These are similarity counterparts of aggregated_distances
, which
see. Generally similarities are computed from their distance counterparts
through s=1-d
or s=1/(1+d)
, but this need not always be the
case. See the source code for the exact computation.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 | asim_centroid(x, y = NULL, vecs, ptrans = NULL, strans = NULL,
by = c("primary", "secondary", "row", "column"), precompute = TRUE,
dist_type = "cosine")
asim_semantic_min_sum(x, y = NULL, vecs, ptrans = NULL, strans = NULL,
by = c("primary", "secondary", "row", "column"), precompute = TRUE,
dist_type = "cosine")
asim_semantic_min_max(x, y = NULL, vecs, ptrans = NULL, strans = NULL,
by = c("primary", "secondary", "row", "column"), precompute = TRUE,
dist_type = "cosine")
asim_rwmd(x, y = NULL, vecs, ptrans = NULL, strans = NULL,
by = c("primary", "secondary", "row", "column"), precompute = TRUE,
dist_type = "cosine")
|
x |
sparse or dense objects supported by |
y |
sparse or dense objects supported by |
vecs |
Dense matrix with columns |
ptrans |
Primary and secondary transformations. Can be either a
function, string or a numeric vector. When a function, it must take 3
arguments - an object supported by |
strans |
Primary and secondary transformations. Can be either a
function, string or a numeric vector. When a function, it must take 3
arguments - an object supported by |
by |
Dimension along which to perform distance computation. For all supported data structures computation along primary dimension is more or as efficient than along the secondary dimension. |
precompute |
logical Weather to optimize the computation for speed and
precompute individual distances. The computation is method specific bug
generally should be |
dist_type |
distance to use across individual vectors in |
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.