,3] [,4]
# [1,]    1    8    1    3
# [2,]    2    1    2    7
# [3,]    3    7    4    5
# [4,]    1    3    3    0
# [5,]    5    4    9    4

# Convert it to LazyTensor:
g_i <- LazyTensor(g, index = 'i')

# Then extract some elements:
ext_g <- extract(g_i, 1, 3)

# In this case, `ext_g` is a LazyTensor encoding, symbolically,
# the following part of g:
#       [,1] [,2] [,3]
# [1,]    8    1    3
# [2,]    1    2    7
# [3,]    7    4    5
# [4,]    3    3    0
# [5,]    4    9    4


# Same principle with a LazyTensor encoding a vector:
v <- c(1, 2, 3, 1, 5)
Pm_v <- LazyTensor(v)

ext_Pm_v <- extract(Pm_v, 2, 3)

# In this case, `ext_Pm_v` is a LazyTensor encoding, symbolically,
# the following part of v:
#       [,1] [,2] [,3]
# [1,]    3    1    5


# A more general example
# ----------------------

x <- matrix(runif(150 * 5), 150, 5) # arbitrary R matrix, 150 rows, 5 columns
x_i <- LazyTensor(x, index = 'i')   # LazyTensor from matrix x, indexed by 'i'
m <- 2
d <- 2

extract_x <- extract(x_i, m, d) # symbolic matrix
}