# November 2016

HierDAG_library files:

- Functions implementing Hierarchical methods for DAG:
	- htd.R: 				implementation of the basic HTD algorithm;
	- tpr.R: 				implementation of the TPR algorithms;
	- Do.HTD.R: 			High level functions to compute the HTD algorithm;
	- Do.TPR.R:				High level functions to compute all the variants of TPR algorithm;

- Utility functions:
	- graph.utils.R: 		utility functions for the processing and the analysis of graphs;
	- IOgraph.R:			IO functions for graphs and to construct graphs in different formats;
	- flat.score.norm.R:	function to normalize flat scores of classes according to the Maximum score of each class;	
	- Do.full.annotations.table.R: 	High level functions to compute the full annotation table;
	- Do.FLAT.best.F.score:			High level functions to select the best F-score by choosing an appropriate threshold for the flat scores matrix;
	- Do.FLAT.scores.normalization: High level functions to normalize flat scores matrix w.r.t. MaxNorm or Qnorm;
	- F.hier.R:		functions to compute precision, recall, F-measure, specificity and accuracy for multiclass multilabel classification task (Kiritchenko-like   multi-label F-scores).

The hierDAG library is publicly avaialble under the GNU General Public License, version 3 (GPL-3.0).