Julia script to perform the combinatorial embedding.

usage: comb.jl -d DATASET -e EPS [-r DIM] [-s] [-m EMBEDDING-SAVE]
               [-v] [-t SCALE] [-c] [-z STATS-SAMPLE] [-p PRECISION]
               [-h]

  -d, --dataset         Dataset to embed
  -e, --eps EPS         Epsilon distortion 
  -r, --dim DIM         Dimension r 
  -p, --precision       Internal precision in bits
  -s, --get-stats       Get statistics
  -m, --embedding-save  Save embedding to file
  -v, --verbose         Prints out row-by-row stats
  -t, --scale SCALE     Use a particular scaling factor 
      (otherwise, auto-generate from epsilon) 
  -a, --auto-tau-float  Calculate scale assuming 64-bit final
                        embedding  
  -c, --use-codes       Use coding-theoretic child placement   
  -z, --stats-sample    Number of rows to sample when computing
                        statistics   
  -y, --save-distances  Save the distance matrix
  -q, --procs PROCS     Number of processes to use for stats computation
  -f, --forest          Forest embedding; use all the edgelists in the
                        given folder (i.e., separate components)
  -w, --visualize       Visualize the embedding (only for 2
                        dimensions; requires matplotlib)
  -h, --help            Show this help message and exit

Example on small attached tree. Epsilon = 1.0, custom scaling factor 2.0,
    output saved in savefile:

julia comb.jl -d ../combinatorial/phylotree.edges -e 1.0 -p 256 -s -r 100 -t 2.0 -c -z 100 -m savefile

Output:

Combinatorial Embedding. Info:
Data set = ../combinatorial/phylotree.edges
Dimensions = 100
Epsilon  = 1.0
Precision = 256
Save embedding to savefile

Graph information
Number of vertices = 344
Number of edges = 343
Max degree = 16

Performing the embedding
Using coding theoretic child placement
Placing children of node 100
Placing children of node 200
Placing children of node 300
elapsed time: 13.989601903 seconds
Scaling factor tau = 2.0

Computing quality statistics
Using 100 sample rows for statistics
Final MAP = 1.0
Final d_avg = 0.23174154829754887, d_wc = 1.4062284394628504
