OPTION VALUES: Model directory: ./IOB2+0.05/ 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: 886 (one data partition) Number of testing sequences: 43 (one data partition) Number of unlabeled sequences: 0 Number of context predicates: 709326 Number of features: 1351627 Feature rare threshold: 1 Context predicate rare threshold: 1 Using multiple rare thresholds for features: 0 Highlight feature: 0 Number of training iterations: 130 Initial lambda value: 0.0500 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 = -1890509.289523 Norm (log-likelihood gradient vector) = 466212.883099 Norm (lambda vector) = 58.129747 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 21090 20082 95.22 97.62 96.41 b-np 12220 11838 11054 93.38 90.46 91.89 i-np 13660 13523 12651 93.55 92.61 93.08 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.05 93.56 93.81 Avg2. 46451 46451 43787 94.26 94.26 94.26 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 11838 10390 87.77 85.02 86.37 ----- ------ ----- ----- ------- ------- ------------- Avg1. 87.77 85.02 86.37 Avg2. 12220 11838 10390 87.77 85.02 86.37 Current max chunk-based F1: 86.37 (iteration 1) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 2 Log-likelihood = -1465026.758519 Norm (log-likelihood gradient vector) = 372052.230017 Norm (lambda vector) = 58.160868 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 21277 19415 91.25 94.38 92.79 b-np 12220 11065 9859 89.10 80.68 84.68 i-np 13660 14109 12319 87.31 90.18 88.72 ----- ------ ----- ----- ------- ------- ------------- Avg1. 89.22 88.41 88.82 Avg2. 46451 46451 41593 89.54 89.54 89.54 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 11065 8625 77.95 70.58 74.08 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.95 70.58 74.08 Avg2. 12220 11065 8625 77.95 70.58 74.08 Current max chunk-based F1: 86.37 (iteration 1) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 3 Log-likelihood = -626239.775339 Norm (log-likelihood gradient vector) = 163799.575405 Norm (lambda vector) = 58.589013 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20872 19834 95.03 96.42 95.72 b-np 12220 10799 10197 94.43 83.45 88.60 i-np 13660 14780 12945 87.58 94.77 91.03 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.35 91.54 91.94 Avg2. 46451 46451 42976 92.52 92.52 92.52 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 10799 9148 84.71 74.86 79.48 ----- ------ ----- ----- ------- ------- ------------- Avg1. 84.71 74.86 79.48 Avg2. 12220 10799 9148 84.71 74.86 79.48 Current max chunk-based F1: 86.37 (iteration 1) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 4 Log-likelihood = -520436.366147 Norm (log-likelihood gradient vector) = 89987.396063 Norm (lambda vector) = 58.676580 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 21097 20105 95.30 97.73 96.50 b-np 12220 11649 10965 94.13 89.73 91.88 i-np 13660 13705 12773 93.20 93.51 93.35 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.21 93.66 93.93 Avg2. 46451 46451 43843 94.39 94.39 94.39 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 11649 10230 87.82 83.72 85.72 ----- ------ ----- ----- ------- ------- ------------- Avg1. 87.82 83.72 85.72 Avg2. 12220 11649 10230 87.82 83.72 85.72 Current max chunk-based F1: 86.37 (iteration 1) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 5 Log-likelihood = -473307.979962 Norm (log-likelihood gradient vector) = 70965.074665 Norm (lambda vector) = 58.765037 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 21258 20266 95.33 98.52 96.90 b-np 12220 11963 11311 94.55 92.56 93.55 i-np 13660 13230 12734 96.25 93.22 94.71 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.38 94.77 95.07 Avg2. 46451 46451 44311 95.39 95.39 95.39 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 11963 10816 90.41 88.51 89.45 ----- ------ ----- ----- ------- ------- ------------- Avg1. 90.41 88.51 89.45 Avg2. 12220 11963 10816 90.41 88.51 89.45 Current max chunk-based F1: 89.45 (iteration 5) Training iteration elapsed (including evaluation time): 25 seconds Iteration: 6 Log-likelihood = -409845.200265 Norm (log-likelihood gradient vector) = 56417.832952 Norm (lambda vector) = 59.012105 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20826 20162 96.81 98.01 97.41 b-np 12220 12091 11492 95.05 94.04 94.54 i-np 13660 13534 12935 95.57 94.69 95.13 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.81 95.58 95.70 Avg2. 46451 46451 44589 95.99 95.99 95.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12091 11009 91.05 90.09 90.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 91.05 90.09 90.57 Avg2. 12220 12091 11009 91.05 90.09 90.57 Current max chunk-based F1: 90.57 (iteration 6) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 7 Log-likelihood = -360416.164183 Norm (log-likelihood gradient vector) = 51146.518429 Norm (lambda vector) = 59.391287 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 21051 20330 96.57 98.83 97.69 b-np 12220 12142 11598 95.52 94.91 95.21 i-np 13660 13258 12853 96.95 94.09 95.50 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.35 95.94 96.14 Avg2. 46451 46451 44781 96.40 96.40 96.40 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12142 11167 91.97 91.38 91.68 ----- ------ ----- ----- ------- ------- ------------- Avg1. 91.97 91.38 91.68 Avg2. 12220 12142 11167 91.97 91.38 91.68 Current max chunk-based F1: 91.68 (iteration 7) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 8 Log-likelihood = -313345.440709 Norm (log-likelihood gradient vector) = 48636.481468 Norm (lambda vector) = 60.099627 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20797 20270 97.47 98.54 98.00 b-np 12220 12052 11626 96.47 95.14 95.80 i-np 13660 13602 13099 96.30 95.89 96.10 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.74 96.52 96.63 Avg2. 46451 46451 44995 96.87 96.87 96.87 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12052 11216 93.06 91.78 92.42 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.06 91.78 92.42 Avg2. 12220 12052 11216 93.06 91.78 92.42 Current max chunk-based F1: 92.42 (iteration 8) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 9 Log-likelihood = -277654.156557 Norm (log-likelihood gradient vector) = 30573.330398 Norm (lambda vector) = 61.277179 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20768 20278 97.64 98.58 98.11 b-np 12220 12073 11662 96.60 95.43 96.01 i-np 13660 13610 13122 96.41 96.06 96.24 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.88 96.69 96.79 Avg2. 46451 46451 45062 97.01 97.01 97.01 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12073 11258 93.25 92.13 92.69 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.25 92.13 92.69 Avg2. 12220 12073 11258 93.25 92.13 92.69 Current max chunk-based F1: 92.69 (iteration 9) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 10 Log-likelihood = -260549.654556 Norm (log-likelihood gradient vector) = 21395.914775 Norm (lambda vector) = 61.790210 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20748 20296 97.82 98.66 98.24 b-np 12220 12168 11750 96.56 96.15 96.36 i-np 13660 13535 13107 96.84 95.95 96.39 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.07 96.92 97.00 Avg2. 46451 46451 45153 97.21 97.21 97.21 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12168 11360 93.36 92.96 93.16 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.36 92.96 93.16 Avg2. 12220 12168 11360 93.36 92.96 93.16 Current max chunk-based F1: 93.16 (iteration 10) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 11 Log-likelihood = -235073.793148 Norm (log-likelihood gradient vector) = 17232.963029 Norm (lambda vector) = 63.055596 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20732 20303 97.93 98.70 98.31 b-np 12220 12303 11836 96.20 96.86 96.53 i-np 13660 13416 13075 97.46 95.72 96.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.20 97.09 97.14 Avg2. 46451 46451 45214 97.34 97.34 97.34 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12303 11481 93.32 93.95 93.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.32 93.95 93.63 Avg2. 12220 12303 11481 93.32 93.95 93.63 Current max chunk-based F1: 93.63 (iteration 11) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 12 Log-likelihood = -238036.923893 Norm (log-likelihood gradient vector) = 63305.742162 Norm (lambda vector) = 64.202050 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20750 20311 97.88 98.74 98.31 b-np 12220 12239 11802 96.43 96.58 96.50 i-np 13660 13462 13097 97.29 95.88 96.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.20 97.06 97.13 Avg2. 46451 46451 45210 97.33 97.33 97.33 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12239 11440 93.47 93.62 93.54 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.47 93.62 93.54 Avg2. 12220 12239 11440 93.47 93.62 93.54 Current max chunk-based F1: 93.63 (iteration 11) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 13 Log-likelihood = -229183.094239 Norm (log-likelihood gradient vector) = 26031.469793 Norm (lambda vector) = 63.530826 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20671 20287 98.14 98.62 98.38 b-np 12220 12175 11792 96.85 96.50 96.68 i-np 13660 13605 13189 96.94 96.55 96.75 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.31 97.22 97.27 Avg2. 46451 46451 45268 97.45 97.45 97.45 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12175 11438 93.95 93.60 93.77 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.95 93.60 93.77 Avg2. 12220 12175 11438 93.95 93.60 93.77 Current max chunk-based F1: 93.77 (iteration 13) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 14 Log-likelihood = -220570.233370 Norm (log-likelihood gradient vector) = 17903.695977 Norm (lambda vector) = 64.309699 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20690 20311 98.17 98.74 98.45 b-np 12220 12187 11812 96.92 96.66 96.79 i-np 13660 13574 13190 97.17 96.56 96.86 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.42 97.32 97.37 Avg2. 46451 46451 45313 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12187 11471 94.12 93.87 94.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.12 93.87 94.00 Avg2. 12220 12187 11471 94.12 93.87 94.00 Current max chunk-based F1: 94.00 (iteration 14) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 15 Log-likelihood = -216165.331374 Norm (log-likelihood gradient vector) = 17676.925475 Norm (lambda vector) = 65.324606 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20708 20324 98.15 98.80 98.47 b-np 12220 12187 11831 97.08 96.82 96.95 i-np 13660 13556 13202 97.39 96.65 97.02 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.54 97.42 97.48 Avg2. 46451 46451 45357 97.64 97.64 97.64 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12187 11512 94.46 94.21 94.33 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.46 94.21 94.33 Avg2. 12220 12187 11512 94.46 94.21 94.33 Current max chunk-based F1: 94.33 (iteration 15) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 16 Log-likelihood = -207054.696368 Norm (log-likelihood gradient vector) = 14474.857086 Norm (lambda vector) = 66.384706 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20633 20287 98.32 98.62 98.47 b-np 12220 12132 11811 97.35 96.65 97.00 i-np 13660 13686 13267 96.94 97.12 97.03 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.54 97.47 97.50 Avg2. 46451 46451 45365 97.66 97.66 97.66 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12132 11494 94.74 94.06 94.40 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.74 94.06 94.40 Avg2. 12220 12132 11494 94.74 94.06 94.40 Current max chunk-based F1: 94.40 (iteration 16) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 17 Log-likelihood = -186642.251769 Norm (log-likelihood gradient vector) = 19234.923526 Norm (lambda vector) = 68.680771 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20809 20366 97.87 99.00 98.43 b-np 12220 12243 11865 96.91 97.09 97.00 i-np 13660 13399 13130 97.99 96.12 97.05 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.59 97.41 97.50 Avg2. 46451 46451 45361 97.65 97.65 97.65 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12243 11571 94.51 94.69 94.60 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.51 94.69 94.60 Avg2. 12220 12243 11571 94.51 94.69 94.60 Current max chunk-based F1: 94.60 (iteration 17) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 18 Log-likelihood = -176102.131358 Norm (log-likelihood gradient vector) = 20889.260832 Norm (lambda vector) = 70.036188 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20729 20338 98.11 98.87 98.49 b-np 12220 12185 11848 97.23 96.96 97.09 i-np 13660 13537 13212 97.60 96.72 97.16 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.65 97.51 97.58 Avg2. 46451 46451 45398 97.73 97.73 97.73 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12185 11549 94.78 94.51 94.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.78 94.51 94.64 Avg2. 12220 12185 11549 94.78 94.51 94.64 Current max chunk-based F1: 94.64 (iteration 18) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 19 Log-likelihood = -168806.641427 Norm (log-likelihood gradient vector) = 13663.869594 Norm (lambda vector) = 70.239425 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20645 20305 98.35 98.71 98.53 b-np 12220 12133 11829 97.49 96.80 97.15 i-np 13660 13673 13281 97.13 97.23 97.18 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.66 97.58 97.62 Avg2. 46451 46451 45415 97.77 97.77 97.77 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12133 11524 94.98 94.30 94.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.98 94.30 94.64 Avg2. 12220 12133 11524 94.98 94.30 94.64 Current max chunk-based F1: 94.64 (iteration 18) Training iteration elapsed (including evaluation time): 25 seconds Iteration: 20 Log-likelihood = -158199.215658 Norm (log-likelihood gradient vector) = 8480.673547 Norm (lambda vector) = 70.507413 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20605 20285 98.45 98.61 98.53 b-np 12220 12130 11823 97.47 96.75 97.11 i-np 13660 13716 13300 96.97 97.36 97.17 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.63 97.58 97.60 Avg2. 46451 46451 45408 97.75 97.75 97.75 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12130 11517 94.95 94.25 94.60 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.95 94.25 94.60 Avg2. 12220 12130 11517 94.95 94.25 94.60 Current max chunk-based F1: 94.64 (iteration 18) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 21 Log-likelihood = -153002.951498 Norm (log-likelihood gradient vector) = 10099.287127 Norm (lambda vector) = 70.838540 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20586 20281 98.52 98.59 98.55 b-np 12220 12140 11833 97.47 96.83 97.15 i-np 13660 13725 13308 96.96 97.42 97.19 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.65 97.62 97.63 Avg2. 46451 46451 45422 97.78 97.78 97.78 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12140 11523 94.92 94.30 94.61 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.92 94.30 94.61 Avg2. 12220 12140 11523 94.92 94.30 94.61 Current max chunk-based F1: 94.64 (iteration 18) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 22 Log-likelihood = -149255.376520 Norm (log-likelihood gradient vector) = 9616.793327 Norm (lambda vector) = 71.483943 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20622 20302 98.45 98.69 98.57 b-np 12220 12245 11894 97.13 97.33 97.23 i-np 13660 13584 13246 97.51 96.97 97.24 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.70 97.66 97.68 Avg2. 46451 46451 45442 97.83 97.83 97.83 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12245 11603 94.76 94.95 94.85 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.76 94.95 94.85 Avg2. 12220 12245 11603 94.76 94.95 94.85 Current max chunk-based F1: 94.85 (iteration 22) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 23 Log-likelihood = -142203.729422 Norm (log-likelihood gradient vector) = 20085.791470 Norm (lambda vector) = 73.438355 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20611 20293 98.46 98.65 98.55 b-np 12220 12171 11859 97.44 97.05 97.24 i-np 13660 13669 13287 97.21 97.27 97.24 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.70 97.65 97.68 Avg2. 46451 46451 45439 97.82 97.82 97.82 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12171 11561 94.99 94.61 94.80 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.99 94.61 94.80 Avg2. 12220 12171 11561 94.99 94.61 94.80 Current max chunk-based F1: 94.85 (iteration 22) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 24 Log-likelihood = -138126.125385 Norm (log-likelihood gradient vector) = 9291.780472 Norm (lambda vector) = 74.964317 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20621 20303 98.46 98.70 98.58 b-np 12220 12176 11873 97.51 97.16 97.34 i-np 13660 13654 13289 97.33 97.28 97.31 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.77 97.71 97.74 Avg2. 46451 46451 45465 97.88 97.88 97.88 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12176 11580 95.11 94.76 94.93 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.11 94.76 94.93 Avg2. 12220 12176 11580 95.11 94.76 94.93 Current max chunk-based F1: 94.93 (iteration 24) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 25 Log-likelihood = -136592.366555 Norm (log-likelihood gradient vector) = 6955.862462 Norm (lambda vector) = 74.871955 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20617 20310 98.51 98.73 98.62 b-np 12220 12162 11879 97.67 97.21 97.44 i-np 13660 13672 13312 97.37 97.45 97.41 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.85 97.80 97.82 Avg2. 46451 46451 45501 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12162 11596 95.35 94.89 95.12 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.35 94.89 95.12 Avg2. 12220 12162 11596 95.35 94.89 95.12 Current max chunk-based F1: 95.12 (iteration 25) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 26 Log-likelihood = -130999.883547 Norm (log-likelihood gradient vector) = 4752.048474 Norm (lambda vector) = 76.164166 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20639 20325 98.48 98.80 98.64 b-np 12220 12181 11887 97.59 97.27 97.43 i-np 13660 13631 13289 97.49 97.28 97.39 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.85 97.79 97.82 Avg2. 46451 46451 45501 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12181 11609 95.30 95.00 95.15 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.30 95.00 95.15 Avg2. 12220 12181 11609 95.30 95.00 95.15 Current max chunk-based F1: 95.15 (iteration 26) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 27 Log-likelihood = -122175.401003 Norm (log-likelihood gradient vector) = 6493.478780 Norm (lambda vector) = 78.523031 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20395 20194 99.01 98.17 98.59 b-np 12220 12083 11833 97.93 96.83 97.38 i-np 13660 13973 13442 96.20 98.40 97.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.72 97.80 97.76 Avg2. 46451 46451 45469 97.89 97.89 97.89 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12083 11531 95.43 94.36 94.89 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.43 94.36 94.89 Avg2. 12220 12083 11531 95.43 94.36 94.89 Current max chunk-based F1: 95.15 (iteration 26) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 28 Log-likelihood = -120712.915191 Norm (log-likelihood gradient vector) = 20444.453823 Norm (lambda vector) = 81.612009 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20572 20304 98.70 98.70 98.70 b-np 12220 12186 11900 97.65 97.38 97.52 i-np 13660 13693 13334 97.38 97.61 97.50 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.91 97.90 97.90 Avg2. 46451 46451 45538 98.03 98.03 98.03 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12186 11634 95.47 95.20 95.34 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.47 95.20 95.34 Avg2. 12220 12186 11634 95.47 95.20 95.34 Current max chunk-based F1: 95.34 (iteration 28) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 29 Log-likelihood = -116463.563250 Norm (log-likelihood gradient vector) = 6072.542521 Norm (lambda vector) = 81.732770 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20610 20326 98.62 98.81 98.72 b-np 12220 12211 11917 97.59 97.52 97.56 i-np 13660 13630 13306 97.62 97.41 97.52 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.95 97.91 97.93 Avg2. 46451 46451 45549 98.06 98.06 98.06 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12211 11654 95.44 95.37 95.40 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.44 95.37 95.40 Avg2. 12220 12211 11654 95.44 95.37 95.40 Current max chunk-based F1: 95.40 (iteration 29) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 30 Log-likelihood = -114871.431327 Norm (log-likelihood gradient vector) = 5430.209350 Norm (lambda vector) = 82.141612 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20628 20338 98.59 98.87 98.73 b-np 12220 12220 11933 97.65 97.65 97.65 i-np 13660 13603 13298 97.76 97.35 97.55 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.00 97.96 97.98 Avg2. 46451 46451 45569 98.10 98.10 98.10 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12220 11669 95.49 95.49 95.49 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.49 95.49 95.49 Avg2. 12220 12220 11669 95.49 95.49 95.49 Current max chunk-based F1: 95.49 (iteration 30) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 31 Log-likelihood = -113618.556275 Norm (log-likelihood gradient vector) = 6262.070995 Norm (lambda vector) = 82.958859 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20633 20346 98.61 98.91 98.76 b-np 12220 12185 11924 97.86 97.58 97.72 i-np 13660 13633 13321 97.71 97.52 97.61 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.06 98.00 98.03 Avg2. 46451 46451 45591 98.15 98.15 98.15 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12185 11668 95.76 95.48 95.62 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.76 95.48 95.62 Avg2. 12220 12185 11668 95.76 95.48 95.62 Current max chunk-based F1: 95.62 (iteration 31) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 32 Log-likelihood = -112312.192955 Norm (log-likelihood gradient vector) = 8944.072796 Norm (lambda vector) = 83.652906 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20621 20349 98.68 98.92 98.80 b-np 12220 12217 11952 97.83 97.81 97.82 i-np 13660 13613 13318 97.83 97.50 97.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.11 98.07 98.09 Avg2. 46451 46451 45619 98.21 98.21 98.21 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12217 11699 95.76 95.74 95.75 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.76 95.74 95.75 Avg2. 12220 12217 11699 95.76 95.74 95.75 Current max chunk-based F1: 95.75 (iteration 32) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 33 Log-likelihood = -111681.035565 Norm (log-likelihood gradient vector) = 7018.270845 Norm (lambda vector) = 85.172818 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20587 20336 98.78 98.86 98.82 b-np 12220 12185 11938 97.97 97.69 97.83 i-np 13660 13679 13357 97.65 97.78 97.71 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.13 98.11 98.12 Avg2. 46451 46451 45631 98.23 98.23 98.23 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12185 11692 95.95 95.68 95.82 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.95 95.68 95.82 Avg2. 12220 12185 11692 95.95 95.68 95.82 Current max chunk-based F1: 95.82 (iteration 33) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 34 Log-likelihood = -109211.792075 Norm (log-likelihood gradient vector) = 5329.844031 Norm (lambda vector) = 84.857924 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20584 20333 98.78 98.84 98.81 b-np 12220 12161 11922 98.03 97.56 97.80 i-np 13660 13706 13368 97.53 97.86 97.70 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.12 98.09 98.10 Avg2. 46451 46451 45623 98.22 98.22 98.22 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12161 11670 95.96 95.50 95.73 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.96 95.50 95.73 Avg2. 12220 12161 11670 95.96 95.50 95.73 Current max chunk-based F1: 95.82 (iteration 33) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 35 Log-likelihood = -106257.611050 Norm (log-likelihood gradient vector) = 5897.814061 Norm (lambda vector) = 84.404389 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20590 20337 98.77 98.86 98.82 b-np 12220 12171 11928 98.00 97.61 97.81 i-np 13660 13690 13364 97.62 97.83 97.73 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.13 98.10 98.12 Avg2. 46451 46451 45629 98.23 98.23 98.23 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12171 11681 95.97 95.59 95.78 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.97 95.59 95.78 Avg2. 12220 12171 11681 95.97 95.59 95.78 Current max chunk-based F1: 95.82 (iteration 33) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 36 Log-likelihood = -103710.948213 Norm (log-likelihood gradient vector) = 4354.456687 Norm (lambda vector) = 84.556741 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20584 20328 98.76 98.82 98.79 b-np 12220 12177 11922 97.91 97.56 97.73 i-np 13660 13690 13354 97.55 97.76 97.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.07 98.05 98.06 Avg2. 46451 46451 45604 98.18 98.18 98.18 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12177 11666 95.80 95.47 95.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.80 95.47 95.63 Avg2. 12220 12177 11666 95.80 95.47 95.63 Current max chunk-based F1: 95.82 (iteration 33) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 37 Log-likelihood = -100211.444617 Norm (log-likelihood gradient vector) = 4848.489680 Norm (lambda vector) = 85.243780 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20689 20381 98.51 99.08 98.79 b-np 12220 12240 11955 97.67 97.83 97.75 i-np 13660 13522 13271 98.14 97.15 97.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.11 98.02 98.06 Avg2. 46451 46451 45607 98.18 98.18 98.18 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12240 11709 95.66 95.82 95.74 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.66 95.82 95.74 Avg2. 12220 12240 11709 95.66 95.82 95.74 Current max chunk-based F1: 95.82 (iteration 33) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 38 Log-likelihood = -93841.417611 Norm (log-likelihood gradient vector) = 10899.078711 Norm (lambda vector) = 86.601354 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20596 20336 98.74 98.86 98.80 b-np 12220 12191 11929 97.85 97.62 97.73 i-np 13660 13664 13346 97.67 97.70 97.69 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.09 98.06 98.07 Avg2. 46451 46451 45611 98.19 98.19 98.19 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12191 11681 95.82 95.59 95.70 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.82 95.59 95.70 Avg2. 12220 12191 11681 95.82 95.59 95.70 Current max chunk-based F1: 95.82 (iteration 33) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 39 Log-likelihood = -89755.282816 Norm (log-likelihood gradient vector) = 4958.437061 Norm (lambda vector) = 88.024069 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20581 20333 98.80 98.84 98.82 b-np 12220 12182 11930 97.93 97.63 97.78 i-np 13660 13688 13366 97.65 97.85 97.75 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.12 98.11 98.12 Avg2. 46451 46451 45629 98.23 98.23 98.23 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12182 11688 95.94 95.65 95.80 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.94 95.65 95.80 Avg2. 12220 12182 11688 95.94 95.65 95.80 Current max chunk-based F1: 95.82 (iteration 33) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 40 Log-likelihood = -89184.040802 Norm (log-likelihood gradient vector) = 3847.169548 Norm (lambda vector) = 87.960804 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20583 20338 98.81 98.87 98.84 b-np 12220 12178 11934 98.00 97.66 97.83 i-np 13660 13690 13375 97.70 97.91 97.81 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.17 98.15 98.16 Avg2. 46451 46451 45647 98.27 98.27 98.27 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12178 11701 96.08 95.75 95.92 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.08 95.75 95.92 Avg2. 12220 12178 11701 96.08 95.75 95.92 Current max chunk-based F1: 95.92 (iteration 40) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 41 Log-likelihood = -87869.183490 Norm (log-likelihood gradient vector) = 3423.110111 Norm (lambda vector) = 88.138866 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20559 20326 98.87 98.81 98.84 b-np 12220 12150 11918 98.09 97.53 97.81 i-np 13660 13742 13397 97.49 98.07 97.78 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.15 98.14 98.14 Avg2. 46451 46451 45641 98.26 98.26 98.26 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12150 11681 96.14 95.59 95.86 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.14 95.59 95.86 Avg2. 12220 12150 11681 96.14 95.59 95.86 Current max chunk-based F1: 95.92 (iteration 40) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 42 Log-likelihood = -85166.926836 Norm (log-likelihood gradient vector) = 8482.384245 Norm (lambda vector) = 89.068255 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20582 20336 98.80 98.86 98.83 b-np 12220 12189 11932 97.89 97.64 97.77 i-np 13660 13680 13359 97.65 97.80 97.72 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.12 98.10 98.11 Avg2. 46451 46451 45627 98.23 98.23 98.23 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12189 11688 95.89 95.65 95.77 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.89 95.65 95.77 Avg2. 12220 12189 11688 95.89 95.65 95.77 Current max chunk-based F1: 95.92 (iteration 40) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 43 Log-likelihood = -83844.127122 Norm (log-likelihood gradient vector) = 5019.287703 Norm (lambda vector) = 89.847056 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20566 20332 98.86 98.84 98.85 b-np 12220 12175 11927 97.96 97.60 97.78 i-np 13660 13710 13383 97.61 97.97 97.79 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.15 98.14 98.14 Avg2. 46451 46451 45642 98.26 98.26 98.26 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12175 11688 96.00 95.65 95.82 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.00 95.65 95.82 Avg2. 12220 12175 11688 96.00 95.65 95.82 Current max chunk-based F1: 95.92 (iteration 40) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 44 Log-likelihood = -81982.947070 Norm (log-likelihood gradient vector) = 3260.265557 Norm (lambda vector) = 90.376901 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20579 20338 98.83 98.87 98.85 b-np 12220 12170 11925 97.99 97.59 97.79 i-np 13660 13702 13383 97.67 97.97 97.82 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.16 98.14 98.15 Avg2. 46451 46451 45646 98.27 98.27 98.27 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12170 11693 96.08 95.69 95.88 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.08 95.69 95.88 Avg2. 12220 12170 11693 96.08 95.69 95.88 Current max chunk-based F1: 95.92 (iteration 40) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 45 Log-likelihood = -80232.505453 Norm (log-likelihood gradient vector) = 3203.493472 Norm (lambda vector) = 91.239227 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20591 20336 98.76 98.86 98.81 b-np 12220 12253 11948 97.51 97.77 97.64 i-np 13660 13607 13322 97.91 97.53 97.72 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.06 98.05 98.06 Avg2. 46451 46451 45606 98.18 98.18 98.18 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12253 11706 95.54 95.79 95.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.54 95.79 95.66 Avg2. 12220 12253 11706 95.54 95.79 95.66 Current max chunk-based F1: 95.92 (iteration 40) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 46 Log-likelihood = -79582.320735 Norm (log-likelihood gradient vector) = 16946.169788 Norm (lambda vector) = 93.866119 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20584 20349 98.86 98.92 98.89 b-np 12220 12202 11950 97.93 97.79 97.86 i-np 13660 13665 13376 97.89 97.92 97.90 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.23 98.21 98.22 Avg2. 46451 46451 45675 98.33 98.33 98.33 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12202 11726 96.10 95.96 96.03 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.10 95.96 96.03 Avg2. 12220 12202 11726 96.10 95.96 96.03 Current max chunk-based F1: 96.03 (iteration 46) Training iteration elapsed (including evaluation time): 25 seconds Iteration: 47 Log-likelihood = -78995.856760 Norm (log-likelihood gradient vector) = 6017.853589 Norm (lambda vector) = 92.177166 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20571 20340 98.88 98.88 98.88 b-np 12220 12184 11938 97.98 97.69 97.84 i-np 13660 13696 13387 97.74 98.00 97.87 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.20 98.19 98.20 Avg2. 46451 46451 45665 98.31 98.31 98.31 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12184 11711 96.12 95.83 95.98 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.12 95.83 95.98 Avg2. 12220 12184 11711 96.12 95.83 95.98 Current max chunk-based F1: 96.03 (iteration 46) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 48 Log-likelihood = -78082.940172 Norm (log-likelihood gradient vector) = 3634.454177 Norm (lambda vector) = 92.783627 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20573 20343 98.88 98.89 98.89 b-np 12220 12177 11933 98.00 97.65 97.82 i-np 13660 13701 13393 97.75 98.05 97.90 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.21 98.20 98.20 Avg2. 46451 46451 45669 98.32 98.32 98.32 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12177 11708 96.15 95.81 95.98 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.15 95.81 95.98 Avg2. 12220 12177 11708 96.15 95.81 95.98 Current max chunk-based F1: 96.03 (iteration 46) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 49 Log-likelihood = -76943.403487 Norm (log-likelihood gradient vector) = 3697.097290 Norm (lambda vector) = 93.873586 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20555 20331 98.91 98.83 98.87 b-np 12220 12173 11931 98.01 97.64 97.82 i-np 13660 13723 13402 97.66 98.11 97.89 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.19 98.19 98.19 Avg2. 46451 46451 45664 98.31 98.31 98.31 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12173 11704 96.15 95.78 95.96 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.15 95.78 95.96 Avg2. 12220 12173 11704 96.15 95.78 95.96 Current max chunk-based F1: 96.03 (iteration 46) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 50 Log-likelihood = -75764.935256 Norm (log-likelihood gradient vector) = 4001.658929 Norm (lambda vector) = 94.364357 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20717 20393 98.44 99.13 98.78 b-np 12220 12266 11961 97.51 97.88 97.70 i-np 13660 13468 13243 98.33 96.95 97.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.09 97.99 98.04 Avg2. 46451 46451 45597 98.16 98.16 98.16 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12266 11726 95.60 95.96 95.78 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.60 95.96 95.78 Avg2. 12220 12266 11726 95.60 95.96 95.78 Current max chunk-based F1: 96.03 (iteration 46) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 51 Log-likelihood = -74260.692674 Norm (log-likelihood gradient vector) = 11101.485863 Norm (lambda vector) = 95.471097 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20639 20364 98.67 98.99 98.83 b-np 12220 12217 11945 97.77 97.75 97.76 i-np 13660 13595 13324 98.01 97.54 97.77 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.15 98.09 98.12 Avg2. 46451 46451 45633 98.24 98.24 98.24 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12217 11716 95.90 95.88 95.89 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.90 95.88 95.89 Avg2. 12220 12217 11716 95.90 95.88 95.89 Current max chunk-based F1: 96.03 (iteration 46) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 52 Log-likelihood = -74170.734478 Norm (log-likelihood gradient vector) = 5743.972026 Norm (lambda vector) = 94.936009 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20582 20344 98.84 98.90 98.87 b-np 12220 12187 11939 97.97 97.70 97.83 i-np 13660 13682 13378 97.78 97.94 97.86 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.20 98.18 98.19 Avg2. 46451 46451 45661 98.30 98.30 98.30 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12187 11715 96.13 95.87 96.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.13 95.87 96.00 Avg2. 12220 12187 11715 96.13 95.87 96.00 Current max chunk-based F1: 96.03 (iteration 46) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 53 Log-likelihood = -72583.832853 Norm (log-likelihood gradient vector) = 3488.986427 Norm (lambda vector) = 95.082140 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20571 20343 98.89 98.89 98.89 b-np 12220 12193 11946 97.97 97.76 97.87 i-np 13660 13687 13387 97.81 98.00 97.90 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.22 98.22 98.22 Avg2. 46451 46451 45676 98.33 98.33 98.33 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12193 11727 96.18 95.97 96.07 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.18 95.97 96.07 Avg2. 12220 12193 11727 96.18 95.97 96.07 Current max chunk-based F1: 96.07 (iteration 53) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 54 Log-likelihood = -71751.008877 Norm (log-likelihood gradient vector) = 2959.919630 Norm (lambda vector) = 95.146220 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20561 20335 98.90 98.85 98.88 b-np 12220 12185 11940 97.99 97.71 97.85 i-np 13660 13705 13391 97.71 98.03 97.87 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.20 98.20 98.20 Avg2. 46451 46451 45666 98.31 98.31 98.31 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12185 11719 96.18 95.90 96.04 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.18 95.90 96.04 Avg2. 12220 12185 11719 96.18 95.90 96.04 Current max chunk-based F1: 96.07 (iteration 53) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 55 Log-likelihood = -70869.713008 Norm (log-likelihood gradient vector) = 3214.564321 Norm (lambda vector) = 95.583961 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20554 20324 98.88 98.80 98.84 b-np 12220 12179 11933 97.98 97.65 97.82 i-np 13660 13718 13383 97.56 97.97 97.76 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.14 98.14 98.14 Avg2. 46451 46451 45640 98.25 98.25 98.25 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12179 11697 96.04 95.72 95.88 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.04 95.72 95.88 Avg2. 12220 12179 11697 96.04 95.72 95.88 Current max chunk-based F1: 96.07 (iteration 53) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 56 Log-likelihood = -69559.185995 Norm (log-likelihood gradient vector) = 2682.496601 Norm (lambda vector) = 96.565085 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20584 20343 98.83 98.89 98.86 b-np 12220 12230 11957 97.77 97.85 97.81 i-np 13660 13637 13351 97.90 97.74 97.82 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.17 98.16 98.16 Avg2. 46451 46451 45651 98.28 98.28 98.28 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12230 11732 95.93 96.01 95.97 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.93 96.01 95.97 Avg2. 12220 12230 11732 95.93 96.01 95.97 Current max chunk-based F1: 96.07 (iteration 53) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 57 Log-likelihood = -68234.201790 Norm (log-likelihood gradient vector) = 9860.007520 Norm (lambda vector) = 98.616805 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20566 20335 98.88 98.85 98.86 b-np 12220 12207 11948 97.88 97.77 97.83 i-np 13660 13678 13373 97.77 97.90 97.83 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.18 98.18 98.18 Avg2. 46451 46451 45656 98.29 98.29 98.29 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12207 11725 96.05 95.95 96.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.05 95.95 96.00 Avg2. 12220 12207 11725 96.05 95.95 96.00 Current max chunk-based F1: 96.07 (iteration 53) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 58 Log-likelihood = -66653.101694 Norm (log-likelihood gradient vector) = 4134.432850 Norm (lambda vector) = 99.437910 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20557 20331 98.90 98.83 98.87 b-np 12220 12195 11944 97.94 97.74 97.84 i-np 13660 13699 13385 97.71 97.99 97.85 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.18 98.19 98.19 Avg2. 46451 46451 45660 98.30 98.30 98.30 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12195 11725 96.15 95.95 96.05 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.15 95.95 96.05 Avg2. 12220 12195 11725 96.15 95.95 96.05 Current max chunk-based F1: 96.07 (iteration 53) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 59 Log-likelihood = -65518.727354 Norm (log-likelihood gradient vector) = 3402.025352 Norm (lambda vector) = 99.872412 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20565 20336 98.89 98.86 98.87 b-np 12220 12202 11946 97.90 97.76 97.83 i-np 13660 13684 13377 97.76 97.93 97.84 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.18 98.18 98.18 Avg2. 46451 46451 45659 98.29 98.29 98.29 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12202 11726 96.10 95.96 96.03 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.10 95.96 96.03 Avg2. 12220 12202 11726 96.10 95.96 96.03 Current max chunk-based F1: 96.07 (iteration 53) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 60 Log-likelihood = -64110.665438 Norm (log-likelihood gradient vector) = 3151.711788 Norm (lambda vector) = 100.581265 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20542 20328 98.96 98.82 98.89 b-np 12220 12182 11942 98.03 97.73 97.88 i-np 13660 13727 13409 97.68 98.16 97.92 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.22 98.24 98.23 Avg2. 46451 46451 45679 98.34 98.34 98.34 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12182 11727 96.26 95.97 96.12 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.26 95.97 96.12 Avg2. 12220 12182 11727 96.26 95.97 96.12 Current max chunk-based F1: 96.12 (iteration 60) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 61 Log-likelihood = -60904.504594 Norm (log-likelihood gradient vector) = 3486.201468 Norm (lambda vector) = 102.318385 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20650 20370 98.64 99.02 98.83 b-np 12220 12263 11972 97.63 97.97 97.80 i-np 13660 13538 13299 98.23 97.36 97.79 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.17 98.12 98.14 Avg2. 46451 46451 45641 98.26 98.26 98.26 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12263 11754 95.85 96.19 96.02 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.85 96.19 96.02 Avg2. 12220 12263 11754 95.85 96.19 96.02 Current max chunk-based F1: 96.12 (iteration 60) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 62 Log-likelihood = -58522.222592 Norm (log-likelihood gradient vector) = 9526.035101 Norm (lambda vector) = 104.259647 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20594 20349 98.81 98.92 98.87 b-np 12220 12217 11954 97.85 97.82 97.84 i-np 13660 13640 13360 97.95 97.80 97.88 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.20 98.18 98.19 Avg2. 46451 46451 45663 98.30 98.30 98.30 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12217 11737 96.07 96.05 96.06 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.07 96.05 96.06 Avg2. 12220 12217 11737 96.07 96.05 96.06 Current max chunk-based F1: 96.12 (iteration 60) Training iteration elapsed (including evaluation time): 25 seconds Iteration: 63 Log-likelihood = -59205.168244 Norm (log-likelihood gradient vector) = 4042.087718 Norm (lambda vector) = 103.268413 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20574 20342 98.87 98.89 98.88 b-np 12220 12211 11954 97.90 97.82 97.86 i-np 13660 13666 13377 97.89 97.93 97.91 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.22 98.21 98.22 Avg2. 46451 46451 45673 98.33 98.33 98.33 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12211 11737 96.12 96.05 96.08 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.12 96.05 96.08 Avg2. 12220 12211 11737 96.12 96.05 96.08 Current max chunk-based F1: 96.12 (iteration 60) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 64 Log-likelihood = -57762.226321 Norm (log-likelihood gradient vector) = 2380.751051 Norm (lambda vector) = 104.550247 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20555 20337 98.94 98.86 98.90 b-np 12220 12196 11949 97.97 97.78 97.88 i-np 13660 13700 13401 97.82 98.10 97.96 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.24 98.25 98.25 Avg2. 46451 46451 45687 98.36 98.36 98.36 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12196 11736 96.23 96.04 96.13 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.23 96.04 96.13 Avg2. 12220 12196 11736 96.23 96.04 96.13 Current max chunk-based F1: 96.13 (iteration 64) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 65 Log-likelihood = -57132.533271 Norm (log-likelihood gradient vector) = 2334.705835 Norm (lambda vector) = 105.606429 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20551 20331 98.93 98.83 98.88 b-np 12220 12197 11945 97.93 97.75 97.84 i-np 13660 13703 13397 97.77 98.07 97.92 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.21 98.22 98.21 Avg2. 46451 46451 45673 98.33 98.33 98.33 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12197 11726 96.14 95.96 96.05 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.14 95.96 96.05 Avg2. 12220 12197 11726 96.14 95.96 96.05 Current max chunk-based F1: 96.13 (iteration 64) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 66 Log-likelihood = -56369.573578 Norm (log-likelihood gradient vector) = 3423.088122 Norm (lambda vector) = 107.319891 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20518 20308 98.98 98.72 98.85 b-np 12220 12155 11926 98.12 97.59 97.85 i-np 13660 13778 13425 97.44 98.28 97.86 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.18 98.20 98.19 Avg2. 46451 46451 45659 98.29 98.29 98.29 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12155 11701 96.26 95.75 96.01 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.26 95.75 96.01 Avg2. 12220 12155 11701 96.26 95.75 96.01 Current max chunk-based F1: 96.13 (iteration 64) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 67 Log-likelihood = -55793.690409 Norm (log-likelihood gradient vector) = 4814.486847 Norm (lambda vector) = 109.878567 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20576 20346 98.88 98.91 98.89 b-np 12220 12208 11957 97.94 97.85 97.90 i-np 13660 13667 13389 97.97 98.02 97.99 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.26 98.26 98.26 Avg2. 46451 46451 45692 98.37 98.37 98.37 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12208 11749 96.24 96.15 96.19 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.24 96.15 96.19 Avg2. 12220 12208 11749 96.24 96.15 96.19 Current max chunk-based F1: 96.19 (iteration 67) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 68 Log-likelihood = -55575.543952 Norm (log-likelihood gradient vector) = 3792.904270 Norm (lambda vector) = 109.546519 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20586 20349 98.85 98.92 98.88 b-np 12220 12205 11955 97.95 97.83 97.89 i-np 13660 13660 13380 97.95 97.95 97.95 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.25 98.23 98.24 Avg2. 46451 46451 45684 98.35 98.35 98.35 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12205 11748 96.26 96.14 96.20 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.26 96.14 96.20 Avg2. 12220 12205 11748 96.26 96.14 96.20 Current max chunk-based F1: 96.20 (iteration 68) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 69 Log-likelihood = -54921.283069 Norm (log-likelihood gradient vector) = 2204.443966 Norm (lambda vector) = 109.498471 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20607 20358 98.79 98.96 98.88 b-np 12220 12216 11963 97.93 97.90 97.91 i-np 13660 13628 13361 98.04 97.81 97.93 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.25 98.22 98.24 Avg2. 46451 46451 45682 98.34 98.34 98.34 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12216 11753 96.21 96.18 96.19 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.21 96.18 96.19 Avg2. 12220 12216 11753 96.21 96.18 96.19 Current max chunk-based F1: 96.20 (iteration 68) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 70 Log-likelihood = -54077.795568 Norm (log-likelihood gradient vector) = 2348.658092 Norm (lambda vector) = 109.817103 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20580 20349 98.88 98.92 98.90 b-np 12220 12182 11952 98.11 97.81 97.96 i-np 13660 13689 13398 97.87 98.08 97.98 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.29 98.27 98.28 Avg2. 46451 46451 45699 98.38 98.38 98.38 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12182 11747 96.43 96.13 96.28 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.43 96.13 96.28 Avg2. 12220 12182 11747 96.43 96.13 96.28 Current max chunk-based F1: 96.28 (iteration 70) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 71 Log-likelihood = -52217.684958 Norm (log-likelihood gradient vector) = 3485.588219 Norm (lambda vector) = 110.894094 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20611 20364 98.80 98.99 98.90 b-np 12220 12212 11968 98.00 97.94 97.97 i-np 13660 13628 13363 98.06 97.83 97.94 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.29 98.25 98.27 Avg2. 46451 46451 45695 98.37 98.37 98.37 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12212 11757 96.27 96.21 96.24 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.27 96.21 96.24 Avg2. 12220 12212 11757 96.27 96.21 96.24 Current max chunk-based F1: 96.28 (iteration 70) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 72 Log-likelihood = -49877.392317 Norm (log-likelihood gradient vector) = 3198.218449 Norm (lambda vector) = 112.355170 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20579 20352 98.90 98.94 98.92 b-np 12220 12205 11969 98.07 97.95 98.01 i-np 13660 13667 13388 97.96 98.01 97.98 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.31 98.30 98.30 Avg2. 46451 46451 45709 98.40 98.40 98.40 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12205 11762 96.37 96.25 96.31 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.37 96.25 96.31 Avg2. 12220 12205 11762 96.37 96.25 96.31 Current max chunk-based F1: 96.31 (iteration 72) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 73 Log-likelihood = -48134.916566 Norm (log-likelihood gradient vector) = 1957.534172 Norm (lambda vector) = 113.613167 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20573 20351 98.92 98.93 98.93 b-np 12220 12193 11964 98.12 97.91 98.01 i-np 13660 13685 13399 97.91 98.09 98.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.32 98.31 98.31 Avg2. 46451 46451 45714 98.41 98.41 98.41 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12193 11754 96.40 96.19 96.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.40 96.19 96.29 Avg2. 12220 12193 11754 96.40 96.19 96.29 Current max chunk-based F1: 96.31 (iteration 72) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 74 Log-likelihood = -47067.494061 Norm (log-likelihood gradient vector) = 1874.928176 Norm (lambda vector) = 114.229073 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20569 20345 98.91 98.90 98.91 b-np 12220 12203 11965 98.05 97.91 97.98 i-np 13660 13679 13384 97.84 97.98 97.91 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.27 98.26 98.27 Avg2. 46451 46451 45694 98.37 98.37 98.37 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12203 11745 96.25 96.11 96.18 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.25 96.11 96.18 Avg2. 12220 12203 11745 96.25 96.11 96.18 Current max chunk-based F1: 96.31 (iteration 72) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 75 Log-likelihood = -44812.923351 Norm (log-likelihood gradient vector) = 2259.904964 Norm (lambda vector) = 115.788815 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20595 20352 98.82 98.94 98.88 b-np 12220 12146 11935 98.26 97.67 97.96 i-np 13660 13710 13391 97.67 98.03 97.85 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.25 98.21 98.23 Avg2. 46451 46451 45678 98.34 98.34 98.34 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12146 11714 96.44 95.86 96.15 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.44 95.86 96.15 Avg2. 12220 12146 11714 96.44 95.86 96.15 Current max chunk-based F1: 96.31 (iteration 72) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 76 Log-likelihood = -43039.521137 Norm (log-likelihood gradient vector) = 5430.232551 Norm (lambda vector) = 118.476148 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20583 20355 98.89 98.95 98.92 b-np 12220 12179 11962 98.22 97.89 98.05 i-np 13660 13689 13391 97.82 98.03 97.93 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.31 98.29 98.30 Avg2. 46451 46451 45708 98.40 98.40 98.40 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12179 11743 96.42 96.10 96.26 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.42 96.10 96.26 Avg2. 12220 12179 11743 96.42 96.10 96.26 Current max chunk-based F1: 96.31 (iteration 72) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 77 Log-likelihood = -41971.090810 Norm (log-likelihood gradient vector) = 1938.318098 Norm (lambda vector) = 118.757116 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20593 20363 98.88 98.99 98.94 b-np 12220 12205 11975 98.12 98.00 98.06 i-np 13660 13653 13379 97.99 97.94 97.97 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.33 98.31 98.32 Avg2. 46451 46451 45717 98.42 98.42 98.42 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12205 11759 96.35 96.23 96.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.35 96.23 96.29 Avg2. 12220 12205 11759 96.35 96.23 96.29 Current max chunk-based F1: 96.31 (iteration 72) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 78 Log-likelihood = -41862.965098 Norm (log-likelihood gradient vector) = 1636.396557 Norm (lambda vector) = 118.711878 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20565 20354 98.97 98.95 98.96 b-np 12220 12193 11973 98.20 97.98 98.09 i-np 13660 13693 13402 97.87 98.11 97.99 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.35 98.35 98.35 Avg2. 46451 46451 45729 98.45 98.45 98.45 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12193 11763 96.47 96.26 96.37 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.47 96.26 96.37 Avg2. 12220 12193 11763 96.47 96.26 96.37 Current max chunk-based F1: 96.37 (iteration 78) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 79 Log-likelihood = -41430.969552 Norm (log-likelihood gradient vector) = 3130.559165 Norm (lambda vector) = 119.500312 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20596 20370 98.90 99.02 98.96 b-np 12220 12224 11985 98.04 98.08 98.06 i-np 13660 13631 13373 98.11 97.90 98.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.35 98.33 98.34 Avg2. 46451 46451 45728 98.44 98.44 98.44 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12224 11773 96.31 96.34 96.33 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.31 96.34 96.33 Avg2. 12220 12224 11773 96.31 96.34 96.33 Current max chunk-based F1: 96.37 (iteration 78) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 80 Log-likelihood = -41019.332819 Norm (log-likelihood gradient vector) = 2532.391938 Norm (lambda vector) = 120.212408 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20602 20374 98.89 99.04 98.97 b-np 12220 12217 11983 98.08 98.06 98.07 i-np 13660 13632 13375 98.11 97.91 98.01 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.36 98.34 98.35 Avg2. 46451 46451 45732 98.45 98.45 98.45 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12217 11772 96.36 96.33 96.35 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.36 96.33 96.35 Avg2. 12220 12217 11772 96.36 96.33 96.35 Current max chunk-based F1: 96.37 (iteration 78) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 81 Log-likelihood = -40265.109043 Norm (log-likelihood gradient vector) = 2165.183557 Norm (lambda vector) = 121.328546 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20613 20378 98.86 99.06 98.96 b-np 12220 12207 11974 98.09 97.99 98.04 i-np 13660 13631 13377 98.14 97.93 98.03 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.36 98.33 98.34 Avg2. 46451 46451 45729 98.45 98.45 98.45 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12207 11768 96.40 96.30 96.35 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.40 96.30 96.35 Avg2. 12220 12207 11768 96.40 96.30 96.35 Current max chunk-based F1: 96.37 (iteration 78) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 82 Log-likelihood = -38952.861442 Norm (log-likelihood gradient vector) = 3058.689159 Norm (lambda vector) = 123.770995 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20584 20365 98.94 99.00 98.97 b-np 12220 12182 11968 98.24 97.94 98.09 i-np 13660 13685 13410 97.99 98.17 98.08 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.39 98.37 98.38 Avg2. 46451 46451 45743 98.48 98.48 98.48 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12182 11766 96.59 96.28 96.43 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.59 96.28 96.43 Avg2. 12220 12182 11766 96.59 96.28 96.43 Current max chunk-based F1: 96.43 (iteration 82) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 83 Log-likelihood = -38566.533342 Norm (log-likelihood gradient vector) = 2903.847666 Norm (lambda vector) = 124.531221 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20577 20358 98.94 98.96 98.95 b-np 12220 12182 11965 98.22 97.91 98.07 i-np 13660 13692 13412 97.96 98.18 98.07 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.37 98.35 98.36 Avg2. 46451 46451 45735 98.46 98.46 98.46 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12182 11767 96.59 96.29 96.44 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.59 96.29 96.44 Avg2. 12220 12182 11767 96.59 96.29 96.44 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 84 Log-likelihood = -38472.957465 Norm (log-likelihood gradient vector) = 2107.651602 Norm (lambda vector) = 124.441981 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20581 20362 98.94 98.98 98.96 b-np 12220 12205 11980 98.16 98.04 98.10 i-np 13660 13665 13396 98.03 98.07 98.05 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.37 98.36 98.37 Avg2. 46451 46451 45738 98.47 98.47 98.47 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12205 11776 96.49 96.37 96.43 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.49 96.37 96.43 Avg2. 12220 12205 11776 96.49 96.37 96.43 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 85 Log-likelihood = -38530.910335 Norm (log-likelihood gradient vector) = 1536.104328 Norm (lambda vector) = 124.305506 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20584 20362 98.92 98.98 98.95 b-np 12220 12191 11971 98.20 97.96 98.08 i-np 13660 13676 13402 98.00 98.11 98.05 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.37 98.35 98.36 Avg2. 46451 46451 45735 98.46 98.46 98.46 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12191 11770 96.55 96.32 96.43 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.55 96.32 96.43 Avg2. 12220 12191 11770 96.55 96.32 96.43 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 86 Log-likelihood = -38437.388846 Norm (log-likelihood gradient vector) = 1727.277267 Norm (lambda vector) = 124.406690 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20582 20359 98.92 98.97 98.94 b-np 12220 12205 11975 98.12 98.00 98.06 i-np 13660 13664 13392 98.01 98.04 98.02 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.35 98.33 98.34 Avg2. 46451 46451 45726 98.44 98.44 98.44 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12205 11770 96.44 96.32 96.38 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.44 96.32 96.38 Avg2. 12220 12205 11770 96.44 96.32 96.38 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 87 Log-likelihood = -38379.659726 Norm (log-likelihood gradient vector) = 1806.445650 Norm (lambda vector) = 124.690079 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20583 20354 98.89 98.95 98.92 b-np 12220 12210 11970 98.03 97.95 97.99 i-np 13660 13658 13381 97.97 97.96 97.96 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.30 98.29 98.29 Avg2. 46451 46451 45705 98.39 98.39 98.39 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12210 11760 96.31 96.24 96.28 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.31 96.24 96.28 Avg2. 12220 12210 11760 96.31 96.24 96.28 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 88 Log-likelihood = -37875.320428 Norm (log-likelihood gradient vector) = 2118.438889 Norm (lambda vector) = 125.520376 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20513 20321 99.06 98.78 98.92 b-np 12220 12147 11938 98.28 97.69 97.98 i-np 13660 13791 13447 97.51 98.44 97.97 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.28 98.31 98.29 Avg2. 46451 46451 45706 98.40 98.40 98.40 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12147 11721 96.49 95.92 96.20 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.49 95.92 96.20 Avg2. 12220 12147 11721 96.49 95.92 96.20 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 89 Log-likelihood = -37329.265967 Norm (log-likelihood gradient vector) = 9432.889016 Norm (lambda vector) = 127.596669 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20570 20354 98.95 98.95 98.95 b-np 12220 12204 11967 98.06 97.93 97.99 i-np 13660 13677 13398 97.96 98.08 98.02 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.32 98.32 98.32 Avg2. 46451 46451 45719 98.42 98.42 98.42 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12204 11757 96.34 96.21 96.27 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.34 96.21 96.27 Avg2. 12220 12204 11757 96.34 96.21 96.27 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 90 Log-likelihood = -35453.043070 Norm (log-likelihood gradient vector) = 1986.037429 Norm (lambda vector) = 129.809962 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20590 20366 98.91 99.00 98.96 b-np 12220 12208 11973 98.08 97.98 98.03 i-np 13660 13653 13387 98.05 98.00 98.03 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.35 98.33 98.34 Avg2. 46451 46451 45726 98.44 98.44 98.44 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12208 11761 96.34 96.24 96.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.34 96.24 96.29 Avg2. 12220 12208 11761 96.34 96.24 96.29 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 91 Log-likelihood = -35075.691513 Norm (log-likelihood gradient vector) = 1536.861223 Norm (lambda vector) = 129.468320 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20588 20365 98.92 99.00 98.96 b-np 12220 12202 11971 98.11 97.96 98.03 i-np 13660 13661 13391 98.02 98.03 98.03 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.35 98.33 98.34 Avg2. 46451 46451 45727 98.44 98.44 98.44 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12202 11762 96.39 96.25 96.32 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.39 96.25 96.32 Avg2. 12220 12202 11762 96.39 96.25 96.32 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 92 Log-likelihood = -34581.359012 Norm (log-likelihood gradient vector) = 1428.122096 Norm (lambda vector) = 129.529129 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20599 20374 98.91 99.04 98.97 b-np 12220 12198 11970 98.13 97.95 98.04 i-np 13660 13654 13392 98.08 98.04 98.06 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.37 98.34 98.36 Avg2. 46451 46451 45736 98.46 98.46 98.46 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12198 11759 96.40 96.23 96.31 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.40 96.23 96.31 Avg2. 12220 12198 11759 96.40 96.23 96.31 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 93 Log-likelihood = -33788.937736 Norm (log-likelihood gradient vector) = 1587.596073 Norm (lambda vector) = 129.957082 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20576 20360 98.95 98.97 98.96 b-np 12220 12182 11965 98.22 97.91 98.07 i-np 13660 13693 13408 97.92 98.16 98.04 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.36 98.35 98.36 Avg2. 46451 46451 45733 98.45 98.45 98.45 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12182 11755 96.49 96.19 96.34 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.49 96.19 96.34 Avg2. 12220 12182 11755 96.49 96.19 96.34 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 94 Log-likelihood = -33087.706865 Norm (log-likelihood gradient vector) = 3515.249248 Norm (lambda vector) = 130.896323 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20580 20366 98.96 99.00 98.98 b-np 12220 12189 11969 98.20 97.95 98.07 i-np 13660 13682 13408 98.00 98.16 98.08 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.38 98.37 98.38 Avg2. 46451 46451 45743 98.48 98.48 98.48 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12189 11761 96.49 96.24 96.37 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.49 96.24 96.37 Avg2. 12220 12189 11761 96.49 96.24 96.37 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 95 Log-likelihood = -32782.705325 Norm (log-likelihood gradient vector) = 1974.504602 Norm (lambda vector) = 131.404424 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20592 20365 98.90 99.00 98.95 b-np 12220 12202 11966 98.07 97.92 97.99 i-np 13660 13657 13388 98.03 98.01 98.02 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.33 98.31 98.32 Avg2. 46451 46451 45719 98.42 98.42 98.42 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12202 11754 96.33 96.19 96.26 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.33 96.19 96.26 Avg2. 12220 12202 11754 96.33 96.19 96.26 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 96 Log-likelihood = -32304.631952 Norm (log-likelihood gradient vector) = 1766.137642 Norm (lambda vector) = 132.864355 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20590 20364 98.90 98.99 98.95 b-np 12220 12203 11968 98.07 97.94 98.01 i-np 13660 13658 13390 98.04 98.02 98.03 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.34 98.32 98.33 Avg2. 46451 46451 45722 98.43 98.43 98.43 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12203 11753 96.31 96.18 96.25 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.31 96.18 96.25 Avg2. 12220 12203 11753 96.31 96.18 96.25 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 97 Log-likelihood = -31890.016397 Norm (log-likelihood gradient vector) = 1779.455399 Norm (lambda vector) = 133.845427 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 27 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20599 20366 98.87 99.00 98.94 b-np 12220 12247 11985 97.86 98.08 97.97 i-np 13660 13605 13357 98.18 97.78 97.98 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.30 98.29 98.29 Avg2. 46451 46451 45708 98.40 98.40 98.40 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12247 11768 96.09 96.30 96.19 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.09 96.30 96.19 Avg2. 12220 12247 11768 96.09 96.30 96.19 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 27 seconds Iteration: 98 Log-likelihood = -30546.115244 Norm (log-likelihood gradient vector) = 4632.304506 Norm (lambda vector) = 138.239708 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20536 20332 99.01 98.84 98.92 b-np 12220 12171 11951 98.19 97.80 98.00 i-np 13660 13744 13427 97.69 98.29 97.99 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.30 98.31 98.30 Avg2. 46451 46451 45710 98.40 98.40 98.40 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12171 11735 96.42 96.03 96.22 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.42 96.03 96.22 Avg2. 12220 12171 11735 96.42 96.03 96.22 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 99 Log-likelihood = -30167.441597 Norm (log-likelihood gradient vector) = 4528.622554 Norm (lambda vector) = 140.067794 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20576 20358 98.94 98.96 98.95 b-np 12220 12197 11966 98.11 97.92 98.01 i-np 13660 13678 13403 97.99 98.12 98.05 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.35 98.33 98.34 Avg2. 46451 46451 45727 98.44 98.44 98.44 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12197 11756 96.38 96.20 96.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.38 96.20 96.29 Avg2. 12220 12197 11756 96.38 96.20 96.29 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 100 Log-likelihood = -30136.798996 Norm (log-likelihood gradient vector) = 1507.396645 Norm (lambda vector) = 138.820881 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20579 20360 98.94 98.97 98.96 b-np 12220 12202 11969 98.09 97.95 98.02 i-np 13660 13670 13398 98.01 98.08 98.05 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.35 98.33 98.34 Avg2. 46451 46451 45727 98.44 98.44 98.44 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12202 11756 96.34 96.20 96.27 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.34 96.20 96.27 Avg2. 12220 12202 11756 96.34 96.20 96.27 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 101 Log-likelihood = -30074.879389 Norm (log-likelihood gradient vector) = 1164.477653 Norm (lambda vector) = 138.992632 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20578 20359 98.94 98.97 98.95 b-np 12220 12198 11971 98.14 97.96 98.05 i-np 13660 13675 13396 97.96 98.07 98.01 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.34 98.33 98.34 Avg2. 46451 46451 45726 98.44 98.44 98.44 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12198 11757 96.38 96.21 96.30 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.38 96.21 96.30 Avg2. 12220 12198 11757 96.38 96.21 96.30 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 102 Log-likelihood = -29979.078933 Norm (log-likelihood gradient vector) = 1931.079530 Norm (lambda vector) = 139.949120 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20584 20364 98.93 98.99 98.96 b-np 12220 12209 11979 98.12 98.03 98.07 i-np 13660 13658 13392 98.05 98.04 98.05 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.37 98.35 98.36 Avg2. 46451 46451 45735 98.46 98.46 98.46 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12209 11769 96.40 96.31 96.35 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.40 96.31 96.35 Avg2. 12220 12209 11769 96.40 96.31 96.35 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 103 Log-likelihood = -29803.222631 Norm (log-likelihood gradient vector) = 2022.034069 Norm (lambda vector) = 141.066373 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20589 20352 98.85 98.94 98.89 b-np 12220 12194 11961 98.09 97.88 97.98 i-np 13660 13668 13376 97.86 97.92 97.89 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.27 98.25 98.26 Avg2. 46451 46451 45689 98.36 98.36 98.36 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12194 11739 96.27 96.06 96.17 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.27 96.06 96.17 Avg2. 12220 12194 11739 96.27 96.06 96.17 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 104 Log-likelihood = -29090.709557 Norm (log-likelihood gradient vector) = 3743.033851 Norm (lambda vector) = 145.960676 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20630 20382 98.80 99.08 98.94 b-np 12220 12227 11981 97.99 98.04 98.02 i-np 13660 13594 13352 98.22 97.75 97.98 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.34 98.29 98.31 Avg2. 46451 46451 45715 98.42 98.42 98.42 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12227 11768 96.25 96.30 96.27 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.25 96.30 96.27 Avg2. 12220 12227 11768 96.25 96.30 96.27 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 105 Log-likelihood = -28694.602906 Norm (log-likelihood gradient vector) = 3532.732963 Norm (lambda vector) = 146.423031 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20602 20367 98.86 99.01 98.93 b-np 12220 12219 11976 98.01 98.00 98.01 i-np 13660 13630 13369 98.09 97.87 97.98 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.32 98.29 98.31 Avg2. 46451 46451 45712 98.41 98.41 98.41 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12219 11760 96.24 96.24 96.24 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.24 96.24 96.24 Avg2. 12220 12219 11760 96.24 96.24 96.24 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 106 Log-likelihood = -28323.889513 Norm (log-likelihood gradient vector) = 1523.750468 Norm (lambda vector) = 145.772347 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20582 20360 98.92 98.97 98.95 b-np 12220 12204 11976 98.13 98.00 98.07 i-np 13660 13665 13389 97.98 98.02 98.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.34 98.33 98.34 Avg2. 46451 46451 45725 98.44 98.44 98.44 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12204 11760 96.36 96.24 96.30 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.36 96.24 96.30 Avg2. 12220 12204 11760 96.36 96.24 96.30 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 107 Log-likelihood = -27959.476685 Norm (log-likelihood gradient vector) = 1119.780872 Norm (lambda vector) = 146.041343 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20571 20356 98.95 98.95 98.95 b-np 12220 12194 11973 98.19 97.98 98.08 i-np 13660 13686 13402 97.92 98.11 98.02 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.36 98.35 98.35 Avg2. 46451 46451 45731 98.45 98.45 98.45 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12194 11760 96.44 96.24 96.34 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.44 96.24 96.34 Avg2. 12220 12194 11760 96.44 96.24 96.34 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 108 Log-likelihood = -27570.857737 Norm (log-likelihood gradient vector) = 1384.798645 Norm (lambda vector) = 146.540251 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20576 20361 98.96 98.98 98.97 b-np 12220 12203 11974 98.12 97.99 98.06 i-np 13660 13672 13403 98.03 98.12 98.08 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.37 98.36 98.37 Avg2. 46451 46451 45738 98.47 98.47 98.47 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12203 11766 96.42 96.28 96.35 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.42 96.28 96.35 Avg2. 12220 12203 11766 96.42 96.28 96.35 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 109 Log-likelihood = -26753.025136 Norm (log-likelihood gradient vector) = 1469.426254 Norm (lambda vector) = 148.002048 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20472 20269 99.01 98.53 98.77 b-np 12220 12085 11885 98.35 97.26 97.80 i-np 13660 13894 13443 96.75 98.41 97.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.04 98.07 98.05 Avg2. 46451 46451 45597 98.16 98.16 98.16 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12085 11632 96.25 95.19 95.72 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.25 95.19 95.72 Avg2. 12220 12085 11632 96.25 95.19 95.72 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 110 Log-likelihood = -27417.193404 Norm (log-likelihood gradient vector) = 12219.176284 Norm (lambda vector) = 150.504444 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20535 20328 98.99 98.82 98.91 b-np 12220 12158 11946 98.26 97.76 98.01 i-np 13660 13758 13427 97.59 98.29 97.94 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.28 98.29 98.29 Avg2. 46451 46451 45701 98.39 98.39 98.39 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12158 11725 96.44 95.95 96.19 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.44 95.95 96.19 Avg2. 12220 12158 11725 96.44 95.95 96.19 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 111 Log-likelihood = -26494.141537 Norm (log-likelihood gradient vector) = 4546.592348 Norm (lambda vector) = 148.898054 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20576 20354 98.92 98.95 98.93 b-np 12220 12191 11965 98.15 97.91 98.03 i-np 13660 13684 13398 97.91 98.08 98.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.33 98.31 98.32 Avg2. 46451 46451 45717 98.42 98.42 98.42 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12191 11756 96.43 96.20 96.32 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.43 96.20 96.32 Avg2. 12220 12191 11756 96.43 96.20 96.32 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 112 Log-likelihood = -25862.271912 Norm (log-likelihood gradient vector) = 1815.979112 Norm (lambda vector) = 149.960916 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20593 20365 98.89 99.00 98.95 b-np 12220 12209 11972 98.06 97.97 98.01 i-np 13660 13649 13383 98.05 97.97 98.01 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.33 98.31 98.32 Avg2. 46451 46451 45720 98.43 98.43 98.43 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12209 11758 96.31 96.22 96.26 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.31 96.22 96.26 Avg2. 12220 12209 11758 96.31 96.22 96.26 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 113 Log-likelihood = -25585.902286 Norm (log-likelihood gradient vector) = 1037.077851 Norm (lambda vector) = 150.520898 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20596 20363 98.87 98.99 98.93 b-np 12220 12219 11975 98.00 98.00 98.00 i-np 13660 13636 13370 98.05 97.88 97.96 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.31 98.29 98.30 Avg2. 46451 46451 45708 98.40 98.40 98.40 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12219 11755 96.20 96.19 96.20 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.20 96.19 96.20 Avg2. 12220 12219 11755 96.20 96.19 96.20 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 114 Log-likelihood = -25216.622369 Norm (log-likelihood gradient vector) = 1250.453566 Norm (lambda vector) = 151.159515 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20600 20365 98.86 99.00 98.93 b-np 12220 12215 11973 98.02 97.98 98.00 i-np 13660 13636 13369 98.04 97.87 97.96 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.31 98.28 98.29 Avg2. 46451 46451 45707 98.40 98.40 98.40 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12215 11753 96.22 96.18 96.20 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.22 96.18 96.20 Avg2. 12220 12215 11753 96.22 96.18 96.20 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 115 Log-likelihood = -24741.696986 Norm (log-likelihood gradient vector) = 1412.884112 Norm (lambda vector) = 151.860775 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20583 20359 98.91 98.97 98.94 b-np 12220 12233 11983 97.96 98.06 98.01 i-np 13660 13635 13372 98.07 97.89 97.98 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.31 98.31 98.31 Avg2. 46451 46451 45714 98.41 98.41 98.41 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12233 11760 96.13 96.24 96.18 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.13 96.24 96.18 Avg2. 12220 12233 11760 96.13 96.24 96.18 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 116 Log-likelihood = -23828.640788 Norm (log-likelihood gradient vector) = 2369.802023 Norm (lambda vector) = 153.582375 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20585 20357 98.89 98.96 98.93 b-np 12220 12193 11965 98.13 97.91 98.02 i-np 13660 13673 13392 97.94 98.04 97.99 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.32 98.30 98.31 Avg2. 46451 46451 45714 98.41 98.41 98.41 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12193 11755 96.41 96.19 96.30 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.41 96.19 96.30 Avg2. 12220 12193 11755 96.41 96.19 96.30 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 117 Log-likelihood = -23060.442104 Norm (log-likelihood gradient vector) = 2114.037169 Norm (lambda vector) = 155.626175 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20595 20364 98.88 98.99 98.94 b-np 12220 12205 11972 98.09 97.97 98.03 i-np 13660 13651 13386 98.06 97.99 98.03 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.34 98.32 98.33 Avg2. 46451 46451 45722 98.43 98.43 98.43 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12205 11765 96.39 96.28 96.34 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.39 96.28 96.34 Avg2. 12220 12205 11765 96.39 96.28 96.34 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 118 Log-likelihood = -23078.225072 Norm (log-likelihood gradient vector) = 1077.066692 Norm (lambda vector) = 155.068588 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20587 20359 98.89 98.97 98.93 b-np 12220 12198 11968 98.11 97.94 98.03 i-np 13660 13666 13389 97.97 98.02 97.99 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.33 98.31 98.32 Avg2. 46451 46451 45716 98.42 98.42 98.42 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12198 11758 96.39 96.22 96.31 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.39 96.22 96.31 Avg2. 12220 12198 11758 96.39 96.22 96.31 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 25 seconds Iteration: 119 Log-likelihood = -23036.164761 Norm (log-likelihood gradient vector) = 1616.637332 Norm (lambda vector) = 155.469060 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20587 20364 98.92 98.99 98.96 b-np 12220 12204 11975 98.12 98.00 98.06 i-np 13660 13660 13395 98.06 98.06 98.06 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.37 98.35 98.36 Avg2. 46451 46451 45734 98.46 98.46 98.46 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12204 11769 96.44 96.31 96.37 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.44 96.31 96.37 Avg2. 12220 12204 11769 96.44 96.31 96.37 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 27 seconds Iteration: 120 Log-likelihood = -23005.629385 Norm (log-likelihood gradient vector) = 1037.218343 Norm (lambda vector) = 155.239849 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20591 20368 98.92 99.01 98.97 b-np 12220 12205 11975 98.12 98.00 98.06 i-np 13660 13655 13394 98.09 98.05 98.07 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.37 98.35 98.36 Avg2. 46451 46451 45737 98.46 98.46 98.46 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12205 11769 96.43 96.31 96.37 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.43 96.31 96.37 Avg2. 12220 12205 11769 96.43 96.31 96.37 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 25 seconds Iteration: 121 Log-likelihood = -22924.688800 Norm (log-likelihood gradient vector) = 1045.836225 Norm (lambda vector) = 155.452760 Log-likelihood and gradient computational time: 26 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20509 20316 99.06 98.76 98.91 b-np 12220 12151 11944 98.30 97.74 98.02 i-np 13660 13791 13441 97.46 98.40 97.93 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.27 98.30 98.29 Avg2. 46451 46451 45701 98.39 98.39 98.39 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12151 11722 96.47 95.92 96.20 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.47 95.92 96.20 Avg2. 12220 12151 11722 96.47 95.92 96.20 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 27 seconds Iteration: 122 Log-likelihood = -22822.875658 Norm (log-likelihood gradient vector) = 6167.410197 Norm (lambda vector) = 157.572929 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 25 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20547 20345 99.02 98.90 98.96 b-np 12220 12181 11969 98.26 97.95 98.10 i-np 13660 13723 13425 97.83 98.28 98.05 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.37 98.38 98.37 Avg2. 46451 46451 45739 98.47 98.47 98.47 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12181 11758 96.53 96.22 96.37 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.53 96.22 96.37 Avg2. 12220 12181 11758 96.53 96.22 96.37 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 25 seconds Iteration: 123 Log-likelihood = -22673.870398 Norm (log-likelihood gradient vector) = 3236.996189 Norm (lambda vector) = 156.500550 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20592 20366 98.90 99.00 98.95 b-np 12220 12213 11979 98.08 98.03 98.06 i-np 13660 13646 13389 98.12 98.02 98.07 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.37 98.35 98.36 Avg2. 46451 46451 45734 98.46 98.46 98.46 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12213 11773 96.40 96.34 96.37 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.40 96.34 96.37 Avg2. 12220 12213 11773 96.40 96.34 96.37 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 124 Log-likelihood = -22236.505928 Norm (log-likelihood gradient vector) = 1232.685037 Norm (lambda vector) = 157.908202 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20600 20367 98.87 99.01 98.94 b-np 12220 12221 11979 98.02 98.03 98.02 i-np 13660 13630 13376 98.14 97.92 98.03 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.34 98.32 98.33 Avg2. 46451 46451 45722 98.43 98.43 98.43 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12221 11772 96.33 96.33 96.33 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.33 96.33 96.33 Avg2. 12220 12221 11772 96.33 96.33 96.33 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 125 Log-likelihood = -21950.022801 Norm (log-likelihood gradient vector) = 1062.858234 Norm (lambda vector) = 158.732500 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20585 20365 98.93 99.00 98.96 b-np 12220 12223 11984 98.04 98.07 98.06 i-np 13660 13643 13387 98.12 98.00 98.06 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.37 98.36 98.36 Avg2. 46451 46451 45736 98.46 98.46 98.46 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12223 11774 96.33 96.35 96.34 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.33 96.35 96.34 Avg2. 12220 12223 11774 96.33 96.35 96.34 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 126 Log-likelihood = -21507.324737 Norm (log-likelihood gradient vector) = 1152.902126 Norm (lambda vector) = 159.934537 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20586 20363 98.92 98.99 98.95 b-np 12220 12214 11981 98.09 98.04 98.07 i-np 13660 13651 13385 98.05 97.99 98.02 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.35 98.34 98.35 Avg2. 46451 46451 45729 98.45 98.45 98.45 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12214 11772 96.38 96.33 96.36 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.38 96.33 96.36 Avg2. 12220 12214 11772 96.38 96.33 96.36 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 127 Log-likelihood = -21040.403393 Norm (log-likelihood gradient vector) = 1248.175803 Norm (lambda vector) = 161.345208 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20604 20364 98.84 98.99 98.91 b-np 12220 12234 11983 97.95 98.06 98.00 i-np 13660 13613 13359 98.13 97.80 97.97 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.31 98.28 98.29 Avg2. 46451 46451 45706 98.40 98.40 98.40 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12234 11776 96.26 96.37 96.31 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.26 96.37 96.31 Avg2. 12220 12234 11776 96.26 96.37 96.31 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 128 Log-likelihood = -20981.604777 Norm (log-likelihood gradient vector) = 4271.143738 Norm (lambda vector) = 164.166990 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20596 20367 98.89 99.01 98.95 b-np 12220 12226 11985 98.03 98.08 98.05 i-np 13660 13629 13374 98.13 97.91 98.02 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.35 98.33 98.34 Avg2. 46451 46451 45726 98.44 98.44 98.44 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12226 11778 96.34 96.38 96.36 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.34 96.38 96.36 Avg2. 12220 12226 11778 96.34 96.38 96.36 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 129 Log-likelihood = -20829.270870 Norm (log-likelihood gradient vector) = 1729.878175 Norm (lambda vector) = 162.626838 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20584 20356 98.89 98.95 98.92 b-np 12220 12206 11975 98.11 98.00 98.05 i-np 13660 13661 13387 97.99 98.00 98.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.33 98.32 98.32 Avg2. 46451 46451 45718 98.42 98.42 98.42 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12206 11770 96.43 96.32 96.37 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.43 96.32 96.37 Avg2. 12220 12206 11770 96.43 96.32 96.37 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds Iteration: 130 Log-likelihood = -20688.334720 Norm (log-likelihood gradient vector) = 917.041121 Norm (lambda vector) = 163.237548 Log-likelihood and gradient computational time: 25 seconds Training iteration elapsed: 26 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- o 20571 20584 20356 98.89 98.95 98.92 b-np 12220 12210 11977 98.09 98.01 98.05 i-np 13660 13657 13384 98.00 97.98 97.99 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.33 98.32 98.32 Avg2. 46451 46451 45717 98.42 98.42 98.42 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12210 11771 96.40 96.33 96.37 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.40 96.33 96.37 Avg2. 12220 12210 11771 96.40 96.33 96.37 Current max chunk-based F1: 96.44 (iteration 83) Training iteration elapsed (including evaluation time): 26 seconds The training process elapsed: 3376 seconds