mlp_set1.log

init_randem_seed:     27863875
targ_offset:          5
init_randem_weight_max:           0.100000
init_randem_weight_min:           -0.100000
init_randem_bias_max:             0.100000
init_randem_bias_min:             -0.100000
learnrate:                        0.768000
layersizes:                       429,2048,2048,2048,2048,2048,2048,8991,
nmod:                             1
remainder:                        0
discard_prob:                     0.000000
discardLabs:                      
Please check...
Get pfile info over: Training data has 390729 frames, 1001 sentences.
Get chunk info over: Training sentences have 4 chunks, 380689 samples.
Get cv chunk info over: CV sentences have 1 chunks, 73 samples.
start training:
--chunk(0) : containing samples 72704
--chunk(1) : containing samples 102400
--chunk(2) : containing samples 102400
--chunk(3) : containing samples 102400
training over
total time cost: 379378.250000ms