reconstruct {tempted} | R Documentation |
Reconstruct tensor from low dimensional components
Description
This function reconstructs the temporal tensor from the low dimensional components extracted by tempted
and svd_centralize
.
Usage
reconstruct(res_tempted, res_svd = NULL, datlist = NULL, r_reconstruct = NULL)
Arguments
res_tempted |
Output of function |
res_svd |
Output of function |
datlist |
Output of function |
r_reconstruct |
The number of components from TEMPTED to be used for the tensor reconstruction. |
Value
datlist_est The reconstructed tensor stored in a datlist format same as the output of format_tempted
.
Examples
# Take a subset of the samples so the example runs faster
# Here we are taking samples from the odd months
sub_sample <- rownames(meta_table)[(meta_table$day_of_life%/%12)%%2==1]
count_table_sub <- count_table[sub_sample,]
processed_table_sub <- processed_table[sub_sample,]
meta_table_sub <- meta_table[sub_sample,]
# reconstruct with mean subtraction
datlist <- format_tempted(processed_table_sub,
meta_table_sub$day_of_life,
meta_table_sub$studyid,
pseudo=NULL,
transform="none")
mean_svd <- svd_centralize(datlist, r=1)
res_tempted <- tempted(mean_svd$datlist, r=2, smooth=1e-5)
datlist_est <- reconstruct(res_tempted, mean_svd, datlist=NULL, r_reconstruct=2)
vec_est <- unlist(sapply(datlist_est, function(x){x[-1,]}))
vec_obs <- unlist(sapply(datlist, function(x){x[-1,]}))
R2 <- 1-sum((vec_est-vec_obs)^2)/sum(vec_obs^2)
R2
# reconstruct without mean subtraction
datlist <- format_tempted(processed_table_sub,
meta_table_sub$day_of_life,
meta_table_sub$studyid,
pseudo=NULL,
transform="none")
res_tempted <- tempted(datlist, r=2, smooth=1e-5)
datlist_est <- reconstruct(res_tempted, res_svd=NULL, datlist=datlist, r_reconstruct=2)
vec_est <- unlist(sapply(datlist_est, function(x){x[-1,]}))
vec_obs <- unlist(sapply(datlist, function(x){x[-1,]}))
R2 <- 1-sum((vec_est-vec_obs)^2)/sum(vec_obs^2)
R2
[Package tempted version 0.1.1 Index]