📄 trn_6.err
字号:
Classifier MLP Training run Patterns file: h6_d.pat; using all 61094 patterns Final pattern-wts: set all equal, no files read Error function: sum of squares Reg. factor: 1.000e-03 Activation fns. on hidden, output nodes: sinusoid, sinusoid Nos. of input, hidden, output nodes: 128, 128, 10 Boltzmann pruning, thresh. exp(-w^2/T), T 1.000e-04 Will use SCG Initial network weights: from file trn_5.wts Final network weights will be written as file trn_6.wts Stopping criteria (max. no. of iterations 50): (RMS err) <= 0.000e+00 OR (RMS g) <= 0.000e+00 * (RMS w) OR (RMS err) > 9.900e-01 * (RMS err 10 iters ago) OR (OK - NG count) < (count 10 iters ago) + 1. (OK level: 0.000) Long outfile not made SCG: doing <= 50 iterations; 17802 variables. pruned 8635 144 8779 C 9.96894e+04 H 3.30720e+04 R 66.82 M -0.01 T 0.0335 Iter Err ( Ep Ew) OK UNK NG OK UNK NG 0 0.048 (0.048 0.130) 60762 0 332 = 99.5 0.0 0.5 % 99.1 99 99 99 99 100 99 100 100 99 99 pruned 8672 146 8818 C 8.02151e+04 H 3.27822e+04 R 59.13 M -0.01 T 0.0336 pruned 8688 146 8834 C 1.12624e+05 H 3.26778e+04 R 70.98 M -0.01 T 0.0337 pruned 8696 146 8842 C 1.28928e+05 H 3.26423e+04 R 74.68 M -0.01 T 0.0338 pruned 8707 146 8853 C 1.00377e+05 H 3.25496e+04 R 67.57 M -0.01 T 0.0338 pruned 8707 146 8853 C 1.20802e+05 H 3.25679e+04 R 73.04 M -0.01 T 0.0338 pruned 8706 146 8852 C 1.15230e+05 H 3.26028e+04 R 71.71 M -0.01 T 0.0338 pruned 8697 145 8842 C 1.14749e+05 H 3.27217e+04 R 71.48 M -0.01 T 0.0338 pruned 8684 146 8830 C 1.24232e+05 H 3.28439e+04 R 73.56 M -0.01 T 0.0338 pruned 8665 144 8809 C 1.14208e+05 H 3.30485e+04 R 71.06 M -0.01 T 0.0338 pruned 8642 143 8785 C 1.12597e+05 H 3.32652e+04 R 70.46 M -0.01 T 0.0337 10 0.048 (0.047 0.131) 60771 0 323 = 99.5 0.0 0.5 % 99.2 100 99 99 99 100 99 100 100 99 99 pruned 8665 146 8811 C 1.19475e+05 H 3.30468e+04 R 72.34 M -0.01 T 0.0339 pruned 8594 145 8739 C 1.09331e+05 H 3.37008e+04 R 69.18 M -0.01 T 0.0337 pruned 8626 146 8772 C 1.19877e+05 H 3.34393e+04 R 72.11 M -0.01 T 0.0338 pruned 8441 139 8580 C 1.19268e+05 H 3.51021e+04 R 70.57 M -0.01 T 0.0332 pruned 8596 140 8736 C 1.23834e+05 H 3.37448e+04 R 72.75 M -0.01 T 0.0338 pruned 8547 141 8688 C 1.16326e+05 H 3.41443e+04 R 70.65 M -0.01 T 0.0337 pruned 8547 145 8692 C 1.29216e+05 H 3.41520e+04 R 73.57 M -0.01 T 0.0338 pruned 8431 143 8574 C 1.12943e+05 H 3.51633e+04 R 68.87 M -0.01 T 0.0334 pruned 8542 144 8686 C 1.23124e+05 H 3.41762e+04 R 72.24 M -0.01 T 0.0339 pruned 8537 140 8677 C 1.30063e+05 H 3.43136e+04 R 73.62 M -0.01 T 0.0339 20 0.047 (0.047 0.132) 60783 0 311 = 99.5 0.0 0.5 % 99.2 99 99 99 99 100 99 100 100 99 99 pruned 8497 139 8636 C 1.22360e+05 H 3.46870e+04 R 71.65 M -0.01 T 0.0338 pruned 8483 136 8619 C 1.21906e+05 H 3.48371e+04 R 71.42 M -0.01 T 0.0338 pruned 8475 141 8616 C 1.16094e+05 H 3.48813e+04 R 69.95 M -0.01 T 0.0338 pruned 8504 140 8644 C 1.11398e+05 H 3.46650e+04 R 68.88 M -0.01 T 0.0339 pruned 8441 140 8581 C 1.14952e+05 H 3.51846e+04 R 69.39 M -0.01 T 0.0338 pruned 8381 143 8524 C 1.17280e+05 H 3.56384e+04 R 69.61 M -0.01 T 0.0336 pruned 8408 142 8550 C 1.14788e+05 H 3.54220e+04 R 69.14 M -0.01 T 0.0338 pruned 8277 140 8417 C 1.14918e+05 H 3.65215e+04 R 68.22 M -0.01 T 0.0333 pruned 8315 144 8459 C 1.17908e+05 H 3.61254e+04 R 69.36 M -0.01 T 0.0335 pruned 8159 137 8296 C 1.19143e+05 H 3.74269e+04 R 68.59 M -0.01 T 0.0330 30 0.047 (0.047 0.133) 60787 0 307 = 99.5 0.0 0.5 % 99.2 99 99 99 99 100 99 100 100 99 99 pruned 8384 135 8519 C 1.21853e+05 H 3.55320e+04 R 70.84 M -0.01 T 0.0339 pruned 8287 138 8425 C 1.15000e+05 H 3.63442e+04 R 68.40 M -0.01 T 0.0336 pruned 8286 141 8427 C 1.14410e+05 H 3.63130e+04 R 68.26 M -0.01 T 0.0337 pruned 8266 139 8405 C 1.18392e+05 H 3.64598e+04 R 69.20 M -0.01 T 0.0336 pruned 8336 138 8474 C 1.33350e+05 H 3.58388e+04 R 73.12 M -0.01 T 0.0339 pruned 8257 139 8396 C 1.19386e+05 H 3.64826e+04 R 69.44 M -0.01 T 0.0337 pruned 8253 135 8388 C 1.29592e+05 H 3.64741e+04 R 71.85 M -0.01 T 0.0337 pruned 8150 134 8284 C 1.13247e+05 H 3.72614e+04 R 67.10 M -0.01 T 0.0334 pruned 8257 139 8396 C 1.26837e+05 H 3.62731e+04 R 71.40 M -0.01 T 0.0339 pruned 8091 140 8231 C 1.16184e+05 H 3.75872e+04 R 67.65 M -0.01 T 0.0333 40 0.047 (0.047 0.134) 60795 0 299 = 99.5 0.0 0.5 % 99.3 99 99 100 99 100 99 100 100 99 99 pruned 8204 139 8343 C 1.20907e+05 H 3.65920e+04 R 69.74 M -0.01 T 0.0338 pruned 8183 133 8316 C 1.17443e+05 H 3.68133e+04 R 68.65 M -0.01 T 0.0338 pruned 8184 137 8321 C 1.12381e+05 H 3.67101e+04 R 67.33 M -0.01 T 0.0338 pruned 8186 138 8324 C 1.14486e+05 H 3.67056e+04 R 67.94 M -0.01 T 0.0339 pruned 8221 137 8358 C 1.27989e+05 H 3.64064e+04 R 71.56 M -0.01 T 0.0341 pruned 8228 140 8368 C 1.20719e+05 H 3.63196e+04 R 69.91 M -0.01 T 0.0342 pruned 8218 141 8359 C 1.20387e+05 H 3.64022e+04 R 69.76 M -0.01 T 0.0343 pruned 8206 141 8347 C 1.17135e+05 H 3.65005e+04 R 68.84 M -0.01 T 0.0343 pruned 8194 140 8334 C 1.25501e+05 H 3.66367e+04 R 70.81 M -0.01 T 0.0343 pruned 8255 140 8395 C 1.40825e+05 H 3.61247e+04 R 74.35 M -0.01 T 0.0346 50 0.046 (0.046 0.135) 60803 0 291 = 99.5 0.0 0.5 % 99.4 99 99 99 99 100 100 100 100 99 99 oklvl 0.00 # Highest two outputs (mean) 0.959 0.059; mean diff 0.900 key name 0 0 1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9 # key: 0 1 2 3 4 5 6 7 8 9 # row: correct, column: actual # 0: 5908 10 2 0 5 1 5 0 7 1 # 1: 0 6673 9 3 7 2 3 6 6 1 # 2: 3 4 6055 6 3 3 5 6 1 0 # 3: 2 1 7 6046 1 8 0 14 6 0 # 4: 0 0 4 0 5988 0 3 1 0 14 # 5: 1 2 3 9 0 5811 10 0 1 1 # 6: 4 4 2 0 0 2 6037 0 2 0 # 7: 0 1 6 0 4 0 0 6314 0 9 # 8: 1 12 1 2 2 6 2 2 5933 5 # 9: 2 0 1 2 6 2 0 19 5 6038 # unknown # * 0 0 0 0 0 0 0 0 0 0 percent of true IDs correctly identified (rows) 99 99 99 99 100 100 100 100 99 99percent of predicted IDs correctly identified (cols) 100 99 99 100 100 100 100 99 100 99 # mean highest activation level # row: correct, column: actual # key: 0 1 2 3 4 5 6 7 8 9 # 0: 97 56 30 0 46 40 58 0 42 44 # 1: 0 97 42 51 48 80 63 58 59 40 # 2: 53 42 95 47 32 60 64 44 85 0 # 3: 41 52 59 96 36 59 0 60 48 0 # 4: 0 0 36 0 96 0 46 35 0 65 # 5: 77 38 77 62 0 96 64 0 64 55 # 6: 38 52 36 0 0 45 97 0 25 0 # 7: 0 55 54 0 56 0 0 97 0 46 # 8: 42 62 88 45 55 58 69 33 94 52 # 9: 49 0 31 44 49 64 0 59 54 96 # unknown # * 0 0 0 0 0 0 0 0 0 0 Histogram of errors, from 2^(-10) to 1 133781 52693 68976 85497 93670 84248 55610 24329 8405 2984 747 21.9 8.6 11.3 14.0 15.3 13.8 9.1 4.0 1.4 0.5 0.1% Iter Err ( Ep Ew) OK UNK NG OK UNK NG F 50 0.046 (0.046 0.135) 60803 0 291 = 99.5 0.0 0.5 % 99.4 99 99 99 99 100 100 100 100 99 99 thresh right unknown wrong correct rejected 1tr 0.000000 60803 0 291 99.52 0.00 2tr 0.050000 60803 0 291 99.52 0.00 3tr 0.100000 60803 0 291 99.52 0.00 4tr 0.150000 60802 1 291 99.52 0.00 5tr 0.200000 60798 10 286 99.53 0.02 6tr 0.250000 60793 18 283 99.54 0.03 7tr 0.300000 60784 41 269 99.56 0.07 8tr 0.350000 60748 96 250 99.59 0.16 9tr 0.400000 60701 179 214 99.65 0.2910tr 0.450000 60632 269 193 99.68 0.4411tr 0.500000 60517 421 156 99.74 0.6912tr 0.525000 60450 502 142 99.77 0.8213tr 0.550000 60340 630 124 99.79 1.0314tr 0.575000 60228 751 115 99.81 1.2315tr 0.600000 60095 899 100 99.83 1.4716tr 0.625000 59955 1051 88 99.85 1.7217tr 0.650000 59803 1217 74 99.88 1.9918tr 0.675000 59620 1410 64 99.89 2.3119tr 0.700000 59427 1613 54 99.91 2.6420tr 0.725000 59179 1873 42 99.93 3.0721tr 0.750000 58880 2178 36 99.94 3.5622tr 0.775000 58509 2553 32 99.95 4.1823tr 0.800000 58046 3019 29 99.95 4.9424tr 0.810000 57854 3213 27 99.95 5.2625tr 0.820000 57610 3459 25 99.96 5.6626tr 0.830000 57323 3750 21 99.96 6.1427tr 0.840000 57035 4038 21 99.96 6.6128tr 0.850000 56749 4326 19 99.97 7.0829tr 0.860000 56406 4672 16 99.97 7.6530tr 0.870000 56036 5043 15 99.97 8.2531tr 0.880000 55585 5495 14 99.97 8.9932tr 0.890000 55092 5989 13 99.98 9.8033tr 0.900000 54489 6592 13 99.98 10.7934tr 0.905000 54176 6905 13 99.98 11.3035tr 0.910000 53802 7279 13 99.98 11.9136tr 0.915000 53405 7676 13 99.98 12.5637tr 0.920000 52953 8130 11 99.98 13.3138tr 0.925000 52465 8618 11 99.98 14.1139tr 0.930000 51908 9176 10 99.98 15.0240tr 0.935000 51292 9792 10 99.98 16.0341tr 0.940000 50577 10507 10 99.98 17.2042tr 0.945000 49769 11315 10 99.98 18.5243tr 0.950000 48802 12283 9 99.98 20.1144tr 0.955000 47654 13432 8 99.98 21.9945tr 0.960000 46302 14785 7 99.98 24.2046tr 0.965000 44714 16373 7 99.98 26.8047tr 0.970000 42889 18198 7 99.98 29.7948tr 0.975000 40596 20491 7 99.98 33.5449tr 0.980000 37657 23432 5 99.99 38.3550tr 0.985000 33851 27238 5 99.99 44.5851tr 0.990000 28676 32414 4 99.99 53.0652tr 0.995000 21033 40059 2 99.99 65.57 Iter 50; ierr 1 : iteration limit Used 51 iterations; 151 function calls; Err 0.046; |g|/|w| 1.539e-05 Rms change in weights 0.021 User+system time used: 15877.4 (s) 4:24:37.4 (h:m:s) Wrote weights as file trn_6.wts⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -