DIM() is to dim() as NROW() is to nrow(). That is, it is identical to
125 | dim() in most cases except if the input is a bare atomic vector with no
126 | dim attribute, in which case, the length of the vector is returned instead
127 | of NULL.
DROP first calls base::drop and then completely removes the dim
129 | attribute if the result is a 1-d array
DIM(x) 133 | 134 | DROP(x)135 | 136 |
| x | 141 |an R vector, potentially with a dim attribute |
142 |
|---|
For DIM, the dim attribute, or if that's not found, then length(x)
For DROP an array with 2 or more axes, or a vector with no dim
149 | attributes.
157 |#> NULL#> [1] 3154 | DIM(x)#> [1] 3#> [1] 3#> int [1:3(1d)] 1 2 3#> int [1:3] 1 2 3
R/dim2.R
117 | array2.RdThese functions reshape or make an array using C-style, row-major semantics. 122 | The returned array is still R's native F-style, (meaning, the underlying 123 | vector has been reordered).
124 |array2(data, dim = length(data), dimnames = NULL) 127 | 128 | matrix2(...) 129 | 130 | dim2(x) <- value 131 | 132 | set_dim2(...)133 | 134 |
| data | 139 |what to fill the array with |
140 |
|---|---|
| dim | 143 |numeric vector of dimensions |
144 |
| dimnames | 147 |a list of dimnames, must be the same length as |
148 |
| ... | 151 |passed on to |
152 |
| x | 155 |object to set dimensions on (array or atomic vector) |
156 |
| value | 159 |a numeric (integerish) vector of new dimensions |
160 |
Other than the C-style semantics, these functions behave identically to their
166 | counterparts (array2() behaves identically to array(), `dim2<-`()
167 | to `dim<-`()). set_dim2() is just a wrapper around set_dim(..., order = "C").
See examples for a drop-in pure R replacement to reticulate::array_reshape()
181 |#> [,1] [,2] 172 | #> [1,] 1 3 173 | #> [2,] 2 4#> [,1] [,2] 174 | #> [1,] 1 2 175 | #> [2,] 3 4176 | # for a drop-in replacement to reticulate::array_reshape 177 | array_reshape <- listarrays:::array_reshape 178 | array_reshape(1:4, c(2,2))#> [,1] [,2] 179 | #> [1,] 1 2 180 | #> [2,] 3 4
A pipe-friendly wrapper for dim(x) <- NULL and dimnames(x) <- NULL or, if
121 | which_dim is not NULL, dimnames(x)[which_dim] <- list(NULL)
drop_dimnames(x, which_dim = NULL, keep_axis_names = FALSE) 125 | 126 | drop_dim(x) 127 | 128 | drop_dim2(x)129 | 130 |
| x | 135 |an object, potentially with dimnames |
136 |
|---|---|
| which_dim | 139 |If |
141 |
| keep_axis_names | 144 |TRUE or FALSE, whether to preserve the axis names when 145 | dropping the dimnames |
146 |
This is the inverse operation of base::drop().
122 | It is analogous to python's numpy.expand_dims(), but vectorized on
123 | which_dim.
expand_dims(x, which_dim = -1L)127 | 128 |
| x | 133 |an array. Bare vectors are treated as 1-d arrays. |
134 |
|---|---|
| which_dim | 137 |numeric. Desired index position of the new axis or axes in 138 | the returned array. Negative numbers count from the back. Can be any 139 | length.Throws a warning if any duplicates are provided. |
140 |
the array x with new dim
155 |#> [1] 2 3 4#> [1] 2 3 4 1#> [1] 2 1 3 4#> [1] 1 1 2 3 4#> [1] 1 2 3 4 1#> [1] 2 3 4 1 1 1#> [1] 2 3 4 1 1 1#> Error in expand_dims(x, -6) : 152 | #> Implicit additional dims for expansion with negative indexes not supported#> [1] 1 1 1 2 3 4
R/modify.R
117 | modify_along_dim.RdThis function can be thought of as a version of base::apply() that is
122 | guaranteed to return a object of the same dimensions as it was input. It also
123 | generally preserves attributes, as it's built on top of [<-.
modify_along_dim(X, which_dim, .f, ...) 127 | 128 | modify_along_rows(X, .f, ...) 129 | 130 | modify_along_cols(X, .f, ...)131 | 132 |
| X | 137 |An array, or a list of arrays |
138 |
|---|---|
| which_dim | 141 |integer vector of dimensions to modify at |
142 |
| .f | 145 |a function or formula defining a function(same semantics as
146 | |
149 |
| ... | 152 |passed on to |
153 |
An array, or if X was a list, a list of arrays of the same shape as
159 | was passed in.
192 |#> , , 1 164 | #> 165 | #> [,1] [,2] 166 | #> [1,] 1.5 1.5 167 | #> 168 | #> , , 2 169 | #> 170 | #> [,1] [,2] 171 | #> [1,] 3.5 3.5 172 | #> 173 | #> , , 3 174 | #> 175 | #> [,1] [,2] 176 | #> [1,] 5.5 5.5 177 | #>#> , , 1 178 | #> 179 | #> [,1] [,2] 180 | #> [1,] 0.6666667 1.333333 181 | #> 182 | #> , , 2 183 | #> 184 | #> [,1] [,2] 185 | #> [1,] 0.8571429 1.142857 186 | #> 187 | #> , , 3 188 | #> 189 | #> [,1] [,2] 190 | #> [1,] 0.9090909 1.090909 191 | #>
Sequence along a dimension
120 |seq_along_dim(x, which_dim) 123 | 124 | seq_along_rows(x) 125 | 126 | seq_along_cols(x)127 | 128 |
| x | 133 |a dataframe, array or vector. For |
138 |
|---|---|
| which_dim | 141 |a scalar integer or character string, specifying which 142 | dimension to generate a sequence for. Negative numbers count from the back. |
143 |
a vector of integers 1:nrow(x), safe for use in for loops and
149 | vectorized equivalents.
164 |#> mpg cyl disp hp drat wt qsec vs am gear carb 154 | #> Mazda RX4 21 6 160 110 3.9 2.62 16.46 0 1 4 4 155 | #> mpg cyl disp hp drat wt qsec vs am gear carb 156 | #> Mazda RX4 Wag 21 6 160 110 3.9 2.875 17.02 0 1 4 4 157 | #> mpg cyl disp hp drat wt qsec vs am gear carb 158 | #> Datsun 710 22.8 4 108 93 3.85 2.32 18.61 1 1 4 1 159 | #> mpg cyl disp hp drat wt qsec vs am gear carb 160 | #> Hornet 4 Drive 21.4 6 258 110 3.08 3.215 19.44 1 0 3 1#> [1] TRUE163 |
R/reshape.R
115 | set_as_rows.RdReshape an array to send a dimension forward or back
120 |set_as_rows(X, which_dim) 123 | 124 | set_as_cols(X, which_dim)125 | 126 |
| X | 131 |an array |
132 |
|---|---|
| which_dim | 135 |scalar integer or string, which dim to bring forward. 136 | Negative numbers count from the back 137 |This is a powered by |
138 |
a reshaped array
144 |base::aperm() set_dim() keras::array_reshape()
154 |
R/shuffle.R
115 | shuffle_rows.RdShuffle along the first dimension multiple arrays in sync
120 |shuffle_rows(...)123 | 124 |
| ... | 129 |arrays of various dimensions (vectors and data.frames OK too) |
130 |
|---|
A list of objects passed on to ..., or if a single object was
136 | supplied, then the single object shuffled
157 |x <- 1:3 140 | y <- matrix(1:9, ncol = 3) 141 | z <- array(1:27, c(3,3,3)) 142 | 143 | if(require(zeallot)) { 144 | c(xs, ys, zs) %<-% shuffle_rows(x, y, z) 145 | 146 | l <- lapply(seq_along_rows(y), function(r) { 147 | list(x = x[r], y = y[r,], z = z[r,,]) 148 | }) 149 | 150 | ls <- lapply(seq_along_rows(y), function(r) { 151 | list(x = xs[r], y = ys[r,], z = zs[r,,]) 152 | }) 153 | 154 | stopifnot( 155 | length(unique(c(l, ls))) == length(l)) 156 | }
transpose an array
120 |# S3 method for array 123 | t(x)124 | 125 |
| x | 130 |an array 131 |This reverses the dimensions of an array |
132 |
|---|
142 |