Input data size [72,1041500] = 72 channels, 1041500 frames
Finding 72 ICA components using extended ICA.
Kurtosis will be calculated initially every 1 blocks using 6000 data points.
Decomposing 200 frames per ICA weight ((5184)^2 = 1041500 weights, Initial learning rate will be 0.001, block size 70.
Learning rate will be multiplied by 0.98 whenever angledelta >= 60 deg.
More than 32 channels: default stopping weight change 1E-7
Training will end when wchange < 1e-07 or after 512 steps.
Online bias adjustment will be used.
Removing mean of each channel ...
Final training data range: -4.73468 to 7.71204
Computing the sphering matrix...
Starting weights are the identity matrix ...
Sphering the data ...
Beginning ICA training ... first training step may be slow ...
Lowering learning rate to 0.0009 and starting again.
Lowering learning rate to 0.00081 and starting again.
Lowering learning rate to 0.000729 and starting again.
Lowering learning rate to 0.0006561 and starting again.
Lowering learning rate to 0.00059049 and starting again.
Lowering learning rate to 0.000531441 and starting again.
Lowering learning rate to 0.000478297 and starting again.
Lowering learning rate to 0.000430467 and starting again.
Lowering learning rate to 0.00038742 and starting again.
Lowering learning rate to 0.000348678 and starting again.
Lowering learning rate to 0.000313811 and starting again.
Lowering learning rate to 0.00028243 and starting again.
Lowering learning rate to 0.000254187 and starting again.
Lowering learning rate to 0.000228768 and starting again.
Lowering learning rate to 0.000205891 and starting again.
Lowering learning rate to 0.000185302 and starting again.
Lowering learning rate to 0.000166772 and starting again.
Lowering learning rate to 0.000150095 and starting again.
Lowering learning rate to 0.000135085 and starting again.
Lowering learning rate to 0.000121577 and starting again.
Lowering learning rate to 0.000109419 and starting again.
Lowering learning rate to 9.84771e-05 and starting again.
Lowering learning rate to 8.86294e-05 and starting again.
Lowering learning rate to 7.97664e-05 and starting again.
Lowering learning rate to 7.17898e-05 and starting again.
Lowering learning rate to 6.46108e-05 and starting again.
Lowering learning rate to 5.81497e-05 and starting again.
Lowering learning rate to 5.23348e-05 and starting again.
Lowering learning rate to 4.71013e-05 and starting again.
Lowering learning rate to 4.23912e-05 and starting again.
Lowering learning rate to 3.8152e-05 and starting again.
Lowering learning rate to 3.43368e-05 and starting again.
Lowering learning rate to 3.09032e-05 and starting again.
Lowering learning rate to 2.78128e-05 and starting again.
Lowering learning rate to 2.50316e-05 and starting again.
Lowering learning rate to 2.25284e-05 and starting again.
Lowering learning rate to 2.02756e-05 and starting again.
Lowering learning rate to 1.8248e-05 and starting again.
step 1 - lrate 0.000018, wchange 29419789262971268.00000000, angledelta  0.0 deg
Lowering learning rate to 1.31386e-05 and starting again.
step 1 - lrate 0.000013, wchange 758033595541.61657715, angledelta  0.0 deg
Lowering learning rate to 9.45977e-06 and starting again.
step 1 - lrate 0.000009, wchange 358453028.92838448, angledelta  0.0 deg
Lowering learning rate to 8.51379e-06 and starting again.
step 1 - lrate 0.000009, wchange 50022152.80841193, angledelta  0.0 deg
step 2 - lrate 0.000009, wchange 2489408467994166.00000000, angledelta  0.0 deg
Lowering learning rate to 6.12993e-06 and starting again.
step 1 - lrate 0.000006, wchange 348747.23407844, angledelta  0.0 deg
step 2 - lrate 0.000006, wchange 122018407893.82626343, angledelta  0.0 deg
step 3 - lrate 0.000005, wchange 3271921000434687.50000000, angledelta  0.8 deg
Lowering learning rate to 3.53084e-06 and starting again.
step 1 - lrate 0.000004, wchange 1509.99964731, angledelta  0.0 deg
step 2 - lrate 0.000004, wchange 2317108.47545171, angledelta  0.0 deg
step 3 - lrate 0.000004, wchange 3618683340.00620461, angledelta  7.6 deg
step 4 - lrate 0.000003, wchange 1226460291125.74438477, angledelta  7.6 deg
step 5 - lrate 0.000002, wchange 123833432386934.17187500, angledelta  7.6 deg
step 6 - lrate 0.000002, wchange 4617045200568676.00000000, angledelta  7.6 deg
Lowering learning rate to 1.30161e-06 and starting again.
step 1 - lrate 0.000001, wchange 33.16609731, angledelta  0.0 deg
step 2 - lrate 0.000001, wchange 124.69635422, angledelta  0.0 deg
step 3 - lrate 0.000001, wchange 1875.73275392, angledelta 60.0 deg
step 4 - lrate 0.000001, wchange 28219.70168450, angledelta 60.0 deg
step 5 - lrate 0.000001, wchange 394508.76421405, angledelta  0.0 deg
step 6 - lrate 0.000001, wchange 5622016.48338219, angledelta  0.1 deg
step 7 - lrate 0.000001, wchange 80117539.36308955, angledelta  0.1 deg
step 8 - lrate 0.000001, wchange 1141729248.85157108, angledelta  0.2 deg
step 9 - lrate 0.000001, wchange 7581446924.78474331, angledelta  0.2 deg
step 10 - lrate 0.000001, wchange 31784665546.27150345, angledelta  0.3 deg
step 11 - lrate 0.000001, wchange 91981006683.52648926, angledelta  0.3 deg
step 12 - lrate 0.000001, wchange 197511044738.29031372, angledelta  0.3 deg
step 13 - lrate 0.000000, wchange 333650141296.92010498, angledelta  0.4 deg
step 14 - lrate 0.000000, wchange 464832015197.54797363, angledelta  0.4 deg
step 15 - lrate 0.000000, wchange 554784523942.69726562, angledelta  0.4 deg
step 16 - lrate 0.000000, wchange 584883942182.62573242, angledelta  0.4 deg
step 17 - lrate 0.000000, wchange 558235745932.87536621, angledelta  0.4 deg
step 18 - lrate 0.000000, wchange 491982899543.72949219, angledelta  0.4 deg
step 19 - lrate 0.000000, wchange 406773893514.73242188, angledelta  0.5 deg
step 20 - lrate 0.000000, wchange 319558151692.32836914, angledelta  0.5 deg
step 21 - lrate 0.000000, wchange 240972407039.08496094, angledelta  0.5 deg
step 22 - lrate 0.000000, wchange 175853948493.63534546, angledelta  0.5 deg