Infotheory
1.0
|
Public Member Functions | |
InfoTools (int dims, int nreps=0) | |
void | displayConfig () |
Display the config for analyses such as number of bins, dimensionality etc. | |
void | displaySnapshot () |
Display current status such as number of points added, number of non-empty bins etc. | |
void | setEqualIntervalBinning (TVector< int > &nbs, TVector< double > &mins, TVector< double > &maxs) |
Set bounds on probs depending on max(ids) More... | |
void | setBinBoundaries (TVector< double > boundaries, int dimIndex) |
Set left-boundaries for the binning of specified dimension. More... | |
void | setBinBoundaries (TVector< TVector< double > > boundaries) |
Set left-boundaries for the binning of all dimensions. More... | |
void | addDataPoint (TVector< double > &dp) |
Identify bin for given datapoint and add to count of points in that bin. More... | |
void | addData (TVector< TVector< double > > &data) |
Add list of points at a time. | |
double | entropy (TVector< int > &vIDs) |
Returns entropy of var along dims with varIDs==0. More... | |
double | mutualInfo (TVector< int > &vIDs) |
Return mutual information between vars along dims varIDs==0 and varIDs==1. More... | |
double | redundantInfo (TVector< int > &vIDs) |
Returns redundant information about varIDs==0, in varIDs==1 and varIDs==2 (and varIDs==3 in the case of three sources). More... | |
double | uniqueInfo (TVector< int > &vIDs) |
Returns amount of unique information about varIDs==0, from varIDs==1 and not from varIDs==2 (or varIDs==3 in the case of three sources). More... | |
double | synergy (TVector< int > &vIDs) |
Returns amount of synergistic information about varIDs==0, in varIDs==1 and varIDs==2 (and varIDs==3 in the case of three sources). More... | |
void | clearAllData () |
Clear all data and start over - create another object instead. | |
|
inline |
Constructor
ARGS
dims - dimensionality of data (all variables combined)
nreps - number of reps on each side of bin boundary to compute average shifted histogram
|
inline |
Identify bin for given datapoint and add to count of points in that bin.
dataPoint - TVector with length=dims contains the datapoint to be added
|
inline |
Returns entropy of var along dims with varIDs==0.
Example: if dims = 4, 4D datapoints will be added, but if only the first 2 dimensions make up the variable of interest, then set varIDs[1] = 1; varIDs[2] = 1; varIDs[3] = -1; varIDs[4] = -1
The dims with varID==-1 will be ignored
|
inline |
Return mutual information between vars along dims varIDs==0 and varIDs==1.
Set other varIDs of dims to be ignored to -1
|
inline |
Returns redundant information about varIDs==0, in varIDs==1 and varIDs==2 (and varIDs==3 in the case of three sources).
Set other varIDs of dims to be ignored to -1
|
inline |
Set left-boundaries for the binning of specified dimension.
boundaries - TVector that has bin-margins for specified dimension. Length of list = number_of_bins-1, left most bin is (-inf,list[0]) and right most bin is (list[-1],inf) dim_index - int with dimension of choice - 0-indexing
|
inline |
Set left-boundaries for the binning of all dimensions.
boundaries - TVector that has TVectors for left-margin of all dims. Length of each TVector = number_of_bins-1, left most bin is (-inf,list[0]) and right most bin is (list[-1],inf)
|
inline |
Set bounds on probs depending on max(ids)
nbs - TVector with length=dims and has number of bins for each dimension
mins - TVector with length=dims that has min on each dim
maxs - TVector with length=dims that has max on each dim
|
inline |
Returns amount of synergistic information about varIDs==0, in varIDs==1 and varIDs==2 (and varIDs==3 in the case of three sources).
Set other varIDs of dims to be ignored to -1
|
inline |
Returns amount of unique information about varIDs==0, from varIDs==1 and not from varIDs==2 (or varIDs==3 in the case of three sources).
Set other varIDs of dims to be ignored to -1