Nameexact_genome_1499015103_38_9792_2
Workunit1303444
Created6 Jul 2017, 7:23:05 UTC
Sent6 Jul 2017, 8:30:48 UTC
Report deadline30 Aug 2017, 3:49:42 UTC
Received10 Jul 2017, 7:13:37 UTC
Server stateOver
OutcomeSuccess
Client stateDone
Exit status0 (0x0)
Computer ID54172
Run time1 days 2 hours 31 min 37 sec
CPU time1 days 2 hours 26 min 47 sec
Validate stateWorkunit error - check skipped
Credit0.00
Device peak FLOPS3.30 GFLOPS
Application versionEXACT MNIST Batch Norm CNN Trainer v0.30
Peak working set size162.00 MB
Peak swap size163.00 MB
Peak disk usage2.61 MB

Stderr output

<core_client_version>7.6.31</core_client_version>
<![CDATA[
<stderr_txt>
arguments:
	'../../projects/csgrid.org_csg/exact_client_0.30_x86_64-pc-linux-gnu'
	'--training_file'
	'training_samples.bin'
	'--generalizability_file'
	'generalizability_samples.bin'
	'--testing_file'
	'testing_samples.bin'
	'--genome_file'
	'input_genome.txt'
	'--output_file'
	'output_genome.txt'
	'--checkpoint_file'
	'checkpoint.txt'
converting arguments to vector
boincified training filename: '../../projects/csgrid.org_csg/mnist_training_new.bin'
boincified generalizability filename: '../../projects/csgrid.org_csg/mnist_split_generalizability2.bin'
boincified testing filename: '../../projects/csgrid.org_csg/mnist_split_test2.bin'
boincified genome filename: '../../projects/csgrid.org_csg/exact_genome_1499015103_38_9792.txt'
boincified output filename: '../../projects/csgrid.org_csg/exact_genome_1499015103_38_9792_2_r167764399_0'
boincified checkpoint filename: 'checkpoint.txt'
parsed arguments, loading images
number_classes: 10
channels: 1
cols: 28
rows: 28
reading image set with 5923 images.
reading image set with 6742 images.
reading image set with 5958 images.
reading image set with 6131 images.
reading image set with 5842 images.
reading image set with 5421 images.
reading image set with 5918 images.
reading image set with 6265 images.
reading image set with 5851 images.
reading image set with 5949 images.
image_size: 1x28x28 = 784
read 60000 images.
    class    0: 5923
    class    1: 6742
    class    2: 5958
    class    3: 6131
    class    4: 5842
    class    5: 5421
    class    6: 5918
    class    7: 6265
    class    8: 5851
    class    9: 5949
calculating averages and standard deviations for images
average pixel value for channel 0: 0.13066
pixel variance for channel 0: 0.0949303
pixel standard deviation for channel 0: 0.308108
number_classes: 10
channels: 1
cols: 28
rows: 28
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
image_size: 1x28x28 = 784
read 5000 images.
    class    0: 500
    class    1: 500
    class    2: 500
    class    3: 500
    class    4: 500
    class    5: 500
    class    6: 500
    class    7: 500
    class    8: 500
    class    9: 500
number_classes: 10
channels: 1
cols: 28
rows: 28
reading image set with 480 images.
reading image set with 635 images.
reading image set with 532 images.
reading image set with 510 images.
reading image set with 482 images.
reading image set with 392 images.
reading image set with 458 images.
reading image set with 528 images.
reading image set with 474 images.
reading image set with 509 images.
image_size: 1x28x28 = 784
read 5000 images.
    class    0: 480
    class    1: 635
    class    2: 532
    class    3: 510
    class    4: 482
    class    5: 392
    class    6: 458
    class    7: 528
    class    8: 474
    class    9: 509
loaded images
starting from input file: '../../projects/csgrid.org_csg/exact_genome_1499015103_38_9792.txt'
read CNN_Genome file with version string: 'v0.25'
read exact_id: -1
read genome_id: -1
read initial_mu: 0.5
read mu: 0.5
read mu_delta: 0.95
read initial_learning_rate: 0.0025
read learning_rate: 0.0025
read learning_rate_delta: 0.95
read initial_weight_decay: 0.0005
read weight_decay: 0.0005
read weight_decay_delta: 0.95
read batch_size: 50
read epsilon: 1e-07
read alpha: 0.1
read input_dropout_probability: 0
read hidden_dropout_probability: 0
read velocity_reset: 1000
read epoch: 0
read max_epochs: 100
read reset_weights: 0
read number_training_images: 60000
read best_predictions: 0
read best_error: 1e+07
read best_predictions_epoch: 0
read best_error_epoch: 0
read number_generalizability_images: 5000
read generalizability_predictions: 0
read generalizability_error: 1e+07
read number_test_images: 5000
read test_predictions: 0
read test_error: 1e+07
read generated_by_disable_edge: 1
read generated_by_enable_edge: 1
read generated_by_split_edge: 0
read generated_by_add_edge: 0
read generated_by_change_size: 1
read generated_by_change_size_x: 0
read generated_by_change_size_y: 0
read generated_by_crossover: 0
read generated_by_reset_weights: 0
read generated_by_add_node: 0
read generation_id: 9792
read normal distribution: '0 -1.39890265464783 -2.6580069065094'
generator_str: '1445031439'
read generator: 1445031439
reading 115 nodes.
reading 364 edges.
number input nodes: 1
number softmax nodes: 10
order_size: 0
parsed input file
starting backpropagation!
generator min: 1, generator max: 2147483646
pre shuffle 1: 750831350
post shuffle 1: 568624575
[          , genome  9792] predictions:   58311/  60000 (97.19%), best:   58311/60000 (97.18%), error:      5317.91455, best error:      5317.91455 on epoch:     0, epoch:    0/100, mu: 0.5000000000, learning_rate: 0.0024999999, weight_decay: 0.0005000000
[          , genome  9792] predictions:   59111/  60000 (98.52%), best:   59111/60000 (98.52%), error:      2959.69434, best error:      2959.69434 on epoch:     1, epoch:    1/100, mu: 0.5245000124, learning_rate: 0.0023749999, weight_decay: 0.0004750000
[          , genome  9792] predictions:   58830/  60000 (98.05%), best:   59111/60000 (98.52%), error:      3966.03442, best error:      2959.69434 on epoch:     1, epoch:    2/100, mu: 0.5477750301, learning_rate: 0.0022562500, weight_decay: 0.0004512500
[          , genome  9792] predictions:   59043/  60000 (98.41%), best:   59111/60000 (98.52%), error:      3236.96533, best error:      2959.69434 on epoch:     1, epoch:    3/100, mu: 0.5698862672, learning_rate: 0.0021434375, weight_decay: 0.0004286875
[          , genome  9792] predictions:   59038/  60000 (98.40%), best:   59111/60000 (98.52%), error:      3128.75293, best error:      2959.69434 on epoch:     1, epoch:    4/100, mu: 0.5908919573, learning_rate: 0.0020362656, weight_decay: 0.0004072531
[          , genome  9792] predictions:   59114/  60000 (98.52%), best:   59114/60000 (98.52%), error:      2955.41309, best error:      2955.41309 on epoch:     5, epoch:    5/100, mu: 0.6108473539, learning_rate: 0.0019344522, weight_decay: 0.0003868904
[          , genome  9792] predictions:   59077/  60000 (98.46%), best:   59114/60000 (98.52%), error:      3142.47729, best error:      2955.41309 on epoch:     5, epoch:    6/100, mu: 0.6298049688, learning_rate: 0.0018377296, weight_decay: 0.0003675459
[          , genome  9792] predictions:   59312/  60000 (98.85%), best:   59312/60000 (98.85%), error:      2464.84009, best error:      2464.84009 on epoch:     7, epoch:    7/100, mu: 0.6478147507, learning_rate: 0.0017458431, weight_decay: 0.0003491686
[          , genome  9792] predictions:   59089/  60000 (98.48%), best:   59312/60000 (98.85%), error:      3116.18555, best error:      2464.84009 on epoch:     7, epoch:    8/100, mu: 0.6649240255, learning_rate: 0.0016585509, weight_decay: 0.0003317102
[          , genome  9792] predictions:   59027/  60000 (98.38%), best:   59312/60000 (98.85%), error:      3226.94263, best error:      2464.84009 on epoch:     7, epoch:    9/100, mu: 0.6811778545, learning_rate: 0.0015756234, weight_decay: 0.0003151247
[          , genome  9792] predictions:   59161/  60000 (98.60%), best:   59312/60000 (98.85%), error:      2847.90747, best error:      2464.84009 on epoch:     7, epoch:   10/100, mu: 0.6966189742, learning_rate: 0.0014968422, weight_decay: 0.0002993684
[          , genome  9792] predictions:   59297/  60000 (98.83%), best:   59297/60000 (98.83%), error:      2426.58496, best error:      2426.58496 on epoch:    11, epoch:   11/100, mu: 0.7112880349, learning_rate: 0.0014220001, weight_decay: 0.0002844000
[          , genome  9792] predictions:   58694/  60000 (97.82%), best:   59297/60000 (98.83%), error:      4069.11865, best error:      2426.58496 on epoch:    11, epoch:   12/100, mu: 0.7252236605, learning_rate: 0.0013509000, weight_decay: 0.0002701800
[          , genome  9792] predictions:   59163/  60000 (98.61%), best:   59297/60000 (98.83%), error:      2814.63672, best error:      2426.58496 on epoch:    11, epoch:   13/100, mu: 0.7384625077, learning_rate: 0.0012833551, weight_decay: 0.0002566710
[          , genome  9792] predictions:   58838/  60000 (98.06%), best:   59297/60000 (98.83%), error:      3832.75146, best error:      2426.58496 on epoch:    11, epoch:   14/100, mu: 0.7510393858, learning_rate: 0.0012191873, weight_decay: 0.0002438374
[          , genome  9792] predictions:   59303/  60000 (98.84%), best:   59303/60000 (98.84%), error:      2397.59131, best error:      2397.59131 on epoch:    15, epoch:   15/100, mu: 0.7629874349, learning_rate: 0.0011582279, weight_decay: 0.0002316456
[          , genome  9792] predictions:   59431/  60000 (99.05%), best:   59431/60000 (99.05%), error:      2083.21924, best error:      2083.21924 on epoch:    16, epoch:   16/100, mu: 0.7743380666, learning_rate: 0.0011003165, weight_decay: 0.0002200633
[          , genome  9792] predictions:   59274/  60000 (98.79%), best:   59431/60000 (99.05%), error:      2476.01733, best error:      2083.21924 on epoch:    16, epoch:   17/100, mu: 0.7851211429, learning_rate: 0.0010453007, weight_decay: 0.0002090601
[          , genome  9792] predictions:   59266/  60000 (98.78%), best:   59431/60000 (99.05%), error:      2546.82104, best error:      2083.21924 on epoch:    16, epoch:   18/100, mu: 0.7953650951, learning_rate: 0.0009930357, weight_decay: 0.0001986071
[          , genome  9792] predictions:   59396/  60000 (98.99%), best:   59431/60000 (99.05%), error:      2154.28198, best error:      2083.21924 on epoch:    16, epoch:   19/100, mu: 0.8050968647, learning_rate: 0.0009433840, weight_decay: 0.0001886768
[          , genome  9792] predictions:   59458/  60000 (99.10%), best:   59458/60000 (99.10%), error:      2066.03076, best error:      2066.03076 on epoch:    20, epoch:   20/100, mu: 0.8143420219, learning_rate: 0.0008962147, weight_decay: 0.0001792429
[          , genome  9792] predictions:   59476/  60000 (99.13%), best:   59476/60000 (99.13%), error:      1933.19763, best error:      1933.19763 on epoch:    21, epoch:   21/100, mu: 0.8231249452, learning_rate: 0.0008514040, weight_decay: 0.0001702808
[          , genome  9792] predictions:   59475/  60000 (99.12%), best:   59475/60000 (99.12%), error:      1849.81775, best error:      1849.81775 on epoch:    22, epoch:   22/100, mu: 0.8314687014, learning_rate: 0.0008088337, weight_decay: 0.0001617667
[          , genome  9792] predictions:   59498/  60000 (99.16%), best:   59498/60000 (99.16%), error:      1829.86670, best error:      1829.86670 on epoch:    23, epoch:   23/100, mu: 0.8393952847, learning_rate: 0.0007683921, weight_decay: 0.0001536784
[          , genome  9792] predictions:   59514/  60000 (99.19%), best:   59498/60000 (99.16%), error:      1870.20215, best error:      1829.86670 on epoch:    23, epoch:   24/100, mu: 0.8469254971, learning_rate: 0.0007299724, weight_decay: 0.0001459945
[          , genome  9792] predictions:   59368/  60000 (98.95%), best:   59498/60000 (99.16%), error:      2248.48120, best error:      1829.86670 on epoch:    23, epoch:   25/100, mu: 0.8540792465, learning_rate: 0.0006934738, weight_decay: 0.0001386948
[          , genome  9792] predictions:   59435/  60000 (99.06%), best:   59498/60000 (99.16%), error:      2005.62012, best error:      1829.86670 on epoch:    23, epoch:   26/100, mu: 0.8608753085, learning_rate: 0.0006588001, weight_decay: 0.0001317600
[          , genome  9792] predictions:   59395/  60000 (98.99%), best:   59498/60000 (99.16%), error:      2077.57324, best error:      1829.86670 on epoch:    23, epoch:   27/100, mu: 0.8673315644, learning_rate: 0.0006258601, weight_decay: 0.0001251720
[          , genome  9792] predictions:   59595/  60000 (99.33%), best:   59595/60000 (99.32%), error:      1558.83777, best error:      1558.83777 on epoch:    28, epoch:   28/100, mu: 0.8734650016, learning_rate: 0.0005945670, weight_decay: 0.0001189134
[          , genome  9792] predictions:   59481/  60000 (99.14%), best:   59595/60000 (99.32%), error:      1893.23022, best error:      1558.83777 on epoch:    28, epoch:   29/100, mu: 0.8792917728, learning_rate: 0.0005648387, weight_decay: 0.0001129677
[          , genome  9792] predictions:   59515/  60000 (99.19%), best:   59595/60000 (99.32%), error:      1713.75134, best error:      1558.83777 on epoch:    28, epoch:   30/100, mu: 0.8848271966, learning_rate: 0.0005365968, weight_decay: 0.0001073193
[          , genome  9792] predictions:   59349/  60000 (98.92%), best:   59595/60000 (99.32%), error:      2219.65747, best error:      1558.83777 on epoch:    28, epoch:   31/100, mu: 0.8900858164, learning_rate: 0.0005097669, weight_decay: 0.0001019534
[          , genome  9792] predictions:   59635/  60000 (99.39%), best:   59635/60000 (99.39%), error:      1442.84094, best error:      1442.84094 on epoch:    32, epoch:   32/100, mu: 0.8950815201, learning_rate: 0.0004842786, weight_decay: 0.0000968557
[          , genome  9792] predictions:   59675/  60000 (99.46%), best:   59675/60000 (99.46%), error:      1401.61938, best error:      1401.61938 on epoch:    33, epoch:   33/100, mu: 0.8998274207, learning_rate: 0.0004600646, weight_decay: 0.0000920129
[          , genome  9792] predictions:   59537/  60000 (99.23%), best:   59675/60000 (99.46%), error:      1656.17725, best error:      1401.61938 on epoch:    33, epoch:   34/100, mu: 0.9043360353, learning_rate: 0.0004370614, weight_decay: 0.0000874123
[          , genome  9792] predictions:   59698/  60000 (99.50%), best:   59698/60000 (99.50%), error:      1303.98242, best error:      1303.98242 on epoch:    35, epoch:   35/100, mu: 0.9086192250, learning_rate: 0.0004152083, weight_decay: 0.0000830417
[          , genome  9792] predictions:   59615/  60000 (99.36%), best:   59698/60000 (99.50%), error:      1498.23694, best error:      1303.98242 on epoch:    35, epoch:   36/100, mu: 0.9126882553, learning_rate: 0.0003944479, weight_decay: 0.0000788896
[          , genome  9792] predictions:   59571/  60000 (99.28%), best:   59698/60000 (99.50%), error:      1643.69861, best error:      1303.98242 on epoch:    35, epoch:   37/100, mu: 0.9165538549, learning_rate: 0.0003747255, weight_decay: 0.0000749451
[          , genome  9792] predictions:   59716/  60000 (99.53%), best:   59716/60000 (99.53%), error:      1211.66101, best error:      1211.66101 on epoch:    38, epoch:   38/100, mu: 0.9202261567, learning_rate: 0.0003559892, weight_decay: 0.0000711978
[          , genome  9792] predictions:   59743/  60000 (99.57%), best:   59743/60000 (99.57%), error:      1152.62769, best error:      1152.62769 on epoch:    39, epoch:   39/100, mu: 0.9237148762, learning_rate: 0.0003381897, weight_decay: 0.0000676379
[          , genome  9792] predictions:   59569/  60000 (99.28%), best:   59743/60000 (99.57%), error:      1625.55359, best error:      1152.62769 on epoch:    39, epoch:   40/100, mu: 0.9270291328, learning_rate: 0.0003212802, weight_decay: 0.0000642560
[          , genome  9792] predictions:   59635/  60000 (99.39%), best:   59743/60000 (99.57%), error:      1500.76501, best error:      1152.62769 on epoch:    39, epoch:   41/100, mu: 0.9301776886, learning_rate: 0.0003052162, weight_decay: 0.0000610432
[          , genome  9792] predictions:   59668/  60000 (99.45%), best:   59743/60000 (99.57%), error:      1340.98340, best error:      1152.62769 on epoch:    39, epoch:   42/100, mu: 0.9331688285, learning_rate: 0.0002899554, weight_decay: 0.0000579911
[          , genome  9792] predictions:   59731/  60000 (99.55%), best:   59743/60000 (99.57%), error:      1183.53577, best error:      1152.62769 on epoch:    39, epoch:   43/100, mu: 0.9360103607, learning_rate: 0.0002754576, weight_decay: 0.0000550915
[          , genome  9792] predictions:   59760/  60000 (99.60%), best:   59760/60000 (99.60%), error:      1093.46558, best error:      1093.46558 on epoch:    44, epoch:   44/100, mu: 0.9387098551, learning_rate: 0.0002616847, weight_decay: 0.0000523369
[          , genome  9792] predictions:   59798/  60000 (99.66%), best:   59798/60000 (99.66%), error:       966.87903, best error:       966.87903 on epoch:    45, epoch:   45/100, mu: 0.9412743449, learning_rate: 0.0002486005, weight_decay: 0.0000497201
[          , genome  9792] predictions:   59687/  60000 (99.48%), best:   59798/60000 (99.66%), error:      1302.42932, best error:       966.87903 on epoch:    45, epoch:   46/100, mu: 0.9437106252, learning_rate: 0.0002361705, weight_decay: 0.0000472341
[          , genome  9792] predictions:   59651/  60000 (99.42%), best:   59798/60000 (99.66%), error:      1417.35364, best error:       966.87903 on epoch:    45, epoch:   47/100, mu: 0.9460250735, learning_rate: 0.0002243619, weight_decay: 0.0000448724
[          , genome  9792] predictions:   59754/  60000 (99.59%), best:   59798/60000 (99.66%), error:      1096.32227, best error:       966.87903 on epoch:    45, epoch:   48/100, mu: 0.9482238293, learning_rate: 0.0002131438, weight_decay: 0.0000426288
[          , genome  9792] predictions:   59793/  60000 (99.66%), best:   59798/60000 (99.66%), error:       983.77179, best error:       966.87903 on epoch:    45, epoch:   49/100, mu: 0.9503126144, learning_rate: 0.0002024866, weight_decay: 0.0000404973
[          , genome  9792] predictions:   59820/  60000 (99.70%), best:   59820/60000 (99.70%), error:       923.35150, best error:       923.35150 on epoch:    50, epoch:   50/100, mu: 0.9522969723, learning_rate: 0.0001923623, weight_decay: 0.0000384725
[          , genome  9792] predictions:   59815/  60000 (99.69%), best:   59820/60000 (99.70%), error:       940.90887, best error:       923.35150 on epoch:    50, epoch:   51/100, mu: 0.9541821480, learning_rate: 0.0001827442, weight_decay: 0.0000365488
[          , genome  9792] predictions:   59653/  60000 (99.42%), best:   59820/60000 (99.70%), error:      1351.07422, best error:       923.35150 on epoch:    50, epoch:   52/100, mu: 0.9559730291, learning_rate: 0.0001736070, weight_decay: 0.0000347214
[          , genome  9792] predictions:   59807/  60000 (99.68%), best:   59820/60000 (99.70%), error:       941.65796, best error:       923.35150 on epoch:    50, epoch:   53/100, mu: 0.9576743841, learning_rate: 0.0001649266, weight_decay: 0.0000329853
[          , genome  9792] predictions:   59782/  60000 (99.64%), best:   59820/60000 (99.70%), error:      1028.91003, best error:       923.35150 on epoch:    50, epoch:   54/100, mu: 0.9592906833, learning_rate: 0.0001566803, weight_decay: 0.0000313361
[          , genome  9792] predictions:   59821/  60000 (99.70%), best:   59821/60000 (99.70%), error:       888.06622, best error:       888.06622 on epoch:    55, epoch:   55/100, mu: 0.9608261585, learning_rate: 0.0001488463, weight_decay: 0.0000297693
arguments:
	'../../projects/csgrid.org_csg/exact_client_0.30_x86_64-pc-linux-gnu'
	'--training_file'
	'training_samples.bin'
	'--generalizability_file'
	'generalizability_samples.bin'
	'--testing_file'
	'testing_samples.bin'
	'--genome_file'
	'input_genome.txt'
	'--output_file'
	'output_genome.txt'
	'--checkpoint_file'
	'checkpoint.txt'
converting arguments to vector
boincified training filename: '../../projects/csgrid.org_csg/mnist_training_new.bin'
boincified generalizability filename: '../../projects/csgrid.org_csg/mnist_split_generalizability2.bin'
boincified testing filename: '../../projects/csgrid.org_csg/mnist_split_test2.bin'
boincified genome filename: '../../projects/csgrid.org_csg/exact_genome_1499015103_38_9792.txt'
boincified output filename: '../../projects/csgrid.org_csg/exact_genome_1499015103_38_9792_2_r167764399_0'
boincified checkpoint filename: 'checkpoint.txt'
parsed arguments, loading images
number_classes: 10
channels: 1
cols: 28
rows: 28
reading image set with 5923 images.
reading image set with 6742 images.
reading image set with 5958 images.
reading image set with 6131 images.
reading image set with 5842 images.
reading image set with 5421 images.
reading image set with 5918 images.
reading image set with 6265 images.
reading image set with 5851 images.
reading image set with 5949 images.
image_size: 1x28x28 = 784
read 60000 images.
    class    0: 5923
    class    1: 6742
    class    2: 5958
    class    3: 6131
    class    4: 5842
    class    5: 5421
    class    6: 5918
    class    7: 6265
    class    8: 5851
    class    9: 5949
calculating averages and standard deviations for images
average pixel value for channel 0: 0.13066
pixel variance for channel 0: 0.0949303
pixel standard deviation for channel 0: 0.308108
number_classes: 10
channels: 1
cols: 28
rows: 28
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
image_size: 1x28x28 = 784
read 5000 images.
    class    0: 500
    class    1: 500
    class    2: 500
    class    3: 500
    class    4: 500
    class    5: 500
    class    6: 500
    class    7: 500
    class    8: 500
    class    9: 500
number_classes: 10
channels: 1
cols: 28
rows: 28
reading image set with 480 images.
reading image set with 635 images.
reading image set with 532 images.
reading image set with 510 images.
reading image set with 482 images.
reading image set with 392 images.
reading image set with 458 images.
reading image set with 528 images.
reading image set with 474 images.
reading image set with 509 images.
image_size: 1x28x28 = 784
read 5000 images.
    class    0: 480
    class    1: 635
    class    2: 532
    class    3: 510
    class    4: 482
    class    5: 392
    class    6: 458
    class    7: 528
    class    8: 474
    class    9: 509
loaded images
starting from checkpoint file: 'checkpoint.txt'
read CNN_Genome file with version string: 'v0.25'
read exact_id: -1
read genome_id: -1
read initial_mu: 0.5
read mu: 0.962285
read mu_delta: 0.95
read initial_learning_rate: 0.0025
read learning_rate: 0.000141404
read learning_rate_delta: 0.95
read initial_weight_decay: 0.0005
read weight_decay: 2.82808e-05
read weight_decay_delta: 0.95
read batch_size: 50
read epsilon: 1e-07
read alpha: 0.1
read input_dropout_probability: 0
read hidden_dropout_probability: 0
read velocity_reset: 1000
read epoch: 56
read max_epochs: 100
read reset_weights: 0
read number_training_images: 60000
read best_predictions: 59821
read best_error: 888.066
read best_predictions_epoch: 55
read best_error_epoch: 55
read number_generalizability_images: 5000
read generalizability_predictions: 0
read generalizability_error: 1e+07
read number_test_images: 5000
read test_predictions: 0
read test_error: 1e+07
read generated_by_disable_edge: 1
read generated_by_enable_edge: 1
read generated_by_split_edge: 0
read generated_by_add_edge: 0
read generated_by_change_size: 1
read generated_by_change_size_x: 0
read generated_by_change_size_y: 0
read generated_by_crossover: 0
read generated_by_reset_weights: 0
read generated_by_add_node: 0
read generation_id: 9792
read normal distribution: '0 -1.39890265464783 -2.6580069065094'
generator_str: '1706793310'
read generator: 1706793310
reading 115 nodes.
reading 364 edges.
number input nodes: 1
number softmax nodes: 10
order_size: 60000
parsed input file
starting backpropagation!
arguments:
	'../../projects/csgrid.org_csg/exact_client_0.30_x86_64-pc-linux-gnu'
	'--training_file'
	'training_samples.bin'
	'--generalizability_file'
	'generalizability_samples.bin'
	'--testing_file'
	'testing_samples.bin'
	'--genome_file'
	'input_genome.txt'
	'--output_file'
	'output_genome.txt'
	'--checkpoint_file'
	'checkpoint.txt'
converting arguments to vector
boincified training filename: '../../projects/csgrid.org_csg/mnist_training_new.bin'
boincified generalizability filename: '../../projects/csgrid.org_csg/mnist_split_generalizability2.bin'
boincified testing filename: '../../projects/csgrid.org_csg/mnist_split_test2.bin'
boincified genome filename: '../../projects/csgrid.org_csg/exact_genome_1499015103_38_9792.txt'
boincified output filename: '../../projects/csgrid.org_csg/exact_genome_1499015103_38_9792_2_r167764399_0'
boincified checkpoint filename: 'checkpoint.txt'
parsed arguments, loading images
number_classes: 10
channels: 1
cols: 28
rows: 28
reading image set with 5923 images.
reading image set with 6742 images.
reading image set with 5958 images.
reading image set with 6131 images.
reading image set with 5842 images.
reading image set with 5421 images.
reading image set with 5918 images.
reading image set with 6265 images.
reading image set with 5851 images.
reading image set with 5949 images.
image_size: 1x28x28 = 784
read 60000 images.
    class    0: 5923
    class    1: 6742
    class    2: 5958
    class    3: 6131
    class    4: 5842
    class    5: 5421
    class    6: 5918
    class    7: 6265
    class    8: 5851
    class    9: 5949
calculating averages and standard deviations for images
average pixel value for channel 0: 0.13066
pixel variance for channel 0: 0.0949303
pixel standard deviation for channel 0: 0.308108
number_classes: 10
channels: 1
cols: 28
rows: 28
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
reading image set with 500 images.
image_size: 1x28x28 = 784
read 5000 images.
    class    0: 500
    class    1: 500
    class    2: 500
    class    3: 500
    class    4: 500
    class    5: 500
    class    6: 500
    class    7: 500
    class    8: 500
    class    9: 500
number_classes: 10
channels: 1
cols: 28
rows: 28
reading image set with 480 images.
reading image set with 635 images.
reading image set with 532 images.
reading image set with 510 images.
reading image set with 482 images.
reading image set with 392 images.
reading image set with 458 images.
reading image set with 528 images.
reading image set with 474 images.
reading image set with 509 images.
image_size: 1x28x28 = 784
read 5000 images.
    class    0: 480
    class    1: 635
    class    2: 532
    class    3: 510
    class    4: 482
    class    5: 392
    class    6: 458
    class    7: 528
    class    8: 474
    class    9: 509
loaded images
starting from checkpoint file: 'checkpoint.txt'
read CNN_Genome file with version string: 'v0.25'
read exact_id: -1
read genome_id: -1
read initial_mu: 0.5
read mu: 0.962285
read mu_delta: 0.95
read initial_learning_rate: 0.0025
read learning_rate: 0.000141404
read learning_rate_delta: 0.95
read initial_weight_decay: 0.0005
read weight_decay: 2.82808e-05
read weight_decay_delta: 0.95
read batch_size: 50
read epsilon: 1e-07
read alpha: 0.1
read input_dropout_probability: 0
read hidden_dropout_probability: 0
read velocity_reset: 1000
read epoch: 56
read max_epochs: 100
read reset_weights: 0
read number_training_images: 60000
read best_predictions: 59821
read best_error: 888.066
read best_predictions_epoch: 55
read best_error_epoch: 55
read number_generalizability_images: 5000
read generalizability_predictions: 0
read generalizability_error: 1e+07
read number_test_images: 5000
read test_predictions: 0
read test_error: 1e+07
read generated_by_disable_edge: 1
read generated_by_enable_edge: 1
read generated_by_split_edge: 0
read generated_by_add_edge: 0
read generated_by_change_size: 1
read generated_by_change_size_x: 0
read generated_by_change_size_y: 0
read generated_by_crossover: 0
read generated_by_reset_weights: 0
read generated_by_add_node: 0
read generation_id: 9792
read normal distribution: '0 -1.39890265464783 -2.6580069065094'
generator_str: '1706793310'
read generator: 1706793310
reading 115 nodes.
reading 364 edges.
number input nodes: 1
number softmax nodes: 10
order_size: 60000
parsed input file
starting backpropagation!
[          , genome  9792] predictions:   59709/  60000 (99.52%), best:   59821/60000 (99.70%), error:      1209.69165, best error:       888.06622 on epoch:    55, epoch:   56/100, mu: 0.9622848630, learning_rate: 0.0001414040, weight_decay: 0.0000282808
[          , genome  9792] predictions:   59849/  60000 (99.75%), best:   59849/60000 (99.75%), error:       841.79187, best error:       841.79187 on epoch:    57, epoch:   57/100, mu: 0.9636706114, learning_rate: 0.0001343338, weight_decay: 0.0000268668
[          , genome  9792] predictions:   59851/  60000 (99.75%), best:   59849/60000 (99.75%), error:       903.64423, best error:       841.79187 on epoch:    57, epoch:   58/100, mu: 0.9649870992, learning_rate: 0.0001276171, weight_decay: 0.0000255234
[          , genome  9792] predictions:   59846/  60000 (99.74%), best:   59849/60000 (99.75%), error:       867.62311, best error:       841.79187 on epoch:    57, epoch:   59/100, mu: 0.9662377238, learning_rate: 0.0001212362, weight_decay: 0.0000242472
[          , genome  9792] predictions:   59802/  60000 (99.67%), best:   59849/60000 (99.75%), error:      1013.54608, best error:       841.79187 on epoch:    57, epoch:   60/100, mu: 0.9674258232, learning_rate: 0.0001151744, weight_decay: 0.0000230349
[          , genome  9792] predictions:   59867/  60000 (99.78%), best:   59867/60000 (99.78%), error:       834.59991, best error:       834.59991 on epoch:    61, epoch:   61/100, mu: 0.9685545564, learning_rate: 0.0001094157, weight_decay: 0.0000218831
[          , genome  9792] predictions:   59875/  60000 (99.79%), best:   59875/60000 (99.79%), error:       757.11652, best error:       757.11652 on epoch:    62, epoch:   62/100, mu: 0.9696268439, learning_rate: 0.0001039449, weight_decay: 0.0000207890
[          , genome  9792] predictions:   59877/  60000 (99.80%), best:   59875/60000 (99.79%), error:       790.80280, best error:       757.11652 on epoch:    62, epoch:   63/100, mu: 0.9706454873, learning_rate: 0.0000987477, weight_decay: 0.0000197495
[          , genome  9792] predictions:   59896/  60000 (99.83%), best:   59896/60000 (99.83%), error:       736.44995, best error:       736.44995 on epoch:    64, epoch:   64/100, mu: 0.9716132283, learning_rate: 0.0000938103, weight_decay: 0.0000187621
[          , genome  9792] predictions:   59904/  60000 (99.84%), best:   59904/60000 (99.84%), error:       733.69580, best error:       733.69580 on epoch:    65, epoch:   65/100, mu: 0.9725325704, learning_rate: 0.0000891198, weight_decay: 0.0000178240
[          , genome  9792] predictions:   59887/  60000 (99.81%), best:   59904/60000 (99.84%), error:       747.43884, best error:       733.69580 on epoch:    65, epoch:   66/100, mu: 0.9734059572, learning_rate: 0.0000846638, weight_decay: 0.0000169328
[          , genome  9792] predictions:   59917/  60000 (99.86%), best:   59917/60000 (99.86%), error:       656.95325, best error:       656.95325 on epoch:    67, epoch:   67/100, mu: 0.9742356539, learning_rate: 0.0000804306, weight_decay: 0.0000160861
[          , genome  9792] predictions:   59852/  60000 (99.75%), best:   59917/60000 (99.86%), error:       849.78918, best error:       656.95325 on epoch:    67, epoch:   68/100, mu: 0.9750238657, learning_rate: 0.0000764091, weight_decay: 0.0000152818
[          , genome  9792] predictions:   59924/  60000 (99.87%), best:   59924/60000 (99.87%), error:       645.24219, best error:       645.24219 on epoch:    69, epoch:   69/100, mu: 0.9757726789, learning_rate: 0.0000725886, weight_decay: 0.0000145177
[          , genome  9792] predictions:   59934/  60000 (99.89%), best:   59924/60000 (99.87%), error:       645.42535, best error:       645.24219 on epoch:    69, epoch:   70/100, mu: 0.9764840603, learning_rate: 0.0000689592, weight_decay: 0.0000137918
[          , genome  9792] predictions:   59912/  60000 (99.85%), best:   59924/60000 (99.87%), error:       694.47974, best error:       645.24219 on epoch:    69, epoch:   71/100, mu: 0.9771598577, learning_rate: 0.0000655112, weight_decay: 0.0000131022
[          , genome  9792] predictions:   59897/  60000 (99.83%), best:   59924/60000 (99.87%), error:       711.84149, best error:       645.24219 on epoch:    69, epoch:   72/100, mu: 0.9778018594, learning_rate: 0.0000622357, weight_decay: 0.0000124471
[          , genome  9792] predictions:   59927/  60000 (99.88%), best:   59927/60000 (99.88%), error:       638.08905, best error:       638.08905 on epoch:    73, epoch:   73/100, mu: 0.9784117937, learning_rate: 0.0000591239, weight_decay: 0.0000118248
[          , genome  9792] predictions:   59923/  60000 (99.87%), best:   59923/60000 (99.87%), error:       619.78186, best error:       619.78186 on epoch:    74, epoch:   74/100, mu: 0.9789912105, learning_rate: 0.0000561677, weight_decay: 0.0000112335
[          , genome  9792] predictions:   59913/  60000 (99.86%), best:   59923/60000 (99.87%), error:       652.99255, best error:       619.78186 on epoch:    74, epoch:   75/100, mu: 0.9795416594, learning_rate: 0.0000533593, weight_decay: 0.0000106719
[          , genome  9792] predictions:   59938/  60000 (99.90%), best:   59938/60000 (99.90%), error:       588.47168, best error:       588.47168 on epoch:    76, epoch:   76/100, mu: 0.9800645709, learning_rate: 0.0000506913, weight_decay: 0.0000101383
[          , genome  9792] predictions:   59939/  60000 (99.90%), best:   59939/60000 (99.90%), error:       573.46271, best error:       573.46271 on epoch:    77, epoch:   77/100, mu: 0.9805613160, learning_rate: 0.0000481568, weight_decay: 0.0000096314
[          , genome  9792] predictions:   59912/  60000 (99.85%), best:   59939/60000 (99.90%), error:       654.51794, best error:       573.46271 on epoch:    77, epoch:   78/100, mu: 0.9810332656, learning_rate: 0.0000457489, weight_decay: 0.0000091498
[          , genome  9792] predictions:   59948/  60000 (99.91%), best:   59939/60000 (99.90%), error:       595.16510, best error:       573.46271 on epoch:    77, epoch:   79/100, mu: 0.9814816117, learning_rate: 0.0000434615, weight_decay: 0.0000086923
[          , genome  9792] predictions:   59944/  60000 (99.91%), best:   59944/60000 (99.91%), error:       558.12799, best error:       558.12799 on epoch:    80, epoch:   80/100, mu: 0.9819075465, learning_rate: 0.0000412884, weight_decay: 0.0000082577
[          , genome  9792] predictions:   59939/  60000 (99.90%), best:   59944/60000 (99.91%), error:       606.29065, best error:       558.12799 on epoch:    80, epoch:   81/100, mu: 0.9823121428, learning_rate: 0.0000392240, weight_decay: 0.0000078448
[          , genome  9792] predictions:   59943/  60000 (99.91%), best:   59944/60000 (99.91%), error:       574.87225, best error:       558.12799 on epoch:    80, epoch:   82/100, mu: 0.9826965332, learning_rate: 0.0000372628, weight_decay: 0.0000074526
[          , genome  9792] predictions:   59935/  60000 (99.89%), best:   59944/60000 (99.91%), error:       616.20636, best error:       558.12799 on epoch:    80, epoch:   83/100, mu: 0.9830617309, learning_rate: 0.0000353996, weight_decay: 0.0000070799
[          , genome  9792] predictions:   59949/  60000 (99.92%), best:   59949/60000 (99.92%), error:       551.08002, best error:       551.08002 on epoch:    84, epoch:   84/100, mu: 0.9834086299, learning_rate: 0.0000336297, weight_decay: 0.0000067259
[          , genome  9792] predictions:   59952/  60000 (99.92%), best:   59952/60000 (99.92%), error:       545.19684, best error:       545.19684 on epoch:    85, epoch:   85/100, mu: 0.9837381840, learning_rate: 0.0000319482, weight_decay: 0.0000063896
[          , genome  9792] predictions:   59960/  60000 (99.93%), best:   59960/60000 (99.93%), error:       524.48865, best error:       524.48865 on epoch:    86, epoch:   86/100, mu: 0.9840512872, learning_rate: 0.0000303508, weight_decay: 0.0000060702
[          , genome  9792] predictions:   59951/  60000 (99.92%), best:   59960/60000 (99.93%), error:       552.04443, best error:       524.48865 on epoch:    86, epoch:   87/100, mu: 0.9843487144, learning_rate: 0.0000288332, weight_decay: 0.0000057666
[          , genome  9792] predictions:   59942/  60000 (99.90%), best:   59960/60000 (99.93%), error:       603.90100, best error:       524.48865 on epoch:    86, epoch:   88/100, mu: 0.9846313000, learning_rate: 0.0000273916, weight_decay: 0.0000054783
[          , genome  9792] predictions:   59947/  60000 (99.91%), best:   59960/60000 (99.93%), error:       608.14514, best error:       524.48865 on epoch:    86, epoch:   89/100, mu: 0.9848997593, learning_rate: 0.0000260220, weight_decay: 0.0000052044
[          , genome  9792] predictions:   59954/  60000 (99.92%), best:   59960/60000 (99.93%), error:       537.30444, best error:       524.48865 on epoch:    86, epoch:   90/100, mu: 0.9851547480, learning_rate: 0.0000247209, weight_decay: 0.0000049442
[          , genome  9792] predictions:   59956/  60000 (99.93%), best:   59960/60000 (99.93%), error:       524.89691, best error:       524.48865 on epoch:    86, epoch:   91/100, mu: 0.9853969812, learning_rate: 0.0000234848, weight_decay: 0.0000046970
[          , genome  9792] predictions:   59957/  60000 (99.93%), best:   59960/60000 (99.93%), error:       532.16724, best error:       524.48865 on epoch:    86, epoch:   92/100, mu: 0.9856271148, learning_rate: 0.0000223106, weight_decay: 0.0000044621
[          , genome  9792] predictions:   59956/  60000 (99.93%), best:   59956/60000 (99.93%), error:       522.91345, best error:       522.91345 on epoch:    93, epoch:   93/100, mu: 0.9858457446, learning_rate: 0.0000211951, weight_decay: 0.0000042390
[          , genome  9792] predictions:   59955/  60000 (99.92%), best:   59955/60000 (99.93%), error:       502.77228, best error:       502.77228 on epoch:    94, epoch:   94/100, mu: 0.9860534668, learning_rate: 0.0000201353, weight_decay: 0.0000040271
[          , genome  9792] predictions:   59945/  60000 (99.91%), best:   59955/60000 (99.93%), error:       605.42480, best error:       502.77228 on epoch:    94, epoch:   95/100, mu: 0.9862508178, learning_rate: 0.0000191285, weight_decay: 0.0000038257
[          , genome  9792] predictions:   59953/  60000 (99.92%), best:   59955/60000 (99.93%), error:       538.38208, best error:       502.77228 on epoch:    94, epoch:   96/100, mu: 0.9864382744, learning_rate: 0.0000181721, weight_decay: 0.0000036344
[          , genome  9792] predictions:   59963/  60000 (99.94%), best:   59955/60000 (99.93%), error:       504.11462, best error:       502.77228 on epoch:    94, epoch:   97/100, mu: 0.9866163731, learning_rate: 0.0000172635, weight_decay: 0.0000034527
[          , genome  9792] predictions:   59961/  60000 (99.94%), best:   59961/60000 (99.93%), error:       496.05423, best error:       496.05423 on epoch:    98, epoch:   98/100, mu: 0.9867855310, learning_rate: 0.0000164003, weight_decay: 0.0000032801
[          , genome  9792] predictions:   59961/  60000 (99.94%), best:   59961/60000 (99.93%), error:       508.16360, best error:       496.05423 on epoch:    98, epoch:   99/100, mu: 0.9869462252, learning_rate: 0.0000155803, weight_decay: 0.0000031161
[          , genome  9792] predictions:   59960/  60000 (99.93%), best:   59960/60000 (99.93%), error:       493.22141, best error:       493.22141 on epoch:   100, epoch:  100/100, mu: 0.9870989323, learning_rate: 0.0000148013, weight_decay: 0.0000029603
evaluating best weights on test data.
evaluting generalizability set with running mean/variance:
[          , genome  9792] predictions:    4949/   5000 (98.98%), best:   59960/60000 (99.93%), error:       154.01109, best error:       493.22141 on epoch:   100, epoch:  101/100, mu: 0.9872440100, learning_rate: 0.0000140612, weight_decay: 0.0000028122
evaluting test set with running mean/variance:
[          , genome  9792] predictions:    4966/   5000 (99.32%), best:   59960/60000 (99.93%), error:       111.60533, best error:       493.22141 on epoch:   100, epoch:  101/100, mu: 0.9872440100, learning_rate: 0.0000140612, weight_decay: 0.0000028122
backpropagation finished successfully!
15:13:24 (19363): called boinc_finish(0)

</stderr_txt>
]]>