In ilangurudev/approxmapR: ApproxMAP (APPROXimate Multiple Alignment Pattern)

library(approxmapR)
library(tidyverse)

clustered_seqs <- 
  mvad %>%
    aggregate_sequences(format = "%Y-%m-%d", 
                        unit = "month", 
                        n_units = 1, 
                        summary_stats=FALSE) %>%
    cluster_knn(k = 15) 

clustered_seqs %>%
  top_n(2) %>%
  pull(df_sequences) %>%
  map(function(df_cluster){
            df_cluster %>%
              mutate(sequence = map_chr(sequence,
                                        format_sequence))
          })

n_months <- seq(1,10)

agg_dfs <- 
  tibble(n_months = n_months) %>%
  mutate(agg_df = map(n_months, function(n){
      mvad %>%
        aggregate_sequences(format = "%Y-%m-%d", 
                          unit = "month", 
                          n_units = n,
                          summary_stats = F)
    })) %>%
  mutate(summary_info = map(agg_df, generate_summary_stats_dup),
         mean_sets = map_dbl(summary_info, function(x){
           x %>% filter(type == "set") %>% pull(mean)
         }))

agg_dfs %>%
  ggplot(aes(n_months, mean_sets)) +
  geom_line()


agg_df <-
  mvad %>% 
    aggregate_sequences(format = "%Y-%m-%d", 
                        unit = "month", 
                        n_units = 3)

k <- seq(1, 50, 3)

k_vals <-
tibble(k = k) %>%
  mutate(cluster_ws = map(k, function(k){
    agg_df %>%
      cluster_knn(k = k) %>%
        filter_pattern(threshold = 0.4)
  }))

k_vals %>%
  mutate(num_clusters = map_int(cluster_ws, ~nrow(.))) %>% 
  ggplot(aes(k, num_clusters)) + geom_line()

ilangurudev/approxmapR documentation built on March 22, 2022, 1:15 p.m.