OPTION VALUES: Model directory: ./Fold07-IOB2/ Training data file: train.tagged Testing data file: test.tagged Unlabeled data file: data.untagged Model file: model.txt Training log file (this one): trainlog.txt Label representation: IOB2 Second-order Markov CRFs Number of labels: 4 Number of training sequences: 1046 (one data partition) Number of testing sequences: 49 (one data partition) Number of unlabeled sequences: 0 Number of context predicates: 801011 Number of features: 1527555 Feature rare threshold: 1 Context predicate rare threshold: 1 Using multiple rare thresholds for features: 0 Highlight feature: 0 Number of training iterations: 150 Initial lambda value: 0.0000 Sigma square (for smoothing): 100.0000 Epsilon for L-BFGS convergence: 0.000100 Number of approximated hessian matrixes: 7 Start to train ... Iteration: 1 Log-likelihood = -3111978.535140 Norm (log-likelihood gradient vector) = 642118.864183 Norm (lambda vector) = 0.000000 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 11641 9333 80.17 70.19 74.85 i-np 15046 15091 12244 81.13 81.38 81.26 o 23014 24625 20579 83.57 89.42 86.40 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.63 80.33 80.97 Avg2. 51357 51357 42156 82.08 82.08 82.08 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 11641 7252 62.30 54.54 58.16 ----- ------ ----- ----- ------- ------- ------------- Avg1. 62.30 54.54 58.16 Avg2. 13297 11641 7252 62.30 54.54 58.16 Current max chunk-based F1: 58.16 (iteration 1) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 2 Log-likelihood = -2503227.440756 Norm (log-likelihood gradient vector) = 562109.855482 Norm (lambda vector) = 1.000000 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 11948 9821 82.20 73.86 77.81 i-np 15046 15944 12952 81.23 86.08 83.59 o 23014 23465 20520 87.45 89.16 88.30 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.63 83.03 83.33 Avg2. 51357 51357 43293 84.30 84.30 84.30 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 11948 7811 65.37 58.74 61.88 ----- ------ ----- ----- ------- ------- ------------- Avg1. 65.37 58.74 61.88 Avg2. 13297 11948 7811 65.37 58.74 61.88 Current max chunk-based F1: 61.88 (iteration 2) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 3 Log-likelihood = -994219.770329 Norm (log-likelihood gradient vector) = 384994.159268 Norm (lambda vector) = 10.188414 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 10986 10164 92.52 76.44 83.71 i-np 15046 16844 14219 84.42 94.50 89.18 o 23014 23527 21785 92.60 94.66 93.62 ----- ------ ----- ----- ------- ------- ------------- Avg1. 89.84 88.53 89.18 Avg2. 51357 51357 46168 89.90 89.90 89.90 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 10986 8861 80.66 66.64 72.98 ----- ------ ----- ----- ------- ------- ------------- Avg1. 80.66 66.64 72.98 Avg2. 13297 10986 8861 80.66 66.64 72.98 Current max chunk-based F1: 72.98 (iteration 3) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 4 Log-likelihood = -721421.102236 Norm (log-likelihood gradient vector) = 171681.280221 Norm (lambda vector) = 9.821681 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 12807 11569 90.33 87.00 88.64 i-np 15046 14984 13752 91.78 91.40 91.59 o 23014 23566 22188 94.15 96.41 95.27 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.09 91.61 91.85 Avg2. 51357 51357 47509 92.51 92.51 92.51 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 12807 10601 82.78 79.72 81.22 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.78 79.72 81.22 Avg2. 13297 12807 10601 82.78 79.72 81.22 Current max chunk-based F1: 81.22 (iteration 4) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 5 Log-likelihood = -665451.699851 Norm (log-likelihood gradient vector) = 108125.313798 Norm (lambda vector) = 9.349900 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 12953 11889 91.79 89.41 90.58 i-np 15046 14489 13682 94.43 90.93 92.65 o 23014 23915 22449 93.87 97.54 95.67 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.36 92.63 92.99 Avg2. 51357 51357 48020 93.50 93.50 93.50 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 12953 11155 86.12 83.89 84.99 ----- ------ ----- ----- ------- ------- ------------- Avg1. 86.12 83.89 84.99 Avg2. 13297 12953 11155 86.12 83.89 84.99 Current max chunk-based F1: 84.99 (iteration 5) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 6 Log-likelihood = -629863.381014 Norm (log-likelihood gradient vector) = 73935.784457 Norm (lambda vector) = 9.395013 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 12835 11986 93.39 90.14 91.73 i-np 15046 14769 13972 94.60 92.86 93.72 o 23014 23753 22513 94.78 97.82 96.28 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.26 93.61 93.93 Avg2. 51357 51357 48471 94.38 94.38 94.38 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 12835 11384 88.69 85.61 87.13 ----- ------ ----- ----- ------- ------- ------------- Avg1. 88.69 85.61 87.13 Avg2. 13297 12835 11384 88.69 85.61 87.13 Current max chunk-based F1: 87.13 (iteration 6) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 7 Log-likelihood = -575967.440281 Norm (log-likelihood gradient vector) = 71453.842737 Norm (lambda vector) = 10.306118 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 12880 12155 94.37 91.41 92.87 i-np 15046 14732 14055 95.40 93.41 94.40 o 23014 23745 22612 95.23 98.25 96.72 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.00 94.36 94.68 Avg2. 51357 51357 48822 95.06 95.06 95.06 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 12880 11615 90.18 87.35 88.74 ----- ------ ----- ----- ------- ------- ------------- Avg1. 90.18 87.35 88.74 Avg2. 13297 12880 11615 90.18 87.35 88.74 Current max chunk-based F1: 88.74 (iteration 7) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 8 Log-likelihood = -498312.679548 Norm (log-likelihood gradient vector) = 59248.762945 Norm (lambda vector) = 12.130654 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 12739 12207 95.82 91.80 93.77 i-np 15046 15638 14512 92.80 96.45 94.59 o 23014 22980 22347 97.25 97.10 97.17 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.29 95.12 95.20 Avg2. 51357 51357 49066 95.54 95.54 95.54 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 12739 11591 90.99 87.17 89.04 ----- ------ ----- ----- ------- ------- ------------- Avg1. 90.99 87.17 89.04 Avg2. 13297 12739 11591 90.99 87.17 89.04 Current max chunk-based F1: 89.04 (iteration 8) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 9 Log-likelihood = -407592.524954 Norm (log-likelihood gradient vector) = 72851.139380 Norm (lambda vector) = 16.693663 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 12969 12403 95.64 93.28 94.44 i-np 15046 15020 14343 95.49 95.33 95.41 o 23014 23368 22612 96.76 98.25 97.50 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.96 95.62 95.79 Avg2. 51357 51357 49358 96.11 96.11 96.11 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 12969 11912 91.85 89.58 90.70 ----- ------ ----- ----- ------- ------- ------------- Avg1. 91.85 89.58 90.70 Avg2. 13297 12969 11912 91.85 89.58 90.70 Current max chunk-based F1: 90.70 (iteration 9) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 10 Log-likelihood = -369104.969361 Norm (log-likelihood gradient vector) = 30582.090195 Norm (lambda vector) = 18.101868 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13067 12494 95.61 93.96 94.78 i-np 15046 14861 14297 96.20 95.02 95.61 o 23014 23429 22673 96.77 98.52 97.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.20 95.83 96.02 Avg2. 51357 51357 49464 96.31 96.31 96.31 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13067 12026 92.03 90.44 91.23 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.03 90.44 91.23 Avg2. 13297 13067 12026 92.03 90.44 91.23 Current max chunk-based F1: 91.23 (iteration 10) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 11 Log-likelihood = -345135.721952 Norm (log-likelihood gradient vector) = 31408.650048 Norm (lambda vector) = 20.143884 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13161 12575 95.55 94.57 95.06 i-np 15046 14762 14273 96.69 94.86 95.77 o 23014 23434 22700 96.87 98.64 97.74 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.37 96.02 96.19 Avg2. 51357 51357 49548 96.48 96.48 96.48 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13161 12129 92.16 91.22 91.68 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.16 91.22 91.68 Avg2. 13297 13161 12129 92.16 91.22 91.68 Current max chunk-based F1: 91.68 (iteration 11) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 12 Log-likelihood = -320926.670758 Norm (log-likelihood gradient vector) = 28400.051694 Norm (lambda vector) = 22.618805 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 12859 12429 96.66 93.47 95.04 i-np 15046 15502 14566 93.96 96.81 95.36 o 23014 22996 22475 97.73 97.66 97.70 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.12 95.98 96.05 Avg2. 51357 51357 49470 96.33 96.33 96.33 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 12859 11896 92.51 89.46 90.96 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.51 89.46 90.96 Avg2. 13297 12859 11896 92.51 89.46 90.96 Current max chunk-based F1: 91.68 (iteration 11) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 13 Log-likelihood = -318061.933023 Norm (log-likelihood gradient vector) = 92828.596096 Norm (lambda vector) = 27.581849 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13059 12585 96.37 94.65 95.50 i-np 15046 15152 14502 95.71 96.38 96.05 o 23014 23146 22602 97.65 98.21 97.93 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.58 96.41 96.49 Avg2. 51357 51357 49689 96.75 96.75 96.75 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13059 12156 93.09 91.42 92.24 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.09 91.42 92.24 Avg2. 13297 13059 12156 93.09 91.42 92.24 Current max chunk-based F1: 92.24 (iteration 13) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 14 Log-likelihood = -289988.033728 Norm (log-likelihood gradient vector) = 26273.147480 Norm (lambda vector) = 28.048311 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13106 12625 96.33 94.95 95.63 i-np 15046 15089 14493 96.05 96.32 96.19 o 23014 23162 22624 97.68 98.31 97.99 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.69 96.53 96.61 Avg2. 51357 51357 49742 96.86 96.86 96.86 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13106 12207 93.14 91.80 92.47 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.14 91.80 92.47 Avg2. 13297 13106 12207 93.14 91.80 92.47 Current max chunk-based F1: 92.47 (iteration 14) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 15 Log-likelihood = -286326.674882 Norm (log-likelihood gradient vector) = 17710.612843 Norm (lambda vector) = 28.323854 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13149 12653 96.23 95.16 95.69 i-np 15046 15020 14461 96.28 96.11 96.20 o 23014 23188 22642 97.65 98.38 98.01 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.72 96.55 96.63 Avg2. 51357 51357 49756 96.88 96.88 96.88 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13149 12241 93.09 92.06 92.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.09 92.06 92.57 Avg2. 13297 13149 12241 93.09 92.06 92.57 Current max chunk-based F1: 92.57 (iteration 15) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 16 Log-likelihood = -281596.258893 Norm (log-likelihood gradient vector) = 19015.789685 Norm (lambda vector) = 28.938042 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13178 12687 96.27 95.41 95.84 i-np 15046 14972 14457 96.56 96.09 96.32 o 23014 23207 22662 97.65 98.47 98.06 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.83 96.66 96.74 Avg2. 51357 51357 49806 96.98 96.98 96.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13178 12291 93.27 92.43 92.85 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.27 92.43 92.85 Avg2. 13297 13178 12291 93.27 92.43 92.85 Current max chunk-based F1: 92.85 (iteration 16) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 17 Log-likelihood = -268659.952383 Norm (log-likelihood gradient vector) = 18613.042280 Norm (lambda vector) = 30.695784 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13127 12661 96.45 95.22 95.83 i-np 15046 15146 14549 96.06 96.70 96.38 o 23014 23084 22602 97.91 98.21 98.06 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.81 96.71 96.76 Avg2. 51357 51357 49812 96.99 96.99 96.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13127 12266 93.44 92.25 92.84 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.44 92.25 92.84 Avg2. 13297 13127 12266 93.44 92.25 92.84 Current max chunk-based F1: 92.85 (iteration 16) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 18 Log-likelihood = -244258.828298 Norm (log-likelihood gradient vector) = 30791.517621 Norm (lambda vector) = 34.243684 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13342 12767 95.69 96.01 95.85 i-np 15046 14695 14321 97.45 95.18 96.30 o 23014 23320 22709 97.38 98.67 98.02 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.84 96.62 96.73 Avg2. 51357 51357 49797 96.96 96.96 96.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13342 12381 92.80 93.11 92.95 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.80 93.11 92.95 Avg2. 13297 13342 12381 92.80 93.11 92.95 Current max chunk-based F1: 92.95 (iteration 18) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 19 Log-likelihood = -241692.232023 Norm (log-likelihood gradient vector) = 62483.399071 Norm (lambda vector) = 40.074256 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13234 12719 96.11 95.65 95.88 i-np 15046 14930 14455 96.82 96.07 96.44 o 23014 23193 22658 97.69 98.45 98.07 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.87 96.73 96.80 Avg2. 51357 51357 49832 97.03 97.03 97.03 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13234 12347 93.30 92.86 93.08 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.30 92.86 93.08 Avg2. 13297 13234 12347 93.30 92.86 93.08 Current max chunk-based F1: 93.08 (iteration 19) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 20 Log-likelihood = -234117.846959 Norm (log-likelihood gradient vector) = 31998.355374 Norm (lambda vector) = 37.016051 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13173 12692 96.35 95.45 95.90 i-np 15046 15033 14507 96.50 96.42 96.46 o 23014 23151 22642 97.80 98.38 98.09 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.88 96.75 96.82 Avg2. 51357 51357 49841 97.05 97.05 97.05 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13173 12313 93.47 92.60 93.03 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.47 92.60 93.03 Avg2. 13297 13173 12313 93.47 92.60 93.03 Current max chunk-based F1: 93.08 (iteration 19) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 21 Log-likelihood = -221563.491542 Norm (log-likelihood gradient vector) = 18114.369816 Norm (lambda vector) = 39.255243 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13111 12676 96.68 95.33 96.00 i-np 15046 15156 14583 96.22 96.92 96.57 o 23014 23090 22621 97.97 98.29 98.13 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.96 96.85 96.90 Avg2. 51357 51357 49880 97.12 97.12 97.12 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13111 12308 93.88 92.56 93.21 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.88 92.56 93.21 Avg2. 13297 13111 12308 93.88 92.56 93.21 Current max chunk-based F1: 93.21 (iteration 21) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 22 Log-likelihood = -210278.763869 Norm (log-likelihood gradient vector) = 10491.101506 Norm (lambda vector) = 41.000600 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13083 12676 96.89 95.33 96.10 i-np 15046 15231 14634 96.08 97.26 96.67 o 23014 23043 22611 98.13 98.25 98.19 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.03 96.95 96.99 Avg2. 51357 51357 49921 97.20 97.20 97.20 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13083 12314 94.12 92.61 93.36 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.12 92.61 93.36 Avg2. 13297 13083 12314 94.12 92.61 93.36 Current max chunk-based F1: 93.36 (iteration 22) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 23 Log-likelihood = -202018.301290 Norm (log-likelihood gradient vector) = 12515.716304 Norm (lambda vector) = 42.191960 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13130 12718 96.86 95.65 96.25 i-np 15046 15147 14619 96.51 97.16 96.84 o 23014 23080 22645 98.12 98.40 98.26 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.16 97.07 97.12 Avg2. 51357 51357 49982 97.32 97.32 97.32 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13130 12382 94.30 93.12 93.71 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.30 93.12 93.71 Avg2. 13297 13130 12382 94.30 93.12 93.71 Current max chunk-based F1: 93.71 (iteration 23) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 24 Log-likelihood = -192842.925830 Norm (log-likelihood gradient vector) = 9826.308329 Norm (lambda vector) = 43.644840 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13042 12671 97.16 95.29 96.21 i-np 15046 15362 14698 95.68 97.69 96.67 o 23014 22953 22576 98.36 98.10 98.23 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.06 97.03 97.04 Avg2. 51357 51357 49945 97.25 97.25 97.25 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13042 12304 94.34 92.53 93.43 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.34 92.53 93.43 Avg2. 13297 13042 12304 94.34 92.53 93.43 Current max chunk-based F1: 93.71 (iteration 23) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 25 Log-likelihood = -185701.881344 Norm (log-likelihood gradient vector) = 29404.117721 Norm (lambda vector) = 45.537133 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13125 12732 97.01 95.75 96.37 i-np 15046 15177 14642 96.47 97.31 96.89 o 23014 23055 22643 98.21 98.39 98.30 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.23 97.15 97.19 Avg2. 51357 51357 50017 97.39 97.39 97.39 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13125 12401 94.48 93.26 93.87 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.48 93.26 93.87 Avg2. 13297 13125 12401 94.48 93.26 93.87 Current max chunk-based F1: 93.87 (iteration 25) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 26 Log-likelihood = -182829.122577 Norm (log-likelihood gradient vector) = 11269.094288 Norm (lambda vector) = 45.816227 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13195 12780 96.85 96.11 96.48 i-np 15046 15042 14588 96.98 96.96 96.97 o 23014 23120 22678 98.09 98.54 98.31 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.31 97.20 97.26 Avg2. 51357 51357 50046 97.45 97.45 97.45 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13195 12463 94.45 93.73 94.09 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.45 93.73 94.09 Avg2. 13297 13195 12463 94.45 93.73 94.09 Current max chunk-based F1: 94.09 (iteration 26) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 27 Log-likelihood = -180182.049775 Norm (log-likelihood gradient vector) = 9449.682353 Norm (lambda vector) = 47.279143 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13217 12805 96.88 96.30 96.59 i-np 15046 15015 14587 97.15 96.95 97.05 o 23014 23125 22687 98.11 98.58 98.34 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.38 97.28 97.33 Avg2. 51357 51357 50079 97.51 97.51 97.51 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13217 12497 94.55 93.98 94.27 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.55 93.98 94.27 Avg2. 13297 13217 12497 94.55 93.98 94.27 Current max chunk-based F1: 94.27 (iteration 27) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 28 Log-likelihood = -176803.050844 Norm (log-likelihood gradient vector) = 12962.811314 Norm (lambda vector) = 49.404789 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13192 12802 97.04 96.28 96.66 i-np 15046 15090 14632 96.96 97.25 97.11 o 23014 23075 22666 98.23 98.49 98.36 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.41 97.34 97.37 Avg2. 51357 51357 50100 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13192 12495 94.72 93.97 94.34 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.72 93.97 94.34 Avg2. 13297 13192 12495 94.72 93.97 94.34 Current max chunk-based F1: 94.34 (iteration 28) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 29 Log-likelihood = -170907.917220 Norm (log-likelihood gradient vector) = 12361.887279 Norm (lambda vector) = 52.669737 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13231 12828 96.95 96.47 96.71 i-np 15046 14988 14584 97.30 96.93 97.12 o 23014 23138 22700 98.11 98.64 98.37 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.46 97.35 97.40 Avg2. 51357 51357 50112 97.58 97.58 97.58 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13231 12522 94.64 94.17 94.41 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.64 94.17 94.41 Avg2. 13297 13231 12522 94.64 94.17 94.41 Current max chunk-based F1: 94.41 (iteration 29) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 30 Log-likelihood = -166599.679699 Norm (log-likelihood gradient vector) = 28907.976769 Norm (lambda vector) = 58.147118 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13212 12827 97.09 96.47 96.77 i-np 15046 15034 14618 97.23 97.16 97.19 o 23014 23111 22694 98.20 98.61 98.40 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.50 97.41 97.46 Avg2. 51357 51357 50139 97.63 97.63 97.63 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13212 12529 94.83 94.22 94.53 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.83 94.22 94.53 Avg2. 13297 13212 12529 94.83 94.22 94.53 Current max chunk-based F1: 94.53 (iteration 30) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 31 Log-likelihood = -165867.672795 Norm (log-likelihood gradient vector) = 14439.848458 Norm (lambda vector) = 55.442406 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13146 12786 97.26 96.16 96.71 i-np 15046 15165 14677 96.78 97.55 97.16 o 23014 23046 22660 98.33 98.46 98.39 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.46 97.39 97.42 Avg2. 51357 51357 50123 97.60 97.60 97.60 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13146 12481 94.94 93.86 94.40 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.94 93.86 94.40 Avg2. 13297 13146 12481 94.94 93.86 94.40 Current max chunk-based F1: 94.53 (iteration 30) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 32 Log-likelihood = -161301.076196 Norm (log-likelihood gradient vector) = 7368.459738 Norm (lambda vector) = 56.641860 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13099 12763 97.43 95.98 96.70 i-np 15046 15259 14713 96.42 97.79 97.10 o 23014 22999 22631 98.40 98.34 98.37 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.42 97.37 97.39 Avg2. 51357 51357 50107 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13099 12452 95.06 93.65 94.35 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.06 93.65 94.35 Avg2. 13297 13099 12452 95.06 93.65 94.35 Current max chunk-based F1: 94.53 (iteration 30) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 33 Log-likelihood = -156313.631869 Norm (log-likelihood gradient vector) = 8714.992249 Norm (lambda vector) = 57.782959 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13085 12758 97.50 95.95 96.72 i-np 15046 15288 14728 96.34 97.89 97.11 o 23014 22984 22626 98.44 98.31 98.38 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.43 97.38 97.40 Avg2. 51357 51357 50112 97.58 97.58 97.58 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13085 12440 95.07 93.55 94.31 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.07 93.55 94.31 Avg2. 13297 13085 12440 95.07 93.55 94.31 Current max chunk-based F1: 94.53 (iteration 30) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 34 Log-likelihood = -151941.800048 Norm (log-likelihood gradient vector) = 11196.139995 Norm (lambda vector) = 58.876472 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13108 12777 97.47 96.09 96.78 i-np 15046 15238 14716 96.57 97.81 97.19 o 23014 23011 22647 98.42 98.41 98.41 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.49 97.43 97.46 Avg2. 51357 51357 50140 97.63 97.63 97.63 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13108 12472 95.15 93.80 94.47 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.15 93.80 94.47 Avg2. 13297 13108 12472 95.15 93.80 94.47 Current max chunk-based F1: 94.53 (iteration 30) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 35 Log-likelihood = -146243.754424 Norm (log-likelihood gradient vector) = 9999.050678 Norm (lambda vector) = 61.124768 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13034 12725 97.63 95.70 96.65 i-np 15046 15406 14758 95.79 98.09 96.93 o 23014 22917 22585 98.55 98.14 98.34 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.32 97.31 97.32 Avg2. 51357 51357 50068 97.49 97.49 97.49 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13034 12374 94.94 93.06 93.99 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.94 93.06 93.99 Avg2. 13297 13034 12374 94.94 93.06 93.99 Current max chunk-based F1: 94.53 (iteration 30) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 36 Log-likelihood = -148638.043295 Norm (log-likelihood gradient vector) = 33865.142602 Norm (lambda vector) = 63.110783 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13075 12760 97.59 95.96 96.77 i-np 15046 15304 14740 96.31 97.97 97.13 o 23014 22978 22629 98.48 98.33 98.40 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.46 97.42 97.44 Avg2. 51357 51357 50129 97.61 97.61 97.61 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13075 12446 95.19 93.60 94.39 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.19 93.60 94.39 Avg2. 13297 13075 12446 95.19 93.60 94.39 Current max chunk-based F1: 94.53 (iteration 30) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 37 Log-likelihood = -144800.561663 Norm (log-likelihood gradient vector) = 15907.793369 Norm (lambda vector) = 61.887845 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13156 12813 97.39 96.36 96.87 i-np 15046 15166 14695 96.89 97.67 97.28 o 23014 23035 22667 98.40 98.49 98.45 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.56 97.51 97.53 Avg2. 51357 51357 50175 97.70 97.70 97.70 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13156 12515 95.13 94.12 94.62 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.13 94.12 94.62 Avg2. 13297 13156 12515 95.13 94.12 94.62 Current max chunk-based F1: 94.62 (iteration 37) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 38 Log-likelihood = -142813.895080 Norm (log-likelihood gradient vector) = 7947.865749 Norm (lambda vector) = 63.322532 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13200 12835 97.23 96.53 96.88 i-np 15046 15073 14653 97.21 97.39 97.30 o 23014 23084 22693 98.31 98.61 98.46 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.58 97.51 97.55 Avg2. 51357 51357 50181 97.71 97.71 97.71 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13200 12546 95.05 94.35 94.70 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.05 94.35 94.70 Avg2. 13297 13200 12546 95.05 94.35 94.70 Current max chunk-based F1: 94.70 (iteration 38) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 39 Log-likelihood = -142009.561583 Norm (log-likelihood gradient vector) = 6607.049963 Norm (lambda vector) = 64.056410 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13214 12841 97.18 96.57 96.87 i-np 15046 15050 14640 97.28 97.30 97.29 o 23014 23093 22696 98.28 98.62 98.45 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.58 97.50 97.54 Avg2. 51357 51357 50177 97.70 97.70 97.70 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13214 12552 94.99 94.40 94.69 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.99 94.40 94.69 Avg2. 13297 13214 12552 94.99 94.40 94.69 Current max chunk-based F1: 94.70 (iteration 38) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 40 Log-likelihood = -140429.046539 Norm (log-likelihood gradient vector) = 8002.273751 Norm (lambda vector) = 64.457087 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13210 12848 97.26 96.62 96.94 i-np 15046 15053 14646 97.30 97.34 97.32 o 23014 23094 22703 98.31 98.65 98.48 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.62 97.54 97.58 Avg2. 51357 51357 50197 97.74 97.74 97.74 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13210 12558 95.06 94.44 94.75 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.06 94.44 94.75 Avg2. 13297 13210 12558 95.06 94.44 94.75 Current max chunk-based F1: 94.75 (iteration 40) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 41 Log-likelihood = -136882.456921 Norm (log-likelihood gradient vector) = 7991.994203 Norm (lambda vector) = 65.321732 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13284 12881 96.97 96.87 96.92 i-np 15046 14905 14572 97.77 96.85 97.31 o 23014 23168 22739 98.15 98.81 98.48 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.63 97.51 97.57 Avg2. 51357 51357 50192 97.73 97.73 97.73 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13284 12598 94.84 94.74 94.79 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.84 94.74 94.79 Avg2. 13297 13284 12598 94.84 94.74 94.79 Current max chunk-based F1: 94.79 (iteration 41) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 42 Log-likelihood = -137816.808610 Norm (log-likelihood gradient vector) = 42417.303896 Norm (lambda vector) = 66.993020 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13248 12869 97.14 96.78 96.96 i-np 15046 14988 14625 97.58 97.20 97.39 o 23014 23121 22720 98.27 98.72 98.49 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.66 97.57 97.61 Avg2. 51357 51357 50214 97.77 97.77 97.77 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13248 12594 95.06 94.71 94.89 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.06 94.71 94.89 Avg2. 13297 13248 12594 95.06 94.71 94.89 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 43 Log-likelihood = -133804.052727 Norm (log-likelihood gradient vector) = 19835.243268 Norm (lambda vector) = 66.136484 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13200 12843 97.30 96.59 96.94 i-np 15046 15069 14660 97.29 97.43 97.36 o 23014 23088 22703 98.33 98.65 98.49 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.64 97.56 97.60 Avg2. 51357 51357 50206 97.76 97.76 97.76 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13200 12558 95.14 94.44 94.79 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.14 94.44 94.79 Avg2. 13297 13200 12558 95.14 94.44 94.79 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 44 Log-likelihood = -130326.698579 Norm (log-likelihood gradient vector) = 7386.749740 Norm (lambda vector) = 66.654189 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13185 12835 97.35 96.53 96.93 i-np 15046 15098 14672 97.18 97.51 97.35 o 23014 23074 22695 98.36 98.61 98.49 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.63 97.55 97.59 Avg2. 51357 51357 50202 97.75 97.75 97.75 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13185 12548 95.17 94.37 94.77 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.17 94.37 94.77 Avg2. 13297 13185 12548 95.17 94.37 94.77 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 45 Log-likelihood = -127725.971823 Norm (log-likelihood gradient vector) = 5556.210730 Norm (lambda vector) = 67.229455 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13168 12830 97.43 96.49 96.96 i-np 15046 15131 14692 97.10 97.65 97.37 o 23014 23058 22690 98.40 98.59 98.50 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.65 97.58 97.61 Avg2. 51357 51357 50212 97.77 97.77 97.77 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13168 12546 95.28 94.35 94.81 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.28 94.35 94.81 Avg2. 13297 13168 12546 95.28 94.35 94.81 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 46 Log-likelihood = -125199.203656 Norm (log-likelihood gradient vector) = 6782.248689 Norm (lambda vector) = 67.828204 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13148 12817 97.48 96.39 96.93 i-np 15046 15173 14705 96.92 97.73 97.32 o 23014 23036 22676 98.44 98.53 98.48 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.61 97.55 97.58 Avg2. 51357 51357 50198 97.74 97.74 97.74 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13148 12531 95.31 94.24 94.77 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.31 94.24 94.77 Avg2. 13297 13148 12531 95.31 94.24 94.77 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 47 Log-likelihood = -122182.968348 Norm (log-likelihood gradient vector) = 10058.396824 Norm (lambda vector) = 68.987548 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13217 12852 97.24 96.65 96.95 i-np 15046 15008 14629 97.47 97.23 97.35 o 23014 23132 22724 98.24 98.74 98.49 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.65 97.54 97.60 Avg2. 51357 51357 50205 97.76 97.76 97.76 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13217 12571 95.11 94.54 94.83 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.11 94.54 94.83 Avg2. 13297 13217 12571 95.11 94.54 94.83 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 48 Log-likelihood = -119936.591924 Norm (log-likelihood gradient vector) = 18060.674020 Norm (lambda vector) = 71.157403 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13179 12839 97.42 96.56 96.99 i-np 15046 15097 14674 97.20 97.53 97.36 o 23014 23081 22703 98.36 98.65 98.51 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.66 97.58 97.62 Avg2. 51357 51357 50216 97.78 97.78 97.78 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13179 12553 95.25 94.40 94.83 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.25 94.40 94.83 Avg2. 13297 13179 12553 95.25 94.40 94.83 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 49 Log-likelihood = -119604.524064 Norm (log-likelihood gradient vector) = 8051.251039 Norm (lambda vector) = 70.032565 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13171 12836 97.46 96.53 96.99 i-np 15046 15108 14684 97.19 97.59 97.39 o 23014 23078 22703 98.38 98.65 98.51 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.68 97.59 97.63 Avg2. 51357 51357 50223 97.79 97.79 97.79 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13171 12557 95.34 94.43 94.88 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.34 94.43 94.88 Avg2. 13297 13171 12557 95.34 94.43 94.88 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 50 Log-likelihood = -118282.101678 Norm (log-likelihood gradient vector) = 5086.543252 Norm (lambda vector) = 70.179157 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13174 12832 97.40 96.50 96.95 i-np 15046 15096 14676 97.22 97.54 97.38 o 23014 23087 22702 98.33 98.64 98.49 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.65 97.56 97.61 Avg2. 51357 51357 50210 97.77 97.77 97.77 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13174 12557 95.32 94.43 94.87 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.32 94.43 94.87 Avg2. 13297 13174 12557 95.32 94.43 94.87 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 51 Log-likelihood = -116573.637698 Norm (log-likelihood gradient vector) = 6523.095451 Norm (lambda vector) = 70.340747 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13169 12825 97.39 96.45 96.92 i-np 15046 15112 14677 97.12 97.55 97.33 o 23014 23076 22693 98.34 98.61 98.47 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.62 97.53 97.58 Avg2. 51357 51357 50195 97.74 97.74 97.74 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13169 12547 95.28 94.36 94.82 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.28 94.36 94.82 Avg2. 13297 13169 12547 95.28 94.36 94.82 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 52 Log-likelihood = -113799.651357 Norm (log-likelihood gradient vector) = 7755.942732 Norm (lambda vector) = 70.691057 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13137 12804 97.47 96.29 96.88 i-np 15046 15191 14697 96.75 97.68 97.21 o 23014 23029 22662 98.41 98.47 98.44 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.54 97.48 97.51 Avg2. 51357 51357 50163 97.68 97.68 97.68 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13137 12509 95.22 94.07 94.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.22 94.07 94.64 Avg2. 13297 13137 12509 95.22 94.07 94.64 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 53 Log-likelihood = -110127.550234 Norm (log-likelihood gradient vector) = 5568.423880 Norm (lambda vector) = 71.315966 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13207 12851 97.30 96.65 96.97 i-np 15046 15054 14647 97.30 97.35 97.32 o 23014 23096 22706 98.31 98.66 98.49 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.64 97.55 97.59 Avg2. 51357 51357 50204 97.75 97.75 97.75 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13207 12573 95.20 94.56 94.88 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.20 94.56 94.88 Avg2. 13297 13207 12573 95.20 94.56 94.88 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 54 Log-likelihood = -106533.103394 Norm (log-likelihood gradient vector) = 16116.365473 Norm (lambda vector) = 72.268138 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13153 12821 97.48 96.42 96.95 i-np 15046 15169 14698 96.89 97.69 97.29 o 23014 23035 22673 98.43 98.52 98.47 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.60 97.54 97.57 Avg2. 51357 51357 50192 97.73 97.73 97.73 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13153 12537 95.32 94.28 94.80 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.32 94.28 94.80 Avg2. 13297 13153 12537 95.32 94.28 94.80 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 55 Log-likelihood = -104608.878319 Norm (log-likelihood gradient vector) = 4405.527258 Norm (lambda vector) = 72.447855 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13147 12815 97.47 96.38 96.92 i-np 15046 15182 14700 96.83 97.70 97.26 o 23014 23028 22666 98.43 98.49 98.46 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.58 97.52 97.55 Avg2. 51357 51357 50181 97.71 97.71 97.71 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13147 12529 95.30 94.22 94.76 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.30 94.22 94.76 Avg2. 13297 13147 12529 95.30 94.22 94.76 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 56 Log-likelihood = -104489.076250 Norm (log-likelihood gradient vector) = 5147.970102 Norm (lambda vector) = 72.417447 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13157 12824 97.47 96.44 96.95 i-np 15046 15168 14698 96.90 97.69 97.29 o 23014 23032 22672 98.44 98.51 98.48 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.60 97.55 97.58 Avg2. 51357 51357 50194 97.74 97.74 97.74 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13157 12539 95.30 94.30 94.80 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.30 94.30 94.80 Avg2. 13297 13157 12539 95.30 94.30 94.80 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 57 Log-likelihood = -104040.977000 Norm (log-likelihood gradient vector) = 5671.765898 Norm (lambda vector) = 72.582408 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13055 12769 97.81 96.03 96.91 i-np 15046 15383 14787 96.13 98.28 97.19 o 23014 22919 22613 98.66 98.26 98.46 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.53 97.52 97.53 Avg2. 51357 51357 50169 97.69 97.69 97.69 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13055 12459 95.43 93.70 94.56 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.43 93.70 94.56 Avg2. 13297 13055 12459 95.43 93.70 94.56 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 58 Log-likelihood = -103910.705105 Norm (log-likelihood gradient vector) = 20597.138342 Norm (lambda vector) = 73.376525 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13115 12803 97.62 96.28 96.95 i-np 15046 15264 14738 96.55 97.95 97.25 o 23014 22978 22645 98.55 98.40 98.47 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.58 97.54 97.56 Avg2. 51357 51357 50186 97.72 97.72 97.72 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13115 12509 95.38 94.07 94.72 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.38 94.07 94.72 Avg2. 13297 13115 12509 95.38 94.07 94.72 Current max chunk-based F1: 94.89 (iteration 42) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 59 Log-likelihood = -103059.304415 Norm (log-likelihood gradient vector) = 9418.400678 Norm (lambda vector) = 72.926630 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13160 12840 97.57 96.56 97.06 i-np 15046 15164 14713 97.03 97.79 97.40 o 23014 23033 22683 98.48 98.56 98.52 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.69 97.64 97.66 Avg2. 51357 51357 50236 97.82 97.82 97.82 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13160 12569 95.51 94.53 95.01 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.51 94.53 95.01 Avg2. 13297 13160 12569 95.51 94.53 95.01 Current max chunk-based F1: 95.01 (iteration 59) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 60 Log-likelihood = -101549.485507 Norm (log-likelihood gradient vector) = 4506.641543 Norm (lambda vector) = 73.422784 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13165 12844 97.56 96.59 97.08 i-np 15046 15150 14710 97.10 97.77 97.43 o 23014 23042 22691 98.48 98.60 98.54 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.71 97.65 97.68 Avg2. 51357 51357 50245 97.83 97.83 97.83 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13165 12574 95.51 94.56 95.03 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.51 94.56 95.03 Avg2. 13297 13165 12574 95.51 94.56 95.03 Current max chunk-based F1: 95.03 (iteration 60) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 61 Log-likelihood = -99520.979813 Norm (log-likelihood gradient vector) = 4260.061212 Norm (lambda vector) = 74.058280 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13167 12846 97.56 96.61 97.08 i-np 15046 15140 14706 97.13 97.74 97.44 o 23014 23050 22696 98.46 98.62 98.54 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.72 97.66 97.69 Avg2. 51357 51357 50248 97.84 97.84 97.84 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13167 12579 95.53 94.60 95.06 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.53 94.60 95.06 Avg2. 13297 13167 12579 95.53 94.60 95.06 Current max chunk-based F1: 95.06 (iteration 61) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 62 Log-likelihood = -96656.031739 Norm (log-likelihood gradient vector) = 4842.317302 Norm (lambda vector) = 74.729784 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13167 12843 97.54 96.59 97.06 i-np 15046 15139 14708 97.15 97.75 97.45 o 23014 23051 22698 98.47 98.63 98.55 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.72 97.66 97.69 Avg2. 51357 51357 50249 97.84 97.84 97.84 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13167 12579 95.53 94.60 95.06 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.53 94.60 95.06 Avg2. 13297 13167 12579 95.53 94.60 95.06 Current max chunk-based F1: 95.06 (iteration 61) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 63 Log-likelihood = -92238.065657 Norm (log-likelihood gradient vector) = 4217.252344 Norm (lambda vector) = 75.689156 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13223 12867 97.31 96.77 97.04 i-np 15046 15046 14660 97.43 97.43 97.43 o 23014 23088 22715 98.38 98.70 98.54 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.71 97.63 97.67 Avg2. 51357 51357 50242 97.83 97.83 97.83 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13223 12600 95.29 94.76 95.02 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.29 94.76 95.02 Avg2. 13297 13223 12600 95.29 94.76 95.02 Current max chunk-based F1: 95.06 (iteration 61) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 64 Log-likelihood = -90208.907106 Norm (log-likelihood gradient vector) = 17603.106507 Norm (lambda vector) = 76.470527 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13192 12858 97.47 96.70 97.08 i-np 15046 15094 14691 97.33 97.64 97.49 o 23014 23071 22710 98.44 98.68 98.56 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.74 97.67 97.71 Avg2. 51357 51357 50259 97.86 97.86 97.86 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13192 12601 95.52 94.77 95.14 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.52 94.77 95.14 Avg2. 13297 13192 12601 95.52 94.77 95.14 Current max chunk-based F1: 95.14 (iteration 64) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 65 Log-likelihood = -90232.207857 Norm (log-likelihood gradient vector) = 9923.767932 Norm (lambda vector) = 76.084318 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13169 12841 97.51 96.57 97.04 i-np 15046 15148 14711 97.12 97.77 97.44 o 23014 23040 22692 98.49 98.60 98.55 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.70 97.65 97.68 Avg2. 51357 51357 50244 97.83 97.83 97.83 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13169 12574 95.48 94.56 95.02 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.48 94.56 95.02 Avg2. 13297 13169 12574 95.48 94.56 95.02 Current max chunk-based F1: 95.14 (iteration 64) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 66 Log-likelihood = -87622.493409 Norm (log-likelihood gradient vector) = 3570.422804 Norm (lambda vector) = 76.696972 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13158 12841 97.59 96.57 97.08 i-np 15046 15168 14724 97.07 97.86 97.46 o 23014 23031 22692 98.53 98.60 98.56 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.73 97.68 97.70 Avg2. 51357 51357 50257 97.86 97.86 97.86 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13158 12578 95.59 94.59 95.09 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.59 94.59 95.09 Avg2. 13297 13158 12578 95.59 94.59 95.09 Current max chunk-based F1: 95.14 (iteration 64) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 67 Log-likelihood = -87497.053185 Norm (log-likelihood gradient vector) = 3410.098788 Norm (lambda vector) = 76.636733 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13150 12841 97.65 96.57 97.11 i-np 15046 15176 14731 97.07 97.91 97.49 o 23014 23031 22694 98.54 98.61 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.75 97.70 97.72 Avg2. 51357 51357 50266 97.88 97.88 97.88 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13150 12577 95.64 94.59 95.11 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.64 94.59 95.11 Avg2. 13297 13150 12577 95.64 94.59 95.11 Current max chunk-based F1: 95.14 (iteration 64) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 68 Log-likelihood = -87523.632002 Norm (log-likelihood gradient vector) = 3968.901247 Norm (lambda vector) = 76.640571 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13153 12841 97.63 96.57 97.10 i-np 15046 15178 14732 97.06 97.91 97.49 o 23014 23026 22692 98.55 98.60 98.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.75 97.69 97.72 Avg2. 51357 51357 50265 97.87 97.87 97.87 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13153 12578 95.63 94.59 95.11 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.63 94.59 95.11 Avg2. 13297 13153 12578 95.63 94.59 95.11 Current max chunk-based F1: 95.14 (iteration 64) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 69 Log-likelihood = -87443.852602 Norm (log-likelihood gradient vector) = 3420.872376 Norm (lambda vector) = 76.636809 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13089 12809 97.86 96.33 97.09 i-np 15046 15314 14779 96.51 98.23 97.36 o 23014 22954 22653 98.69 98.43 98.56 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.69 97.66 97.67 Avg2. 51357 51357 50241 97.83 97.83 97.83 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13089 12526 95.70 94.20 94.94 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.70 94.20 94.94 Avg2. 13297 13089 12526 95.70 94.20 94.94 Current max chunk-based F1: 95.14 (iteration 64) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 70 Log-likelihood = -86489.621588 Norm (log-likelihood gradient vector) = 9803.220996 Norm (lambda vector) = 77.351121 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13230 12897 97.48 96.99 97.24 i-np 15046 15027 14685 97.72 97.60 97.66 o 23014 23100 22751 98.49 98.86 98.67 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.90 97.82 97.86 Avg2. 51357 51357 50333 98.01 98.01 98.01 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13230 12649 95.61 95.13 95.37 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.61 95.13 95.37 Avg2. 13297 13230 12649 95.61 95.13 95.37 Current max chunk-based F1: 95.37 (iteration 70) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 71 Log-likelihood = -87280.067600 Norm (log-likelihood gradient vector) = 12901.150140 Norm (lambda vector) = 78.292120 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13147 12846 97.71 96.61 97.16 i-np 15046 15198 14741 96.99 97.97 97.48 o 23014 23012 22692 98.61 98.60 98.61 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.77 97.73 97.75 Avg2. 51357 51357 50279 97.90 97.90 97.90 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13147 12577 95.66 94.59 95.12 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.66 94.59 95.12 Avg2. 13297 13147 12577 95.66 94.59 95.12 Current max chunk-based F1: 95.37 (iteration 70) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 72 Log-likelihood = -85913.103121 Norm (log-likelihood gradient vector) = 4477.000587 Norm (lambda vector) = 77.692130 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13162 12858 97.69 96.70 97.19 i-np 15046 15174 14734 97.10 97.93 97.51 o 23014 23021 22700 98.61 98.64 98.62 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.80 97.75 97.78 Avg2. 51357 51357 50292 97.93 97.93 97.93 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13162 12593 95.68 94.71 95.19 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.68 94.71 95.19 Avg2. 13297 13162 12593 95.68 94.71 95.19 Current max chunk-based F1: 95.37 (iteration 70) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 73 Log-likelihood = -85009.034798 Norm (log-likelihood gradient vector) = 3400.132220 Norm (lambda vector) = 78.125872 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13176 12863 97.62 96.74 97.18 i-np 15046 15141 14721 97.23 97.84 97.53 o 23014 23040 22711 98.57 98.68 98.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.81 97.75 97.78 Avg2. 51357 51357 50295 97.93 97.93 97.93 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13176 12602 95.64 94.77 95.21 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.64 94.77 95.21 Avg2. 13297 13176 12602 95.64 94.77 95.21 Current max chunk-based F1: 95.37 (iteration 70) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 74 Log-likelihood = -83000.896299 Norm (log-likelihood gradient vector) = 4029.285434 Norm (lambda vector) = 79.034902 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13180 12867 97.63 96.77 97.19 i-np 15046 15140 14724 97.25 97.86 97.56 o 23014 23037 22712 98.59 98.69 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.82 97.77 97.80 Avg2. 51357 51357 50303 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13180 12608 95.66 94.82 95.24 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.66 94.82 95.24 Avg2. 13297 13180 12608 95.66 94.82 95.24 Current max chunk-based F1: 95.37 (iteration 70) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 75 Log-likelihood = -81829.909792 Norm (log-likelihood gradient vector) = 3610.968514 Norm (lambda vector) = 79.533178 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13209 12874 97.46 96.82 97.14 i-np 15046 15070 14674 97.37 97.53 97.45 o 23014 23078 22724 98.47 98.74 98.60 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.77 97.70 97.73 Avg2. 51357 51357 50272 97.89 97.89 97.89 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13209 12614 95.50 94.86 95.18 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.50 94.86 95.18 Avg2. 13297 13209 12614 95.50 94.86 95.18 Current max chunk-based F1: 95.37 (iteration 70) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 76 Log-likelihood = -82443.743505 Norm (log-likelihood gradient vector) = 23169.837782 Norm (lambda vector) = 81.226506 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13196 12881 97.61 96.87 97.24 i-np 15046 15109 14714 97.39 97.79 97.59 o 23014 23052 22723 98.57 98.74 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.80 97.83 Avg2. 51357 51357 50318 97.98 97.98 97.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13196 12629 95.70 94.98 95.34 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.70 94.98 95.34 Avg2. 13297 13196 12629 95.70 94.98 95.34 Current max chunk-based F1: 95.37 (iteration 70) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 77 Log-likelihood = -80904.053714 Norm (log-likelihood gradient vector) = 10909.612575 Norm (lambda vector) = 80.278749 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13178 12869 97.66 96.78 97.22 i-np 15046 15139 14722 97.25 97.85 97.55 o 23014 23040 22712 98.58 98.69 98.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.83 97.77 97.80 Avg2. 51357 51357 50303 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13178 12617 95.74 94.89 95.31 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.74 94.89 95.31 Avg2. 13297 13178 12617 95.74 94.89 95.31 Current max chunk-based F1: 95.37 (iteration 70) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 78 Log-likelihood = -78969.434055 Norm (log-likelihood gradient vector) = 4490.659142 Norm (lambda vector) = 81.158244 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13174 12866 97.66 96.76 97.21 i-np 15046 15147 14728 97.23 97.89 97.56 o 23014 23036 22710 98.58 98.68 98.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.83 97.77 97.80 Avg2. 51357 51357 50304 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13174 12618 95.78 94.89 95.33 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.78 94.89 95.33 Avg2. 13297 13174 12618 95.78 94.89 95.33 Current max chunk-based F1: 95.37 (iteration 70) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 79 Log-likelihood = -78378.363126 Norm (log-likelihood gradient vector) = 3005.839888 Norm (lambda vector) = 81.494737 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13172 12867 97.68 96.77 97.22 i-np 15046 15153 14736 97.25 97.94 97.59 o 23014 23032 22713 98.61 98.69 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.85 97.80 97.82 Avg2. 51357 51357 50316 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13172 12621 95.82 94.92 95.36 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.82 94.92 95.36 Avg2. 13297 13172 12621 95.82 94.92 95.36 Current max chunk-based F1: 95.37 (iteration 70) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 80 Log-likelihood = -77593.505357 Norm (log-likelihood gradient vector) = 3307.211825 Norm (lambda vector) = 82.035650 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13177 12868 97.66 96.77 97.21 i-np 15046 15135 14726 97.30 97.87 97.58 o 23014 23045 22715 98.57 98.70 98.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.84 97.78 97.81 Avg2. 51357 51357 50309 97.96 97.96 97.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13177 12626 95.82 94.95 95.38 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.82 94.95 95.38 Avg2. 13297 13177 12626 95.82 94.95 95.38 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 81 Log-likelihood = -76154.206724 Norm (log-likelihood gradient vector) = 3490.954544 Norm (lambda vector) = 82.952624 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13077 12809 97.95 96.33 97.13 i-np 15046 15358 14806 96.41 98.40 97.40 o 23014 22922 22647 98.80 98.41 98.60 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.72 97.71 97.72 Avg2. 51357 51357 50262 97.87 97.87 97.87 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13077 12533 95.84 94.25 95.04 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.84 94.25 95.04 Avg2. 13297 13077 12533 95.84 94.25 95.04 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 82 Log-likelihood = -74316.761123 Norm (log-likelihood gradient vector) = 14378.130620 Norm (lambda vector) = 84.694934 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13150 12853 97.74 96.66 97.20 i-np 15046 15203 14753 97.04 98.05 97.54 o 23014 23004 22697 98.67 98.62 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.82 97.78 97.80 Avg2. 51357 51357 50303 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13150 12601 95.83 94.77 95.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.83 94.77 95.29 Avg2. 13297 13150 12601 95.83 94.77 95.29 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 83 Log-likelihood = -72509.666868 Norm (log-likelihood gradient vector) = 3651.777124 Norm (lambda vector) = 85.248950 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13165 12865 97.72 96.75 97.23 i-np 15046 15166 14741 97.20 97.97 97.58 o 23014 23026 22710 98.63 98.68 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.85 97.80 97.82 Avg2. 51357 51357 50316 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13165 12617 95.84 94.89 95.36 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.84 94.89 95.36 Avg2. 13297 13165 12617 95.84 94.89 95.36 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 84 Log-likelihood = -71790.034882 Norm (log-likelihood gradient vector) = 2703.018878 Norm (lambda vector) = 85.328445 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13181 12874 97.67 96.82 97.24 i-np 15046 15131 14722 97.30 97.85 97.57 o 23014 23045 22720 98.59 98.72 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.85 97.80 97.82 Avg2. 51357 51357 50316 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13181 12622 95.76 94.92 95.34 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.76 94.92 95.34 Avg2. 13297 13181 12622 95.76 94.92 95.34 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 85 Log-likelihood = -70337.103607 Norm (log-likelihood gradient vector) = 3286.514246 Norm (lambda vector) = 86.003700 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13184 12877 97.67 96.84 97.25 i-np 15046 15126 14722 97.33 97.85 97.59 o 23014 23047 22722 98.59 98.73 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.81 97.83 Avg2. 51357 51357 50321 97.98 97.98 97.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13184 12627 95.78 94.96 95.37 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.78 94.96 95.37 Avg2. 13297 13184 12627 95.78 94.96 95.37 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 86 Log-likelihood = -68883.031546 Norm (log-likelihood gradient vector) = 2775.082650 Norm (lambda vector) = 86.845379 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13190 12872 97.59 96.80 97.19 i-np 15046 15115 14708 97.31 97.75 97.53 o 23014 23052 22718 98.55 98.71 98.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.82 97.76 97.79 Avg2. 51357 51357 50298 97.94 97.94 97.94 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13190 12616 95.65 94.88 95.26 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.65 94.88 95.26 Avg2. 13297 13190 12616 95.65 94.88 95.26 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 87 Log-likelihood = -69716.209777 Norm (log-likelihood gradient vector) = 20646.186231 Norm (lambda vector) = 89.863898 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13187 12873 97.62 96.81 97.21 i-np 15046 15118 14712 97.31 97.78 97.55 o 23014 23052 22720 98.56 98.72 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.83 97.77 97.80 Avg2. 51357 51357 50305 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13187 12620 95.70 94.91 95.30 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.70 94.91 95.30 Avg2. 13297 13187 12620 95.70 94.91 95.30 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 88 Log-likelihood = -68026.878814 Norm (log-likelihood gradient vector) = 8906.605809 Norm (lambda vector) = 88.132344 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13189 12873 97.60 96.81 97.21 i-np 15046 15122 14714 97.30 97.79 97.55 o 23014 23046 22717 98.57 98.71 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.83 97.77 97.80 Avg2. 51357 51357 50304 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13189 12619 95.68 94.90 95.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.68 94.90 95.29 Avg2. 13297 13189 12619 95.68 94.90 95.29 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 89 Log-likelihood = -66669.872752 Norm (log-likelihood gradient vector) = 3018.549833 Norm (lambda vector) = 88.965251 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13183 12869 97.62 96.78 97.20 i-np 15046 15133 14722 97.28 97.85 97.56 o 23014 23041 22714 98.58 98.70 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.83 97.77 97.80 Avg2. 51357 51357 50305 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13183 12620 95.73 94.91 95.32 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.73 94.91 95.32 Avg2. 13297 13183 12620 95.73 94.91 95.32 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 90 Log-likelihood = -66304.123146 Norm (log-likelihood gradient vector) = 2541.144000 Norm (lambda vector) = 89.290642 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13180 12868 97.63 96.77 97.20 i-np 15046 15140 14726 97.27 97.87 97.57 o 23014 23037 22715 98.60 98.70 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.83 97.78 97.81 Avg2. 51357 51357 50309 97.96 97.96 97.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13180 12620 95.75 94.91 95.33 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.75 94.91 95.33 Avg2. 13297 13180 12620 95.75 94.91 95.33 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 91 Log-likelihood = -65375.689669 Norm (log-likelihood gradient vector) = 3303.887093 Norm (lambda vector) = 90.074185 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13152 12854 97.73 96.67 97.20 i-np 15046 15189 14748 97.10 98.02 97.56 o 23014 23016 22704 98.64 98.65 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.83 97.78 97.80 Avg2. 51357 51357 50306 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13152 12601 95.81 94.77 95.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.81 94.77 95.29 Avg2. 13297 13152 12601 95.81 94.77 95.29 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 92 Log-likelihood = -63610.096557 Norm (log-likelihood gradient vector) = 4641.332593 Norm (lambda vector) = 91.273932 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13254 12898 97.31 97.00 97.16 i-np 15046 14963 14629 97.77 97.23 97.50 o 23014 23140 22755 98.34 98.87 98.60 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.81 97.70 97.75 Avg2. 51357 51357 50282 97.91 97.91 97.91 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13254 12640 95.37 95.06 95.21 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.37 95.06 95.21 Avg2. 13297 13254 12640 95.37 95.06 95.21 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 93 Log-likelihood = -62169.823654 Norm (log-likelihood gradient vector) = 13746.484513 Norm (lambda vector) = 94.020256 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13191 12874 97.60 96.82 97.21 i-np 15046 15105 14707 97.37 97.75 97.56 o 23014 23061 22726 98.55 98.75 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.84 97.77 97.80 Avg2. 51357 51357 50307 97.96 97.96 97.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13191 12618 95.66 94.89 95.27 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.66 94.89 95.27 Avg2. 13297 13191 12618 95.66 94.89 95.27 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 94 Log-likelihood = -62328.547194 Norm (log-likelihood gradient vector) = 5238.716034 Norm (lambda vector) = 92.372851 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13167 12864 97.70 96.74 97.22 i-np 15046 15153 14730 97.21 97.90 97.55 o 23014 23037 22715 98.60 98.70 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.84 97.78 97.81 Avg2. 51357 51357 50309 97.96 97.96 97.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13167 12608 95.75 94.82 95.28 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.75 94.82 95.28 Avg2. 13297 13167 12608 95.75 94.82 95.28 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 95 Log-likelihood = -60878.414164 Norm (log-likelihood gradient vector) = 2813.385030 Norm (lambda vector) = 93.006782 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13159 12855 97.69 96.68 97.18 i-np 15046 15170 14728 97.09 97.89 97.48 o 23014 23028 22703 98.59 98.65 98.62 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.79 97.74 97.76 Avg2. 51357 51357 50286 97.91 97.91 97.91 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13159 12591 95.68 94.69 95.18 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.68 94.69 95.18 Avg2. 13297 13159 12591 95.68 94.69 95.18 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 96 Log-likelihood = -59891.587094 Norm (log-likelihood gradient vector) = 2352.151091 Norm (lambda vector) = 93.273510 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13154 12856 97.73 96.68 97.21 i-np 15046 15183 14741 97.09 97.97 97.53 o 23014 23020 22703 98.62 98.65 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.82 97.77 97.79 Avg2. 51357 51357 50300 97.94 97.94 97.94 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13154 12597 95.77 94.74 95.25 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.77 94.74 95.25 Avg2. 13297 13154 12597 95.77 94.74 95.25 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 97 Log-likelihood = -58572.040182 Norm (log-likelihood gradient vector) = 2895.296618 Norm (lambda vector) = 93.809268 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13165 12865 97.72 96.75 97.23 i-np 15046 15176 14742 97.14 97.98 97.56 o 23014 23016 22705 98.65 98.66 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.84 97.80 97.82 Avg2. 51357 51357 50312 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13165 12606 95.75 94.80 95.28 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.75 94.80 95.28 Avg2. 13297 13165 12606 95.75 94.80 95.28 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 98 Log-likelihood = -56701.087634 Norm (log-likelihood gradient vector) = 2371.733052 Norm (lambda vector) = 94.834156 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13192 12878 97.62 96.85 97.23 i-np 15046 15120 14712 97.30 97.78 97.54 o 23014 23045 22718 98.58 98.71 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.83 97.78 97.81 Avg2. 51357 51357 50308 97.96 97.96 97.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13192 12621 95.67 94.92 95.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.67 94.92 95.29 Avg2. 13297 13192 12621 95.67 94.92 95.29 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 99 Log-likelihood = -55856.864648 Norm (log-likelihood gradient vector) = 11170.090257 Norm (lambda vector) = 96.716810 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13178 12864 97.62 96.74 97.18 i-np 15046 15152 14721 97.16 97.84 97.50 o 23014 23027 22706 98.61 98.66 98.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.79 97.75 97.77 Avg2. 51357 51357 50291 97.92 97.92 97.92 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13178 12602 95.63 94.77 95.20 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.63 94.77 95.20 Avg2. 13297 13178 12602 95.63 94.77 95.20 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 100 Log-likelihood = -55859.557483 Norm (log-likelihood gradient vector) = 4761.268510 Norm (lambda vector) = 95.606094 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13187 12875 97.63 96.83 97.23 i-np 15046 15141 14724 97.25 97.86 97.55 o 23014 23029 22711 98.62 98.68 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.83 97.79 97.81 Avg2. 51357 51357 50310 97.96 97.96 97.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13187 12620 95.70 94.91 95.30 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.70 94.91 95.30 Avg2. 13297 13187 12620 95.70 94.91 95.30 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 101 Log-likelihood = -54620.079790 Norm (log-likelihood gradient vector) = 2761.474015 Norm (lambda vector) = 96.626089 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13189 12877 97.63 96.84 97.24 i-np 15046 15132 14717 97.26 97.81 97.53 o 23014 23036 22714 98.60 98.70 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.83 97.78 97.81 Avg2. 51357 51357 50308 97.96 97.96 97.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13189 12618 95.67 94.89 95.28 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.67 94.89 95.28 Avg2. 13297 13189 12618 95.67 94.89 95.28 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 102 Log-likelihood = -53776.550072 Norm (log-likelihood gradient vector) = 2151.461088 Norm (lambda vector) = 97.386192 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13195 12882 97.63 96.88 97.25 i-np 15046 15115 14715 97.35 97.80 97.58 o 23014 23047 22720 98.58 98.72 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.85 97.80 97.83 Avg2. 51357 51357 50317 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13195 12629 95.71 94.98 95.34 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.71 94.98 95.34 Avg2. 13297 13195 12629 95.71 94.98 95.34 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 103 Log-likelihood = -52941.463449 Norm (log-likelihood gradient vector) = 3000.233833 Norm (lambda vector) = 98.333087 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13146 12855 97.79 96.68 97.23 i-np 15046 15227 14767 96.98 98.15 97.56 o 23014 22984 22688 98.71 98.58 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.83 97.80 97.81 Avg2. 51357 51357 50310 97.96 97.96 97.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13146 12599 95.84 94.75 95.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.84 94.75 95.29 Avg2. 13297 13146 12599 95.84 94.75 95.29 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 104 Log-likelihood = -52439.838352 Norm (log-likelihood gradient vector) = 7315.402538 Norm (lambda vector) = 99.816849 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13182 12874 97.66 96.82 97.24 i-np 15046 15135 14727 97.30 97.88 97.59 o 23014 23040 22717 98.60 98.71 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.80 97.83 Avg2. 51357 51357 50318 97.98 97.98 97.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13182 12624 95.77 94.94 95.35 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.77 94.94 95.35 Avg2. 13297 13182 12624 95.77 94.94 95.35 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 105 Log-likelihood = -51905.555922 Norm (log-likelihood gradient vector) = 3015.625484 Norm (lambda vector) = 100.124520 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13196 12884 97.64 96.89 97.26 i-np 15046 15106 14717 97.42 97.81 97.62 o 23014 23055 22728 98.58 98.76 98.67 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.88 97.82 97.85 Avg2. 51357 51357 50329 98.00 98.00 98.00 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13196 12635 95.75 95.02 95.38 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.75 95.02 95.38 Avg2. 13297 13196 12635 95.75 95.02 95.38 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 106 Log-likelihood = -51644.493820 Norm (log-likelihood gradient vector) = 2184.364642 Norm (lambda vector) = 100.249522 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13194 12885 97.66 96.90 97.28 i-np 15046 15109 14718 97.41 97.82 97.62 o 23014 23054 22729 98.59 98.76 98.68 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.89 97.83 97.86 Avg2. 51357 51357 50332 98.00 98.00 98.00 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13194 12633 95.75 95.01 95.38 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.75 95.01 95.38 Avg2. 13297 13194 12633 95.75 95.01 95.38 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 107 Log-likelihood = -51383.638233 Norm (log-likelihood gradient vector) = 2619.007269 Norm (lambda vector) = 100.727421 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13186 12880 97.68 96.86 97.27 i-np 15046 15125 14722 97.34 97.85 97.59 o 23014 23046 22723 98.60 98.74 98.67 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.87 97.82 97.84 Avg2. 51357 51357 50325 97.99 97.99 97.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13186 12621 95.72 94.92 95.31 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.72 94.92 95.31 Avg2. 13297 13186 12621 95.72 94.92 95.31 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 108 Log-likelihood = -51046.562758 Norm (log-likelihood gradient vector) = 2299.493466 Norm (lambda vector) = 101.504826 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13190 12876 97.62 96.83 97.23 i-np 15046 15123 14716 97.31 97.81 97.56 o 23014 23044 22720 98.59 98.72 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.84 97.79 97.81 Avg2. 51357 51357 50312 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13190 12613 95.63 94.86 95.24 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.63 94.86 95.24 Avg2. 13297 13190 12613 95.63 94.86 95.24 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 109 Log-likelihood = -50735.237554 Norm (log-likelihood gradient vector) = 9373.186626 Norm (lambda vector) = 104.642618 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13189 12882 97.67 96.88 97.27 i-np 15046 15117 14718 97.36 97.82 97.59 o 23014 23051 22728 98.60 98.76 98.68 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.88 97.82 97.85 Avg2. 51357 51357 50328 98.00 98.00 98.00 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13189 12621 95.69 94.92 95.30 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.69 94.92 95.30 Avg2. 13297 13189 12621 95.69 94.92 95.30 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 110 Log-likelihood = -50486.520965 Norm (log-likelihood gradient vector) = 4998.339820 Norm (lambda vector) = 102.996312 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13197 12879 97.59 96.86 97.22 i-np 15046 15117 14706 97.28 97.74 97.51 o 23014 23043 22718 98.59 98.71 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.82 97.77 97.80 Avg2. 51357 51357 50303 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13197 12610 95.55 94.83 95.19 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.55 94.83 95.19 Avg2. 13297 13197 12610 95.55 94.83 95.19 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 111 Log-likelihood = -49985.879669 Norm (log-likelihood gradient vector) = 2541.491394 Norm (lambda vector) = 103.852871 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13193 12880 97.63 96.86 97.24 i-np 15046 15128 14715 97.27 97.80 97.53 o 23014 23036 22717 98.62 98.71 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.84 97.79 97.81 Avg2. 51357 51357 50312 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13193 12614 95.61 94.86 95.24 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.61 94.86 95.24 Avg2. 13297 13193 12614 95.61 94.86 95.24 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 112 Log-likelihood = -49577.280176 Norm (log-likelihood gradient vector) = 1942.465342 Norm (lambda vector) = 104.351319 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13187 12880 97.67 96.86 97.27 i-np 15046 15133 14716 97.24 97.81 97.52 o 23014 23037 22717 98.61 98.71 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.84 97.79 97.82 Avg2. 51357 51357 50313 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13187 12613 95.65 94.86 95.25 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.65 94.86 95.25 Avg2. 13297 13187 12613 95.65 94.86 95.25 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 113 Log-likelihood = -49026.628676 Norm (log-likelihood gradient vector) = 2603.048959 Norm (lambda vector) = 105.123876 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13189 12881 97.66 96.87 97.27 i-np 15046 15121 14710 97.28 97.77 97.52 o 23014 23047 22722 98.59 98.73 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.85 97.79 97.82 Avg2. 51357 51357 50313 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13189 12617 95.66 94.89 95.27 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.66 94.89 95.27 Avg2. 13297 13189 12617 95.66 94.89 95.27 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 114 Log-likelihood = -48240.607875 Norm (log-likelihood gradient vector) = 3088.669026 Norm (lambda vector) = 106.423850 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13063 12814 98.09 96.37 97.22 i-np 15046 15389 14820 96.30 98.50 97.39 o 23014 22905 22640 98.84 98.37 98.61 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.75 97.75 97.75 Avg2. 51357 51357 50274 97.89 97.89 97.89 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13063 12534 95.95 94.26 95.10 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.95 94.26 95.10 Avg2. 13297 13063 12534 95.95 94.26 95.10 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 115 Log-likelihood = -48875.586771 Norm (log-likelihood gradient vector) = 16748.902929 Norm (lambda vector) = 108.293344 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13138 12858 97.87 96.70 97.28 i-np 15046 15223 14755 96.93 98.07 97.49 o 23014 22996 22693 98.68 98.61 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.83 97.79 97.81 Avg2. 51357 51357 50306 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13138 12594 95.86 94.71 95.28 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.86 94.71 95.28 Avg2. 13297 13138 12594 95.86 94.71 95.28 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 116 Log-likelihood = -47828.232029 Norm (log-likelihood gradient vector) = 7192.937192 Norm (lambda vector) = 107.183128 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13170 12871 97.73 96.80 97.26 i-np 15046 15158 14721 97.12 97.84 97.48 o 23014 23029 22708 98.61 98.67 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.82 97.77 97.79 Avg2. 51357 51357 50300 97.94 97.94 97.94 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13170 12601 95.68 94.77 95.22 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.68 94.77 95.22 Avg2. 13297 13170 12601 95.68 94.77 95.22 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 117 Log-likelihood = -47169.331989 Norm (log-likelihood gradient vector) = 3155.088198 Norm (lambda vector) = 107.999519 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13187 12886 97.72 96.91 97.31 i-np 15046 15114 14708 97.31 97.75 97.53 o 23014 23056 22725 98.56 98.74 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.87 97.80 97.83 Avg2. 51357 51357 50319 97.98 97.98 97.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13187 12622 95.72 94.92 95.32 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.72 94.92 95.32 Avg2. 13297 13187 12622 95.72 94.92 95.32 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 118 Log-likelihood = -46769.345872 Norm (log-likelihood gradient vector) = 1977.540356 Norm (lambda vector) = 108.491182 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13183 12882 97.72 96.88 97.30 i-np 15046 15118 14708 97.29 97.75 97.52 o 23014 23056 22723 98.56 98.74 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.85 97.79 97.82 Avg2. 51357 51357 50313 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13183 12616 95.70 94.88 95.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.70 94.88 95.29 Avg2. 13297 13183 12616 95.70 94.88 95.29 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 119 Log-likelihood = -46507.956174 Norm (log-likelihood gradient vector) = 2259.348214 Norm (lambda vector) = 108.808255 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13180 12876 97.69 96.83 97.26 i-np 15046 15119 14707 97.27 97.75 97.51 o 23014 23058 22723 98.55 98.74 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.84 97.77 97.81 Avg2. 51357 51357 50306 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13180 12609 95.67 94.83 95.24 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.67 94.83 95.24 Avg2. 13297 13180 12609 95.67 94.83 95.24 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 120 Log-likelihood = -46112.731697 Norm (log-likelihood gradient vector) = 2294.008493 Norm (lambda vector) = 109.253383 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13194 12881 97.63 96.87 97.25 i-np 15046 15137 14713 97.20 97.79 97.49 o 23014 23026 22706 98.61 98.66 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.81 97.77 97.79 Avg2. 51357 51357 50300 97.94 97.94 97.94 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13194 12615 95.61 94.87 95.24 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.61 94.87 95.24 Avg2. 13297 13194 12615 95.61 94.87 95.24 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 121 Log-likelihood = -46468.763440 Norm (log-likelihood gradient vector) = 12382.094680 Norm (lambda vector) = 110.940293 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13182 12878 97.69 96.85 97.27 i-np 15046 15139 14717 97.21 97.81 97.51 o 23014 23036 22713 98.60 98.69 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.83 97.78 97.81 Avg2. 51357 51357 50308 97.96 97.96 97.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13182 12613 95.68 94.86 95.27 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.68 94.86 95.27 Avg2. 13297 13182 12613 95.68 94.86 95.27 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 122 Log-likelihood = -45730.236435 Norm (log-likelihood gradient vector) = 5788.082395 Norm (lambda vector) = 109.999158 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13177 12875 97.71 96.83 97.27 i-np 15046 15143 14719 97.20 97.83 97.51 o 23014 23037 22712 98.59 98.69 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.83 97.78 97.81 Avg2. 51357 51357 50306 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13177 12613 95.72 94.86 95.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.72 94.86 95.29 Avg2. 13297 13177 12613 95.72 94.86 95.29 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 123 Log-likelihood = -45187.124762 Norm (log-likelihood gradient vector) = 2327.108198 Norm (lambda vector) = 110.351086 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13173 12874 97.73 96.82 97.27 i-np 15046 15157 14725 97.15 97.87 97.51 o 23014 23027 22708 98.61 98.67 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.83 97.79 97.81 Avg2. 51357 51357 50307 97.96 97.96 97.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13173 12611 95.73 94.84 95.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.73 94.84 95.29 Avg2. 13297 13173 12611 95.73 94.84 95.29 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 124 Log-likelihood = -44785.094994 Norm (log-likelihood gradient vector) = 1547.627900 Norm (lambda vector) = 110.792265 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13175 12877 97.74 96.84 97.29 i-np 15046 15159 14726 97.14 97.87 97.51 o 23014 23023 22706 98.62 98.66 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.83 97.79 97.81 Avg2. 51357 51357 50309 97.96 97.96 97.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13175 12614 95.74 94.86 95.30 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.74 94.86 95.30 Avg2. 13297 13175 12614 95.74 94.86 95.30 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 125 Log-likelihood = -44357.826227 Norm (log-likelihood gradient vector) = 1874.389961 Norm (lambda vector) = 111.291486 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13178 12875 97.70 96.83 97.26 i-np 15046 15158 14724 97.14 97.86 97.50 o 23014 23021 22702 98.61 98.64 98.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.82 97.78 97.80 Avg2. 51357 51357 50301 97.94 97.94 97.94 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13178 12611 95.70 94.84 95.27 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.70 94.84 95.27 Avg2. 13297 13178 12611 95.70 94.84 95.27 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 126 Log-likelihood = -43516.870953 Norm (log-likelihood gradient vector) = 2205.668439 Norm (lambda vector) = 112.240754 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13189 12882 97.67 96.88 97.27 i-np 15046 15125 14711 97.26 97.77 97.52 o 23014 23043 22714 98.57 98.70 98.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.84 97.78 97.81 Avg2. 51357 51357 50307 97.96 97.96 97.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13189 12626 95.73 94.95 95.34 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.73 94.95 95.34 Avg2. 13297 13189 12626 95.73 94.95 95.34 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 127 Log-likelihood = -42175.555014 Norm (log-likelihood gradient vector) = 2445.513822 Norm (lambda vector) = 113.889256 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13062 12797 97.97 96.24 97.10 i-np 15046 15427 14820 96.07 98.50 97.27 o 23014 22868 22610 98.87 98.24 98.56 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.64 97.66 97.65 Avg2. 51357 51357 50227 97.80 97.80 97.80 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13062 12501 95.71 94.01 94.85 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.71 94.01 94.85 Avg2. 13297 13062 12501 95.71 94.01 94.85 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 128 Log-likelihood = -43333.587069 Norm (log-likelihood gradient vector) = 20006.988917 Norm (lambda vector) = 117.374571 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13136 12853 97.85 96.66 97.25 i-np 15046 15262 14773 96.80 98.19 97.49 o 23014 22959 22671 98.75 98.51 98.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.80 97.79 97.79 Avg2. 51357 51357 50297 97.94 97.94 97.94 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13136 12595 95.88 94.72 95.30 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.88 94.72 95.30 Avg2. 13297 13136 12595 95.88 94.72 95.30 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 129 Log-likelihood = -41483.512498 Norm (log-likelihood gradient vector) = 7273.414758 Norm (lambda vector) = 115.282190 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13186 12882 97.69 96.88 97.29 i-np 15046 15143 14726 97.25 97.87 97.56 o 23014 23028 22712 98.63 98.69 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.81 97.83 Avg2. 51357 51357 50320 97.98 97.98 97.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13186 12629 95.78 94.98 95.37 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.78 94.98 95.37 Avg2. 13297 13186 12629 95.78 94.98 95.37 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 130 Log-likelihood = -40934.747937 Norm (log-likelihood gradient vector) = 1785.945035 Norm (lambda vector) = 115.491310 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13191 12885 97.68 96.90 97.29 i-np 15046 15129 14719 97.29 97.83 97.56 o 23014 23037 22716 98.61 98.71 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.81 97.84 Avg2. 51357 51357 50320 97.98 97.98 97.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13191 12628 95.73 94.97 95.35 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.73 94.97 95.35 Avg2. 13297 13191 12628 95.73 94.97 95.35 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 131 Log-likelihood = -40892.115567 Norm (log-likelihood gradient vector) = 1359.088519 Norm (lambda vector) = 115.453036 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13195 12887 97.67 96.92 97.29 i-np 15046 15116 14713 97.33 97.79 97.56 o 23014 23046 22720 98.59 98.72 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.81 97.84 Avg2. 51357 51357 50320 97.98 97.98 97.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13195 12626 95.69 94.95 95.32 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.69 94.95 95.32 Avg2. 13297 13195 12626 95.69 94.95 95.32 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 132 Log-likelihood = -40822.404044 Norm (log-likelihood gradient vector) = 1443.410535 Norm (lambda vector) = 115.520286 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13192 12884 97.67 96.89 97.28 i-np 15046 15115 14713 97.34 97.79 97.56 o 23014 23050 22722 98.58 98.73 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.80 97.83 Avg2. 51357 51357 50319 97.98 97.98 97.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13192 12625 95.70 94.95 95.32 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.70 94.95 95.32 Avg2. 13297 13192 12625 95.70 94.95 95.32 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 133 Log-likelihood = -40513.214543 Norm (log-likelihood gradient vector) = 1757.952657 Norm (lambda vector) = 116.044606 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13222 12910 97.64 97.09 97.36 i-np 15046 15110 14715 97.39 97.80 97.59 o 23014 23025 22719 98.67 98.72 98.69 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.90 97.87 97.88 Avg2. 51357 51357 50344 98.03 98.03 98.03 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13222 12646 95.64 95.10 95.37 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.64 95.10 95.37 Avg2. 13297 13222 12646 95.64 95.10 95.37 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 134 Log-likelihood = -39712.700546 Norm (log-likelihood gradient vector) = 5626.021313 Norm (lambda vector) = 118.205482 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13210 12893 97.60 96.96 97.28 i-np 15046 15115 14705 97.29 97.73 97.51 o 23014 23032 22715 98.62 98.70 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.84 97.80 97.82 Avg2. 51357 51357 50313 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13210 12623 95.56 94.93 95.24 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.56 94.93 95.24 Avg2. 13297 13210 12623 95.56 94.93 95.24 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 135 Log-likelihood = -39896.651363 Norm (log-likelihood gradient vector) = 2785.530626 Norm (lambda vector) = 117.049627 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13204 12895 97.66 96.98 97.32 i-np 15046 15121 14714 97.31 97.79 97.55 o 23014 23032 22717 98.63 98.71 98.67 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.87 97.83 97.85 Avg2. 51357 51357 50326 97.99 97.99 97.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13204 12630 95.65 94.98 95.32 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.65 94.98 95.32 Avg2. 13297 13204 12630 95.65 94.98 95.32 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 136 Log-likelihood = -39337.555774 Norm (log-likelihood gradient vector) = 1851.969184 Norm (lambda vector) = 117.881450 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13181 12881 97.72 96.87 97.30 i-np 15046 15146 14726 97.23 97.87 97.55 o 23014 23030 22713 98.62 98.69 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.81 97.84 Avg2. 51357 51357 50320 97.98 97.98 97.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13181 12623 95.77 94.93 95.35 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.77 94.93 95.35 Avg2. 13297 13181 12623 95.77 94.93 95.35 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 137 Log-likelihood = -38709.268997 Norm (log-likelihood gradient vector) = 1472.855191 Norm (lambda vector) = 118.818462 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13177 12880 97.75 96.86 97.30 i-np 15046 15157 14731 97.19 97.91 97.55 o 23014 23023 22710 98.64 98.68 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.82 97.84 Avg2. 51357 51357 50321 97.98 97.98 97.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13177 12622 95.79 94.92 95.35 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.79 94.92 95.35 Avg2. 13297 13177 12622 95.79 94.92 95.35 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 138 Log-likelihood = -37636.241013 Norm (log-likelihood gradient vector) = 1995.629383 Norm (lambda vector) = 120.241531 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13178 12874 97.69 96.82 97.25 i-np 15046 15107 14695 97.27 97.67 97.47 o 23014 23072 22719 98.47 98.72 98.59 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.81 97.73 97.77 Avg2. 51357 51357 50288 97.92 97.92 97.92 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13178 12613 95.71 94.86 95.28 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.71 94.86 95.28 Avg2. 13297 13178 12613 95.71 94.86 95.28 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 139 Log-likelihood = -36225.654451 Norm (log-likelihood gradient vector) = 2790.508001 Norm (lambda vector) = 122.183921 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13158 12868 97.80 96.77 97.28 i-np 15046 15162 14730 97.15 97.90 97.52 o 23014 23037 22708 98.57 98.67 98.62 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.84 97.78 97.81 Avg2. 51357 51357 50306 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13158 12613 95.86 94.86 95.35 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.86 94.86 95.35 Avg2. 13297 13158 12613 95.86 94.86 95.35 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 140 Log-likelihood = -35221.358047 Norm (log-likelihood gradient vector) = 3255.416117 Norm (lambda vector) = 123.226628 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13181 12874 97.67 96.82 97.24 i-np 15046 15127 14705 97.21 97.73 97.47 o 23014 23049 22711 98.53 98.68 98.61 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.80 97.75 97.78 Avg2. 51357 51357 50290 97.92 97.92 97.92 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13181 12612 95.68 94.85 95.26 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.68 94.85 95.26 Avg2. 13297 13181 12612 95.68 94.85 95.26 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 141 Log-likelihood = -34793.183061 Norm (log-likelihood gradient vector) = 1327.094241 Norm (lambda vector) = 123.447848 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13181 12876 97.69 96.83 97.26 i-np 15046 15125 14708 97.24 97.75 97.50 o 23014 23051 22713 98.53 98.69 98.61 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.82 97.76 97.79 Avg2. 51357 51357 50297 97.94 97.94 97.94 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13181 12617 95.72 94.89 95.30 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.72 94.89 95.30 Avg2. 13297 13181 12617 95.72 94.89 95.30 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 142 Log-likelihood = -34674.219143 Norm (log-likelihood gradient vector) = 1436.792943 Norm (lambda vector) = 123.505303 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13191 12884 97.67 96.89 97.28 i-np 15046 15113 14705 97.30 97.73 97.52 o 23014 23053 22717 98.54 98.71 98.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.84 97.78 97.81 Avg2. 51357 51357 50306 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13191 12628 95.73 94.97 95.35 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.73 94.97 95.35 Avg2. 13297 13191 12628 95.73 94.97 95.35 Current max chunk-based F1: 95.38 (iteration 80) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 143 Log-likelihood = -34089.247835 Norm (log-likelihood gradient vector) = 2015.697548 Norm (lambda vector) = 124.707822 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13178 12879 97.73 96.86 97.29 i-np 15046 15117 14717 97.35 97.81 97.58 o 23014 23062 22725 98.54 98.74 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.87 97.80 97.84 Avg2. 51357 51357 50321 97.98 97.98 97.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13178 12631 95.85 94.99 95.42 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.85 94.99 95.42 Avg2. 13297 13178 12631 95.85 94.99 95.42 Current max chunk-based F1: 95.42 (iteration 143) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 144 Log-likelihood = -33515.731702 Norm (log-likelihood gradient vector) = 5712.149300 Norm (lambda vector) = 128.675590 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13189 12887 97.71 96.92 97.31 i-np 15046 15117 14714 97.33 97.79 97.56 o 23014 23051 22720 98.56 98.72 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.87 97.81 97.84 Avg2. 51357 51357 50321 97.98 97.98 97.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13189 12633 95.78 95.01 95.39 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.78 95.01 95.39 Avg2. 13297 13189 12633 95.78 95.01 95.39 Current max chunk-based F1: 95.42 (iteration 143) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 145 Log-likelihood = -33476.488765 Norm (log-likelihood gradient vector) = 2661.036962 Norm (lambda vector) = 126.698212 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13188 12889 97.73 96.93 97.33 i-np 15046 15124 14717 97.31 97.81 97.56 o 23014 23045 22717 98.58 98.71 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.87 97.82 97.85 Avg2. 51357 51357 50323 97.99 97.99 97.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13188 12632 95.78 95.00 95.39 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.78 95.00 95.39 Avg2. 13297 13188 12632 95.78 95.00 95.39 Current max chunk-based F1: 95.42 (iteration 143) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 146 Log-likelihood = -32919.625302 Norm (log-likelihood gradient vector) = 1505.023325 Norm (lambda vector) = 128.508366 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13186 12885 97.72 96.90 97.31 i-np 15046 15127 14714 97.27 97.79 97.53 o 23014 23044 22714 98.57 98.70 98.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.85 97.80 97.82 Avg2. 51357 51357 50313 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13186 12627 95.76 94.96 95.36 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.76 94.96 95.36 Avg2. 13297 13186 12627 95.76 94.96 95.36 Current max chunk-based F1: 95.42 (iteration 143) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 147 Log-likelihood = -32543.604181 Norm (log-likelihood gradient vector) = 1467.433749 Norm (lambda vector) = 129.840390 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13195 12889 97.68 96.93 97.30 i-np 15046 15105 14701 97.33 97.71 97.52 o 23014 23057 22718 98.53 98.71 98.62 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.85 97.78 97.81 Avg2. 51357 51357 50308 97.96 97.96 97.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13195 12624 95.67 94.94 95.30 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.67 94.94 95.30 Avg2. 13297 13195 12624 95.67 94.94 95.30 Current max chunk-based F1: 95.42 (iteration 143) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 148 Log-likelihood = -31824.379463 Norm (log-likelihood gradient vector) = 1546.229391 Norm (lambda vector) = 132.084708 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13122 12848 97.91 96.62 97.26 i-np 15046 15360 14806 96.39 98.40 97.39 o 23014 22875 22619 98.88 98.28 98.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.73 97.77 97.75 Avg2. 51357 51357 50273 97.89 97.89 97.89 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13122 12573 95.82 94.56 95.18 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.82 94.56 95.18 Avg2. 13297 13122 12573 95.82 94.56 95.18 Current max chunk-based F1: 95.42 (iteration 143) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 149 Log-likelihood = -32039.341242 Norm (log-likelihood gradient vector) = 10871.209296 Norm (lambda vector) = 136.056513 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13161 12877 97.84 96.84 97.34 i-np 15046 15225 14767 96.99 98.15 97.57 o 23014 22971 22680 98.73 98.55 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.85 97.85 Avg2. 51357 51357 50324 97.99 97.99 97.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13161 12622 95.90 94.92 95.41 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.90 94.92 95.41 Avg2. 13297 13161 12622 95.90 94.92 95.41 Current max chunk-based F1: 95.42 (iteration 143) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 150 Log-likelihood = -31447.858510 Norm (log-likelihood gradient vector) = 5002.345745 Norm (lambda vector) = 133.953350 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13297 13196 12891 97.69 96.95 97.32 i-np 15046 15121 14712 97.30 97.78 97.54 o 23014 23040 22714 98.59 98.70 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.81 97.83 Avg2. 51357 51357 50317 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13297 13196 12629 95.70 94.98 95.34 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.70 94.98 95.34 Avg2. 13297 13196 12629 95.70 94.98 95.34 Current max chunk-based F1: 95.42 (iteration 143) Training iteration elapsed (including evaluation time): 32 seconds The training process elapsed: 4758 seconds