manish7294

bin/mlpack_lmnn -i covertype.txt -p 50 -b 65536 -P -k 3 -O amsgrad -v | grep -v 'node combinations\|base cases\|DEBUG'
[WARN ] Should pass '--output_file (-o)'; no output will be saved!
[INFO ] Loading 'covertype.txt' as CSV data.  Size is 55 x 581012.
[INFO ] Using last column of input dataset as labels.
[INFO ] Initial learning point have invalid dimensionality.  Identity matrix will be used as initial learning point for optimization.
Iteration 0 : Out of 65536, Impostors will be recalculated for 65536 data points.
Iteration 1 : Out of 65536, Impostors will be recalculated for 65536 data points.
Iteration 2 : Out of 65536, Impostors will be recalculated for 65536 data points.
Iteration 3 : Out of 65536, Impostors will be recalculated for 65536 data points.
Iteration 4 : Out of 65536, Impostors will be recalculated for 65536 data points.

manish7294

$ bin/mlpack_boostmetric -i iris.csv -l iris_labels.txt -k 5 
[DEBUG] Compiled with debugging symbols.
[WARN ] Should pass '--output_file (-o)'; no output will be saved!
Iteration : 0, Out of 3750 instances, u is less than 1e-05 for 0 instances.
Iteration : 1, Out of 3750 instances, u is less than 1e-05 for 2019 instances.
Iteration : 2, Out of 3750 instances, u is less than 1e-05 for 2339 instances.
Iteration : 3, Out of 3750 instances, u is less than 1e-05 for 2723 instances.
Iteration : 4, Out of 3750 instances, u is less than 1e-05 for 2927 instances.
Iteration : 5, Out of 3750 instances, u is less than 1e-05 for 2921 instances.
Iteration : 6, Out of 3750 instances, u is less than 1e-05 for 2947 instances.

manish7294

iris	0.14874	97.3333
satellite	2.802369	94.0793
ecoli	0.016140	93.75
vehicle	0.600228	78.8416
balance	0.074875	93.44
letter	31.89887  97.0

manish7294

Initial -> Initial Accuracy 
Final -> final accuracy after boostmetric learning
Timings -> boostmetric timings
Final_LMNN -> final accuracy after lmnn learning
Timings_LMNN -> lmnn timings

Iris:
k   Initial  Final    Timings   Final_LMNN   Timings_LMNN   
3   96.000   96.000  0.063613   96.000        0.347348     
5   96.667   96.667  0.116031   96.667        0.501501