Run Agglomerative Merging to target a range of block numbers

Run Agglomerative Merging to target a range of block numbers

collapse_run(
  sbm,
  num_final_blocks = 1:10,
  num_mcmc_sweeps = 10,
  sigma = 2,
  eps = 0.1,
  num_block_proposals = 5,
  parallel = FALSE
)

Arguments

sbm	sbm_network object as created by new_sbm_network.
num_final_blocks	Array of integers corresponding to number of blocks to check in run.
num_mcmc_sweeps	How many MCMC sweeps the model does at each agglomerative merge step. This allows the model to allow nodes to find their most natural resting place in a given collapsed state. Larger values will slow down runtime but can potentially lead for more stable results.
sigma	Controls the rate of collapse. At each step of the collapsing the model will try and remove current_num_nodes(1 - 1/sigma) nodes from the model. So a larger sigma means a faster collapse rate.
eps	Controls randomness of move proposals. Effects both the block merging and mcmc sweeps.
num_block_proposals	Controls how many merger proposals are drawn for each block in the model. A larger number will increase the exploration of merge potentials but may lead the model to local minimums. If the number of proposals is greater than then number of blocks then all blocks are searched exhaustively.
parallel	Run in parallel using furrr?

Value

An S3 object of class sbm_network. For details see new_sbm_network section "Class structure."

See also

Other modeling: choose_best_collapse_state(), collapse_blocks(), entropy(), interblock_edge_counts(), mcmc_sweep(), n_blocks(), state()

Examples

set.seed(42)

# Start with a random network of two blocks with 25 nodes each
net <- sim_basic_block_network(n_blocks = 3, n_nodes_per_block = 25) %>%
  collapse_run(num_final_blocks = 1:5, sigma = 1.5)

# Collapse runs can be done in parallel
if (FALSE) {
  net <- sim_basic_block_network(n_blocks = 3, n_nodes_per_block = 25) %>%
    collapse_run(num_final_blocks = 1:5, sigma = 1.5, parallel = TRUE)
}

# We can look directly at the collapse results
net %>% get_collapse_results()
#> # A tibble: 5 x 3
#>   entropy n_blocks state            
#>     <dbl>    <int> <list>           
#> 1   2227.        1 <df[,4] [75 × 4]>
#> 2   2143.        2 <df[,4] [75 × 4]>
#> 3   2033.        3 <df[,4] [75 × 4]>
#> 4   2035.        4 <df[,4] [75 × 4]>
#> 5   2029.        5 <df[,4] [75 × 4]>

# We can visualize the collapse results
net %>% visualize_collapse_results()
#> Warning: Can't use entropy delta for collapse score because collapse results doesn't report entropy delta for results from collapse_run(). Using entropy instead.

# We can choose best result with default heuristic
net <- choose_best_collapse_state(net, verbose = TRUE, use_entropy = TRUE)
#> Choosing collapse with 3 blocks and an entropy of 2032.8128.