In Xinzhu-Fang/shinyNDTr: A web-based interface for neural decoding in R

knitr::opts_chunk$set(echo = TRUE)

binned_file_name <-'/student/15/xf15/GitHub/shinyNDTr/data/binned/ZD_150_samples_binned_every_10_samples_start_-300_end_300.Rda'


variable_to_decode <-'combined_ID_position'


num_cv_splits <- 5


ds <- NDTr::basic_DS$new(binned_file_name, variable_to_decode, num_cv_splits)


ds$num_repeats_per_level_per_cv_split <- 2


cl <- NDTr::max_correlation_CL$new()


fps <- list()


cv <- NDTr::standard_CV$new(ds, cl, fps)


DECODING_RESULTS <- cv$run_decoding()


save('DECODING_RESULTS', file = '/student/15/xf15/GitHub/shinyNDTr/results/rmd.rda')

selected_result <- DECODING_RESULTS$zero_one_loss_results


selected_mean_results <- colMeans(selected_result)


selected_time_bin_names <- NDTr::get_center_bin_time(dimnames(selected_result)[[3]])


image.plot(selected_time_bin_names, selected_time_bin_names, selected_mean_results, legend.lab = 'Classification Accuracy', xlab = 'Test time (ms)', ylab = 'Train time (ms)')


abline(v = 0)