OPTION VALUES: Model directory: ./Fold11-IOB2/ Training data file: train.tagged Testing data file: test.tagged Unlabeled data file: data.untagged Model file: model.txt Training log file (this one): trainlog.txt Label representation: IOB2 Second-order Markov CRFs Number of labels: 4 Number of training sequences: 1044 (one data partition) Number of testing sequences: 50 (one data partition) Number of unlabeled sequences: 0 Number of context predicates: 800101 Number of features: 1525661 Feature rare threshold: 1 Context predicate rare threshold: 1 Using multiple rare thresholds for features: 0 Highlight feature: 0 Number of training iterations: 150 Initial lambda value: 0.0000 Sigma square (for smoothing): 100.0000 Epsilon for L-BFGS convergence: 0.000100 Number of approximated hessian matrixes: 7 Start to train ... Iteration: 1 Log-likelihood = -3107412.081515 Norm (log-likelihood gradient vector) = 641419.414458 Norm (lambda vector) = 0.000000 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 12148 9687 79.74 70.43 74.79 i-np 16033 15762 12797 81.19 79.82 80.50 o 23216 25094 20773 82.78 89.48 86.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.24 79.91 80.57 Avg2. 53004 53004 43257 81.61 81.61 81.61 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 12148 7532 62.00 54.76 58.16 ----- ------ ----- ----- ------- ------- ------------- Avg1. 62.00 54.76 58.16 Avg2. 13755 12148 7532 62.00 54.76 58.16 Current max chunk-based F1: 58.16 (iteration 1) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 2 Log-likelihood = -2499342.355219 Norm (log-likelihood gradient vector) = 561464.532285 Norm (lambda vector) = 1.000000 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 12491 10153 81.28 73.81 77.37 i-np 16033 16808 13600 80.91 84.83 82.82 o 23216 23705 20649 87.11 88.94 88.02 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.10 82.53 82.81 Avg2. 53004 53004 44402 83.77 83.77 83.77 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 12491 8078 64.67 58.73 61.56 ----- ------ ----- ----- ------- ------- ------------- Avg1. 64.67 58.73 61.56 Avg2. 13755 12491 8078 64.67 58.73 61.56 Current max chunk-based F1: 61.56 (iteration 2) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 3 Log-likelihood = -991861.823012 Norm (log-likelihood gradient vector) = 383381.240237 Norm (lambda vector) = 10.185181 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 11378 10548 92.71 76.68 83.94 i-np 16033 17710 14921 84.25 93.06 88.44 o 23216 23916 21951 91.78 94.55 93.15 ----- ------ ----- ----- ------- ------- ------------- Avg1. 89.58 88.10 88.83 Avg2. 53004 53004 47420 89.46 89.46 89.46 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 11378 9095 79.93 66.12 72.37 ----- ------ ----- ----- ------- ------- ------------- Avg1. 79.93 66.12 72.37 Avg2. 13755 11378 9095 79.93 66.12 72.37 Current max chunk-based F1: 72.37 (iteration 3) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 4 Log-likelihood = -718970.529904 Norm (log-likelihood gradient vector) = 169577.114311 Norm (lambda vector) = 9.817675 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13330 12018 90.16 87.37 88.74 i-np 16033 15752 14426 91.58 89.98 90.77 o 23216 23922 22405 93.66 96.51 95.06 ----- ------ ----- ----- ------- ------- ------------- Avg1. 91.80 91.29 91.54 Avg2. 53004 53004 48849 92.16 92.16 92.16 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13330 10932 82.01 79.48 80.72 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.01 79.48 80.72 Avg2. 13755 13330 10932 82.01 79.48 80.72 Current max chunk-based F1: 80.72 (iteration 4) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 5 Log-likelihood = -663377.842465 Norm (log-likelihood gradient vector) = 107531.003201 Norm (lambda vector) = 9.353435 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13494 12349 91.51 89.78 90.64 i-np 16033 15264 14387 94.25 89.73 91.94 o 23216 24246 22638 93.37 97.51 95.39 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.05 92.34 92.69 Avg2. 53004 53004 49374 93.15 93.15 93.15 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13494 11506 85.27 83.65 84.45 ----- ------ ----- ----- ------- ------- ------------- Avg1. 85.27 83.65 84.45 Avg2. 13755 13494 11506 85.27 83.65 84.45 Current max chunk-based F1: 84.45 (iteration 5) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 6 Log-likelihood = -627793.275981 Norm (log-likelihood gradient vector) = 73674.734459 Norm (lambda vector) = 9.402242 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13405 12464 92.98 90.61 91.78 i-np 16033 15571 14719 94.53 91.80 93.15 o 23216 24028 22704 94.49 97.79 96.11 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.00 93.40 93.70 Avg2. 53004 53004 49887 94.12 94.12 94.12 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13405 11783 87.90 85.66 86.77 ----- ------ ----- ----- ------- ------- ------------- Avg1. 87.90 85.66 86.77 Avg2. 13755 13405 11783 87.90 85.66 86.77 Current max chunk-based F1: 86.77 (iteration 6) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 7 Log-likelihood = -574055.638893 Norm (log-likelihood gradient vector) = 71329.127886 Norm (lambda vector) = 10.323663 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13492 12706 94.17 92.37 93.27 i-np 16033 15481 14828 95.78 92.48 94.10 o 23216 24031 22833 95.01 98.35 96.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.99 94.40 94.70 Avg2. 53004 53004 50367 95.02 95.02 95.02 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13492 12098 89.67 87.95 88.80 ----- ------ ----- ----- ------- ------- ------------- Avg1. 89.67 87.95 88.80 Avg2. 13755 13492 12098 89.67 87.95 88.80 Current max chunk-based F1: 88.80 (iteration 7) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 8 Log-likelihood = -496259.055173 Norm (log-likelihood gradient vector) = 59493.955406 Norm (lambda vector) = 12.167226 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13257 12691 95.73 92.26 93.97 i-np 16033 16568 15380 92.83 95.93 94.35 o 23216 23179 22504 97.09 96.93 97.01 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.22 95.04 95.13 Avg2. 53004 53004 50575 95.42 95.42 95.42 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13257 12025 90.71 87.42 89.03 ----- ------ ----- ----- ------- ------- ------------- Avg1. 90.71 87.42 89.03 Avg2. 13755 13257 12025 90.71 87.42 89.03 Current max chunk-based F1: 89.03 (iteration 8) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 9 Log-likelihood = -405236.013740 Norm (log-likelihood gradient vector) = 69584.224858 Norm (lambda vector) = 16.729181 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13526 12909 95.44 93.85 94.64 i-np 16033 15822 15138 95.68 94.42 95.04 o 23216 23656 22811 96.43 98.26 97.33 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.85 95.51 95.68 Avg2. 53004 53004 50858 95.95 95.95 95.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13526 12362 91.39 89.87 90.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 91.39 89.87 90.63 Avg2. 13755 13526 12362 91.39 89.87 90.63 Current max chunk-based F1: 90.63 (iteration 9) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 10 Log-likelihood = -367323.970062 Norm (log-likelihood gradient vector) = 30584.138894 Norm (lambda vector) = 18.187758 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13650 13016 95.36 94.63 94.99 i-np 16033 15627 15078 96.49 94.04 95.25 o 23216 23727 22867 96.38 98.50 97.42 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.07 95.72 95.90 Avg2. 53004 53004 50961 96.15 96.15 96.15 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13650 12505 91.61 90.91 91.26 ----- ------ ----- ----- ------- ------- ------------- Avg1. 91.61 90.91 91.26 Avg2. 13755 13650 12505 91.61 90.91 91.26 Current max chunk-based F1: 91.26 (iteration 10) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 11 Log-likelihood = -342639.277282 Norm (log-likelihood gradient vector) = 30867.495547 Norm (lambda vector) = 20.266179 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13750 13100 95.27 95.24 95.26 i-np 16033 15516 15036 96.91 93.78 95.32 o 23216 23738 22891 96.43 98.60 97.50 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.20 95.87 96.04 Avg2. 53004 53004 51027 96.27 96.27 96.27 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13750 12591 91.57 91.54 91.55 ----- ------ ----- ----- ------- ------- ------------- Avg1. 91.57 91.54 91.55 Avg2. 13755 13750 12591 91.57 91.54 91.55 Current max chunk-based F1: 91.55 (iteration 11) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 12 Log-likelihood = -318580.267036 Norm (log-likelihood gradient vector) = 27724.416646 Norm (lambda vector) = 22.728027 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13364 12950 96.90 94.15 95.50 i-np 16033 16420 15478 94.26 96.54 95.39 o 23216 23220 22663 97.60 97.62 97.61 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.26 96.10 96.18 Avg2. 53004 53004 51091 96.39 96.39 96.39 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13364 12411 92.87 90.23 91.53 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.87 90.23 91.53 Avg2. 13755 13364 12411 92.87 90.23 91.53 Current max chunk-based F1: 91.55 (iteration 11) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 13 Log-likelihood = -316815.713918 Norm (log-likelihood gradient vector) = 93878.677241 Norm (lambda vector) = 27.788211 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13626 13116 96.26 95.35 95.80 i-np 16033 15924 15305 96.11 95.46 95.78 o 23216 23454 22810 97.25 98.25 97.75 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.54 96.35 96.45 Avg2. 53004 53004 51231 96.65 96.65 96.65 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13626 12650 92.84 91.97 92.40 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.84 91.97 92.40 Avg2. 13755 13626 12650 92.84 91.97 92.40 Current max chunk-based F1: 92.40 (iteration 13) Training iteration elapsed (including evaluation time): 33 seconds Iteration: 14 Log-likelihood = -288557.877316 Norm (log-likelihood gradient vector) = 25902.658521 Norm (lambda vector) = 28.195554 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13686 13154 96.11 95.63 95.87 i-np 16033 15814 15262 96.51 95.19 95.85 o 23216 23504 22840 97.17 98.38 97.77 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.60 96.40 96.50 Avg2. 53004 53004 51256 96.70 96.70 96.70 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13686 12702 92.81 92.34 92.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.81 92.34 92.58 Avg2. 13755 13686 12702 92.81 92.34 92.58 Current max chunk-based F1: 92.58 (iteration 14) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 15 Log-likelihood = -285180.466572 Norm (log-likelihood gradient vector) = 17582.806356 Norm (lambda vector) = 28.377971 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13736 13191 96.03 95.90 95.97 i-np 16033 15736 15228 96.77 94.98 95.87 o 23216 23532 22861 97.15 98.47 97.81 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.65 96.45 96.55 Avg2. 53004 53004 51280 96.75 96.75 96.75 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13736 12741 92.76 92.63 92.69 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.76 92.63 92.69 Avg2. 13755 13736 12741 92.76 92.63 92.69 Current max chunk-based F1: 92.69 (iteration 15) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 16 Log-likelihood = -280566.042716 Norm (log-likelihood gradient vector) = 18988.970858 Norm (lambda vector) = 28.906382 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13753 13220 96.12 96.11 96.12 i-np 16033 15706 15234 96.99 95.02 96.00 o 23216 23545 22879 97.17 98.55 97.86 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.76 96.56 96.66 Avg2. 53004 53004 51333 96.85 96.85 96.85 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13753 12791 93.01 92.99 93.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.01 92.99 93.00 Avg2. 13755 13753 12791 93.01 92.99 93.00 Current max chunk-based F1: 93.00 (iteration 16) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 17 Log-likelihood = -268029.977757 Norm (log-likelihood gradient vector) = 18896.966101 Norm (lambda vector) = 30.599325 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13694 13224 96.57 96.14 96.35 i-np 16033 15928 15404 96.71 96.08 96.39 o 23216 23382 22844 97.70 98.40 98.05 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.99 96.87 96.93 Avg2. 53004 53004 51472 97.11 97.11 97.11 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13694 12827 93.67 93.25 93.46 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.67 93.25 93.46 Avg2. 13755 13694 12827 93.67 93.25 93.46 Current max chunk-based F1: 93.46 (iteration 17) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 18 Log-likelihood = -242974.375704 Norm (log-likelihood gradient vector) = 29663.523618 Norm (lambda vector) = 34.086473 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13920 13336 95.80 96.95 96.38 i-np 16033 15424 15106 97.94 94.22 96.04 o 23216 23660 22949 96.99 98.85 97.91 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.91 96.67 96.79 Avg2. 53004 53004 51391 96.96 96.96 96.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13920 12906 92.72 93.83 93.27 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.72 93.83 93.27 Avg2. 13755 13920 12906 92.72 93.83 93.27 Current max chunk-based F1: 93.46 (iteration 17) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 19 Log-likelihood = -237965.692205 Norm (log-likelihood gradient vector) = 53528.761685 Norm (lambda vector) = 40.055684 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13734 13269 96.61 96.47 96.54 i-np 16033 15799 15354 97.18 95.76 96.47 o 23216 23471 22891 97.53 98.60 98.06 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.11 96.94 97.03 Avg2. 53004 53004 51514 97.19 97.19 97.19 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13734 12880 93.78 93.64 93.71 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.78 93.64 93.71 Avg2. 13755 13734 12880 93.78 93.64 93.71 Current max chunk-based F1: 93.71 (iteration 19) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 20 Log-likelihood = -219566.928312 Norm (log-likelihood gradient vector) = 17202.024589 Norm (lambda vector) = 40.031062 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13671 13234 96.80 96.21 96.51 i-np 16033 15940 15417 96.72 96.16 96.44 o 23216 23393 22851 97.68 98.43 98.05 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.07 96.93 97.00 Avg2. 53004 53004 51502 97.17 97.17 97.17 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13671 12841 93.93 93.36 93.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 93.93 93.36 93.64 Avg2. 13755 13671 12841 93.93 93.36 93.64 Current max chunk-based F1: 93.71 (iteration 19) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 21 Log-likelihood = -213023.408173 Norm (log-likelihood gradient vector) = 11121.214694 Norm (lambda vector) = 40.347441 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13629 13218 96.98 96.10 96.54 i-np 16033 16041 15470 96.44 96.49 96.46 o 23216 23334 22829 97.84 98.33 98.08 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.09 96.97 97.03 Avg2. 53004 53004 51517 97.19 97.19 97.19 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13629 12836 94.18 93.32 93.75 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.18 93.32 93.75 Avg2. 13755 13629 12836 94.18 93.32 93.75 Current max chunk-based F1: 93.75 (iteration 21) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 22 Log-likelihood = -205635.523409 Norm (log-likelihood gradient vector) = 10593.674434 Norm (lambda vector) = 41.076209 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13644 13237 97.02 96.23 96.62 i-np 16033 16034 15484 96.57 96.58 96.57 o 23216 23326 22839 97.91 98.38 98.14 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.17 97.06 97.11 Avg2. 53004 53004 51560 97.28 97.28 97.28 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13644 12867 94.31 93.54 93.92 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.31 93.54 93.92 Avg2. 13755 13644 12867 94.31 93.54 93.92 Current max chunk-based F1: 93.92 (iteration 22) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 23 Log-likelihood = -196281.235720 Norm (log-likelihood gradient vector) = 10053.917840 Norm (lambda vector) = 42.379244 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13460 13114 97.43 95.34 96.37 i-np 16033 16478 15640 94.91 97.55 96.21 o 23216 23066 22679 98.32 97.69 98.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.89 96.86 96.87 Avg2. 53004 53004 51433 97.04 97.04 97.04 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13460 12696 94.32 92.30 93.30 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.32 92.30 93.30 Avg2. 13755 13460 12696 94.32 92.30 93.30 Current max chunk-based F1: 93.92 (iteration 22) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 24 Log-likelihood = -193524.762305 Norm (log-likelihood gradient vector) = 55875.615571 Norm (lambda vector) = 46.460558 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13565 13194 97.27 95.92 96.59 i-np 16033 16207 15556 95.98 97.02 96.50 o 23216 23232 22780 98.05 98.12 98.09 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.10 97.02 97.06 Avg2. 53004 53004 51530 97.22 97.22 97.22 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13565 12818 94.49 93.19 93.84 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.49 93.19 93.84 Avg2. 13755 13565 12818 94.49 93.19 93.84 Current max chunk-based F1: 93.92 (iteration 22) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 25 Log-likelihood = -191194.205008 Norm (log-likelihood gradient vector) = 20435.164366 Norm (lambda vector) = 43.782316 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13648 13248 97.07 96.31 96.69 i-np 16033 16042 15511 96.69 96.74 96.72 o 23216 23314 22840 97.97 98.38 98.17 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.24 97.15 97.19 Avg2. 53004 53004 51599 97.35 97.35 97.35 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13648 12905 94.56 93.82 94.19 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.56 93.82 94.19 Avg2. 13755 13648 12905 94.56 93.82 94.19 Current max chunk-based F1: 94.19 (iteration 25) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 26 Log-likelihood = -185180.038815 Norm (log-likelihood gradient vector) = 11731.974893 Norm (lambda vector) = 45.078612 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13715 13273 96.78 96.50 96.64 i-np 16033 15937 15440 96.88 96.30 96.59 o 23216 23352 22849 97.85 98.42 98.13 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.17 97.07 97.12 Avg2. 53004 53004 51562 97.28 97.28 97.28 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13715 12916 94.17 93.90 94.04 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.17 93.90 94.04 Avg2. 13755 13715 12916 94.17 93.90 94.04 Current max chunk-based F1: 94.19 (iteration 25) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 27 Log-likelihood = -178557.573133 Norm (log-likelihood gradient vector) = 9799.135607 Norm (lambda vector) = 47.876353 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13717 13289 96.88 96.61 96.75 i-np 16033 15939 15461 97.00 96.43 96.72 o 23216 23348 22855 97.89 98.45 98.17 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.26 97.16 97.21 Avg2. 53004 53004 51605 97.36 97.36 97.36 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13717 12949 94.40 94.14 94.27 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.40 94.14 94.27 Avg2. 13755 13717 12949 94.40 94.14 94.27 Current max chunk-based F1: 94.27 (iteration 27) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 28 Log-likelihood = -172005.417358 Norm (log-likelihood gradient vector) = 8214.299735 Norm (lambda vector) = 50.009625 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13777 13344 96.86 97.01 96.93 i-np 16033 15819 15420 97.48 96.18 96.82 o 23216 23408 22896 97.81 98.62 98.22 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.38 97.27 97.33 Avg2. 53004 53004 51660 97.46 97.46 97.46 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13777 13013 94.45 94.61 94.53 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.45 94.61 94.53 Avg2. 13755 13777 13013 94.45 94.61 94.53 Current max chunk-based F1: 94.53 (iteration 28) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 29 Log-likelihood = -164682.758882 Norm (log-likelihood gradient vector) = 14832.581876 Norm (lambda vector) = 53.708443 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13613 13264 97.44 96.43 96.93 i-np 16033 16142 15589 96.57 97.23 96.90 o 23216 23249 22820 98.15 98.29 98.22 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.39 97.32 97.35 Avg2. 53004 53004 51673 97.49 97.49 97.49 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13613 12938 95.04 94.06 94.55 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.04 94.06 94.55 Avg2. 13755 13613 12938 95.04 94.06 94.55 Current max chunk-based F1: 94.55 (iteration 29) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 30 Log-likelihood = -162274.860874 Norm (log-likelihood gradient vector) = 20808.429956 Norm (lambda vector) = 56.540080 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13644 13293 97.43 96.64 97.03 i-np 16033 16073 15568 96.86 97.10 96.98 o 23216 23287 22849 98.12 98.42 98.27 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.47 97.39 97.43 Avg2. 53004 53004 51710 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13644 12975 95.10 94.33 94.71 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.10 94.33 94.71 Avg2. 13755 13644 12975 95.10 94.33 94.71 Current max chunk-based F1: 94.71 (iteration 30) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 31 Log-likelihood = -160085.313003 Norm (log-likelihood gradient vector) = 10184.870693 Norm (lambda vector) = 55.232384 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13646 13302 97.48 96.71 97.09 i-np 16033 16038 15557 97.00 97.03 97.02 o 23216 23320 22870 98.07 98.51 98.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.52 97.42 97.47 Avg2. 53004 53004 51729 97.59 97.59 97.59 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13646 12993 95.21 94.46 94.84 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.21 94.46 94.84 Avg2. 13755 13646 12993 95.21 94.46 94.84 Current max chunk-based F1: 94.84 (iteration 31) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 32 Log-likelihood = -158686.601480 Norm (log-likelihood gradient vector) = 6846.563963 Norm (lambda vector) = 55.007989 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13650 13307 97.49 96.74 97.11 i-np 16033 16030 15555 97.04 97.02 97.03 o 23216 23324 22875 98.07 98.53 98.30 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.53 97.43 97.48 Avg2. 53004 53004 51737 97.61 97.61 97.61 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13650 12998 95.22 94.50 94.86 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.22 94.50 94.86 Avg2. 13755 13650 12998 95.22 94.50 94.86 Current max chunk-based F1: 94.86 (iteration 32) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 33 Log-likelihood = -156392.869039 Norm (log-likelihood gradient vector) = 8508.253510 Norm (lambda vector) = 55.621121 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13688 13331 97.39 96.92 97.15 i-np 16033 15958 15521 97.26 96.81 97.03 o 23216 23358 22897 98.03 98.63 98.33 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.56 97.45 97.50 Avg2. 53004 53004 51749 97.63 97.63 97.63 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13688 13019 95.11 94.65 94.88 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.11 94.65 94.88 Avg2. 13755 13688 13019 95.11 94.65 94.88 Current max chunk-based F1: 94.88 (iteration 33) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 34 Log-likelihood = -152197.839528 Norm (log-likelihood gradient vector) = 9213.020675 Norm (lambda vector) = 57.182967 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13425 13169 98.09 95.74 96.90 i-np 16033 16542 15735 95.12 98.14 96.61 o 23216 23037 22708 98.57 97.81 98.19 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.26 97.23 97.25 Avg2. 53004 53004 51612 97.37 97.37 97.37 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13425 12782 95.21 92.93 94.05 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.21 92.93 94.05 Avg2. 13755 13425 12782 95.21 92.93 94.05 Current max chunk-based F1: 94.88 (iteration 33) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 35 Log-likelihood = -158511.634721 Norm (log-likelihood gradient vector) = 48686.357776 Norm (lambda vector) = 61.776439 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13598 13290 97.73 96.62 97.17 i-np 16033 16151 15611 96.66 97.37 97.01 o 23216 23255 22848 98.25 98.41 98.33 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.55 97.47 97.51 Avg2. 53004 53004 51749 97.63 97.63 97.63 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13598 12968 95.37 94.28 94.82 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.37 94.28 94.82 Avg2. 13755 13598 12968 95.37 94.28 94.82 Current max chunk-based F1: 94.88 (iteration 33) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 36 Log-likelihood = -150467.293101 Norm (log-likelihood gradient vector) = 16451.026193 Norm (lambda vector) = 58.679083 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13674 13332 97.50 96.92 97.21 i-np 16033 16008 15553 97.16 97.01 97.08 o 23216 23322 22888 98.14 98.59 98.36 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.60 97.51 97.55 Avg2. 53004 53004 51773 97.68 97.68 97.68 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13674 13025 95.25 94.69 94.97 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.25 94.69 94.97 Avg2. 13755 13674 13025 95.25 94.69 94.97 Current max chunk-based F1: 94.97 (iteration 36) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 37 Log-likelihood = -146744.565488 Norm (log-likelihood gradient vector) = 7609.149537 Norm (lambda vector) = 60.172055 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13686 13337 97.45 96.96 97.20 i-np 16033 15983 15533 97.18 96.88 97.03 o 23216 23335 22894 98.11 98.61 98.36 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.58 97.49 97.53 Avg2. 53004 53004 51764 97.66 97.66 97.66 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13686 13020 95.13 94.66 94.89 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.13 94.66 94.89 Avg2. 13755 13686 13020 95.13 94.66 94.89 Current max chunk-based F1: 94.97 (iteration 36) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 38 Log-likelihood = -144166.531545 Norm (log-likelihood gradient vector) = 6293.245508 Norm (lambda vector) = 61.126512 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13680 13339 97.51 96.98 97.24 i-np 16033 16003 15555 97.20 97.02 97.11 o 23216 23321 22900 98.19 98.64 98.42 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.63 97.54 97.59 Avg2. 53004 53004 51794 97.72 97.72 97.72 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13680 13027 95.23 94.71 94.97 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.23 94.71 94.97 Avg2. 13755 13680 13027 95.23 94.71 94.97 Current max chunk-based F1: 94.97 (iteration 36) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 39 Log-likelihood = -140454.387039 Norm (log-likelihood gradient vector) = 6888.568436 Norm (lambda vector) = 61.995048 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13681 13342 97.52 97.00 97.26 i-np 16033 16002 15557 97.22 97.03 97.13 o 23216 23321 22902 98.20 98.65 98.42 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.65 97.56 97.60 Avg2. 53004 53004 51801 97.73 97.73 97.73 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13681 13033 95.26 94.75 95.01 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.26 94.75 95.01 Avg2. 13755 13681 13033 95.26 94.75 95.01 Current max chunk-based F1: 95.01 (iteration 39) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 40 Log-likelihood = -134433.310731 Norm (log-likelihood gradient vector) = 6082.056274 Norm (lambda vector) = 63.459951 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13595 13309 97.90 96.76 97.32 i-np 16033 16177 15660 96.80 97.67 97.24 o 23216 23232 22869 98.44 98.51 98.47 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.71 97.65 97.68 Avg2. 53004 53004 51838 97.80 97.80 97.80 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13595 13014 95.73 94.61 95.17 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.73 94.61 95.17 Avg2. 13755 13595 13014 95.73 94.61 95.17 Current max chunk-based F1: 95.17 (iteration 40) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 41 Log-likelihood = -125221.128509 Norm (log-likelihood gradient vector) = 26236.200907 Norm (lambda vector) = 65.525173 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13702 13365 97.54 97.16 97.35 i-np 16033 15953 15550 97.47 96.99 97.23 o 23216 23349 22924 98.18 98.74 98.46 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.73 97.63 97.68 Avg2. 53004 53004 51839 97.80 97.80 97.80 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13702 13071 95.39 95.03 95.21 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.39 95.03 95.21 Avg2. 13755 13702 13071 95.39 95.03 95.21 Current max chunk-based F1: 95.21 (iteration 41) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 42 Log-likelihood = -120268.268384 Norm (log-likelihood gradient vector) = 6088.371069 Norm (lambda vector) = 66.890603 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13692 13364 97.60 97.16 97.38 i-np 16033 15955 15556 97.50 97.02 97.26 o 23216 23357 22928 98.16 98.76 98.46 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.76 97.65 97.70 Avg2. 53004 53004 51848 97.82 97.82 97.82 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13692 13077 95.51 95.07 95.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.51 95.07 95.29 Avg2. 13755 13692 13077 95.51 95.07 95.29 Current max chunk-based F1: 95.29 (iteration 42) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 43 Log-likelihood = -119562.749417 Norm (log-likelihood gradient vector) = 5058.812756 Norm (lambda vector) = 66.397293 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13610 13337 97.99 96.96 97.47 i-np 16033 16122 15654 97.10 97.64 97.37 o 23216 23272 22894 98.38 98.61 98.49 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.82 97.74 97.78 Avg2. 53004 53004 51885 97.89 97.89 97.89 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13610 13065 96.00 94.98 95.49 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.00 94.98 95.49 Avg2. 13755 13610 13065 96.00 94.98 95.49 Current max chunk-based F1: 95.49 (iteration 43) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 44 Log-likelihood = -117400.135281 Norm (log-likelihood gradient vector) = 10644.352956 Norm (lambda vector) = 67.505572 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13656 13346 97.73 97.03 97.38 i-np 16033 16018 15593 97.35 97.26 97.30 o 23216 23330 22915 98.22 98.70 98.46 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.77 97.66 97.71 Avg2. 53004 53004 51854 97.83 97.83 97.83 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13656 13072 95.72 95.03 95.38 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.72 95.03 95.38 Avg2. 13755 13656 13072 95.72 95.03 95.38 Current max chunk-based F1: 95.49 (iteration 43) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 45 Log-likelihood = -114727.170982 Norm (log-likelihood gradient vector) = 8265.861553 Norm (lambda vector) = 67.939048 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13655 13357 97.82 97.11 97.46 i-np 16033 16030 15614 97.40 97.39 97.40 o 23216 23319 22919 98.28 98.72 98.50 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.84 97.74 97.79 Avg2. 53004 53004 51890 97.90 97.90 97.90 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13655 13092 95.88 95.18 95.53 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.88 95.18 95.53 Avg2. 13755 13655 13092 95.88 95.18 95.53 Current max chunk-based F1: 95.53 (iteration 45) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 46 Log-likelihood = -114294.006078 Norm (log-likelihood gradient vector) = 5400.263371 Norm (lambda vector) = 67.706809 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13656 13364 97.86 97.16 97.51 i-np 16033 16036 15626 97.44 97.46 97.45 o 23216 23312 22924 98.34 98.74 98.54 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.88 97.79 97.83 Avg2. 53004 53004 51914 97.94 97.94 97.94 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13656 13104 95.96 95.27 95.61 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.96 95.27 95.61 Avg2. 13755 13656 13104 95.96 95.27 95.61 Current max chunk-based F1: 95.61 (iteration 46) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 47 Log-likelihood = -112615.573415 Norm (log-likelihood gradient vector) = 4150.932907 Norm (lambda vector) = 67.777903 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13654 13358 97.83 97.11 97.47 i-np 16033 16053 15627 97.35 97.47 97.41 o 23216 23297 22915 98.36 98.70 98.53 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.85 97.76 97.80 Avg2. 53004 53004 51900 97.92 97.92 97.92 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13654 13089 95.86 95.16 95.51 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.86 95.16 95.51 Avg2. 13755 13654 13089 95.86 95.16 95.51 Current max chunk-based F1: 95.61 (iteration 46) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 48 Log-likelihood = -111374.688499 Norm (log-likelihood gradient vector) = 4548.199971 Norm (lambda vector) = 67.717742 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13632 13349 97.92 97.05 97.48 i-np 16033 16129 15668 97.14 97.72 97.43 o 23216 23243 22895 98.50 98.62 98.56 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.80 97.83 Avg2. 53004 53004 51912 97.94 97.94 97.94 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13632 13078 95.94 95.08 95.51 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.94 95.08 95.51 Avg2. 13755 13632 13078 95.94 95.08 95.51 Current max chunk-based F1: 95.61 (iteration 46) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 49 Log-likelihood = -106878.516859 Norm (log-likelihood gradient vector) = 12119.841067 Norm (lambda vector) = 68.340256 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13653 13361 97.86 97.14 97.50 i-np 16033 16084 15641 97.25 97.56 97.40 o 23216 23267 22905 98.44 98.66 98.55 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.85 97.78 97.82 Avg2. 53004 53004 51907 97.93 97.93 97.93 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13653 13087 95.85 95.14 95.50 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.85 95.14 95.50 Avg2. 13755 13653 13087 95.85 95.14 95.50 Current max chunk-based F1: 95.61 (iteration 46) Training iteration elapsed (including evaluation time): 33 seconds Iteration: 50 Log-likelihood = -105143.705069 Norm (log-likelihood gradient vector) = 5240.782564 Norm (lambda vector) = 69.047328 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13633 13351 97.93 97.06 97.50 i-np 16033 16112 15655 97.16 97.64 97.40 o 23216 23259 22898 98.45 98.63 98.54 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.85 97.78 97.81 Avg2. 53004 53004 51904 97.92 97.92 97.92 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13633 13079 95.94 95.09 95.51 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.94 95.09 95.51 Avg2. 13755 13633 13079 95.94 95.09 95.51 Current max chunk-based F1: 95.61 (iteration 46) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 51 Log-likelihood = -103416.862843 Norm (log-likelihood gradient vector) = 6217.754788 Norm (lambda vector) = 69.596906 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13650 13360 97.88 97.13 97.50 i-np 16033 16068 15632 97.29 97.50 97.39 o 23216 23286 22908 98.38 98.67 98.52 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.85 97.77 97.81 Avg2. 53004 53004 51900 97.92 97.92 97.92 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13650 13089 95.89 95.16 95.52 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.89 95.16 95.52 Avg2. 13755 13650 13089 95.89 95.16 95.52 Current max chunk-based F1: 95.61 (iteration 46) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 52 Log-likelihood = -101847.562743 Norm (log-likelihood gradient vector) = 5025.418588 Norm (lambda vector) = 70.059060 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13650 13363 97.90 97.15 97.52 i-np 16033 16074 15635 97.27 97.52 97.39 o 23216 23280 22907 98.40 98.67 98.53 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.85 97.78 97.82 Avg2. 53004 53004 51905 97.93 97.93 97.93 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13650 13088 95.88 95.15 95.52 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.88 95.15 95.52 Avg2. 13755 13650 13088 95.88 95.15 95.52 Current max chunk-based F1: 95.61 (iteration 46) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 53 Log-likelihood = -100061.119550 Norm (log-likelihood gradient vector) = 4268.944354 Norm (lambda vector) = 70.497548 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13706 13382 97.64 97.29 97.46 i-np 16033 15975 15580 97.53 97.17 97.35 o 23216 23323 22926 98.30 98.75 98.52 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.82 97.74 97.78 Avg2. 53004 53004 51888 97.89 97.89 97.89 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13706 13099 95.57 95.23 95.40 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.57 95.23 95.40 Avg2. 13755 13706 13099 95.57 95.23 95.40 Current max chunk-based F1: 95.61 (iteration 46) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 54 Log-likelihood = -95634.318043 Norm (log-likelihood gradient vector) = 8742.475335 Norm (lambda vector) = 71.834510 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13623 13346 97.97 97.03 97.49 i-np 16033 16156 15676 97.03 97.77 97.40 o 23216 23225 22880 98.51 98.55 98.53 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.84 97.78 97.81 Avg2. 53004 53004 51902 97.92 97.92 97.92 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13623 13071 95.95 95.03 95.49 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.95 95.03 95.49 Avg2. 13755 13623 13071 95.95 95.03 95.49 Current max chunk-based F1: 95.61 (iteration 46) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 55 Log-likelihood = -93164.745984 Norm (log-likelihood gradient vector) = 11264.928239 Norm (lambda vector) = 73.269595 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13658 13367 97.87 97.18 97.52 i-np 16033 16081 15644 97.28 97.57 97.43 o 23216 23265 22903 98.44 98.65 98.55 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.87 97.80 97.83 Avg2. 53004 53004 51914 97.94 97.94 97.94 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13658 13094 95.87 95.19 95.53 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.87 95.19 95.53 Avg2. 13755 13658 13094 95.87 95.19 95.53 Current max chunk-based F1: 95.61 (iteration 46) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 56 Log-likelihood = -92649.082197 Norm (log-likelihood gradient vector) = 4850.862910 Norm (lambda vector) = 72.762706 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13674 13375 97.81 97.24 97.52 i-np 16033 16047 15625 97.37 97.46 97.41 o 23216 23283 22912 98.41 98.69 98.55 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.79 97.83 Avg2. 53004 53004 51912 97.94 97.94 97.94 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13674 13098 95.79 95.22 95.50 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.79 95.22 95.50 Avg2. 13755 13674 13098 95.79 95.22 95.50 Current max chunk-based F1: 95.61 (iteration 46) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 57 Log-likelihood = -92235.465944 Norm (log-likelihood gradient vector) = 3276.485314 Norm (lambda vector) = 72.647926 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13672 13371 97.80 97.21 97.50 i-np 16033 16057 15630 97.34 97.49 97.41 o 23216 23275 22909 98.43 98.68 98.55 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.79 97.82 Avg2. 53004 53004 51910 97.94 97.94 97.94 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13672 13091 95.75 95.17 95.46 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.75 95.17 95.46 Avg2. 13755 13672 13091 95.75 95.17 95.46 Current max chunk-based F1: 95.61 (iteration 46) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 58 Log-likelihood = -91506.877568 Norm (log-likelihood gradient vector) = 3862.246541 Norm (lambda vector) = 72.848950 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13660 13371 97.88 97.21 97.55 i-np 16033 16069 15640 97.33 97.55 97.44 o 23216 23275 22914 98.45 98.70 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.89 97.82 97.85 Avg2. 53004 53004 51925 97.96 97.96 97.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13660 13093 95.85 95.19 95.52 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.85 95.19 95.52 Avg2. 13755 13660 13093 95.85 95.19 95.52 Current max chunk-based F1: 95.61 (iteration 46) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 59 Log-likelihood = -90264.243318 Norm (log-likelihood gradient vector) = 4024.796931 Norm (lambda vector) = 73.375170 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13600 13332 98.03 96.92 97.47 i-np 16033 16211 15687 96.77 97.84 97.30 o 23216 23193 22857 98.55 98.45 98.50 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.78 97.74 97.76 Avg2. 53004 53004 51876 97.87 97.87 97.87 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13600 13028 95.79 94.71 95.25 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.79 94.71 95.25 Avg2. 13755 13600 13028 95.79 94.71 95.25 Current max chunk-based F1: 95.61 (iteration 46) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 60 Log-likelihood = -89690.554501 Norm (log-likelihood gradient vector) = 18405.699093 Norm (lambda vector) = 76.457462 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13635 13362 98.00 97.14 97.57 i-np 16033 16128 15674 97.19 97.76 97.47 o 23216 23241 22897 98.52 98.63 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.90 97.84 97.87 Avg2. 53004 53004 51933 97.98 97.98 97.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13635 13086 95.97 95.14 95.55 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.97 95.14 95.55 Avg2. 13755 13635 13086 95.97 95.14 95.55 Current max chunk-based F1: 95.61 (iteration 46) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 61 Log-likelihood = -88864.035996 Norm (log-likelihood gradient vector) = 9650.281162 Norm (lambda vector) = 74.660877 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13659 13377 97.94 97.25 97.59 i-np 16033 16068 15648 97.39 97.60 97.49 o 23216 23277 22914 98.44 98.70 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.92 97.85 97.89 Avg2. 53004 53004 51939 97.99 97.99 97.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13659 13107 95.96 95.29 95.62 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.96 95.29 95.62 Avg2. 13755 13659 13107 95.96 95.29 95.62 Current max chunk-based F1: 95.62 (iteration 61) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 62 Log-likelihood = -87491.504697 Norm (log-likelihood gradient vector) = 7059.758462 Norm (lambda vector) = 75.779017 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13623 13360 98.07 97.13 97.60 i-np 16033 16136 15686 97.21 97.84 97.52 o 23216 23245 22900 98.52 98.64 98.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.93 97.87 97.90 Avg2. 53004 53004 51946 98.00 98.00 98.00 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13623 13096 96.13 95.21 95.67 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.13 95.21 95.67 Avg2. 13755 13623 13096 96.13 95.21 95.67 Current max chunk-based F1: 95.67 (iteration 62) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 63 Log-likelihood = -86539.624298 Norm (log-likelihood gradient vector) = 4071.069361 Norm (lambda vector) = 76.481683 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13644 13372 98.01 97.22 97.61 i-np 16033 16105 15673 97.32 97.75 97.54 o 23216 23255 22906 98.50 98.66 98.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.94 97.88 97.91 Avg2. 53004 53004 51951 98.01 98.01 98.01 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13644 13115 96.12 95.35 95.73 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.12 95.35 95.73 Avg2. 13755 13644 13115 96.12 95.35 95.73 Current max chunk-based F1: 95.73 (iteration 63) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 64 Log-likelihood = -85731.366711 Norm (log-likelihood gradient vector) = 3815.611238 Norm (lambda vector) = 77.246702 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13624 13364 98.09 97.16 97.62 i-np 16033 16149 15700 97.22 97.92 97.57 o 23216 23231 22894 98.55 98.61 98.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.95 97.90 97.93 Avg2. 53004 53004 51958 98.03 98.03 98.03 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13624 13115 96.26 95.35 95.80 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.26 95.35 95.80 Avg2. 13755 13624 13115 96.26 95.35 95.80 Current max chunk-based F1: 95.80 (iteration 64) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 65 Log-likelihood = -84703.302892 Norm (log-likelihood gradient vector) = 5407.832728 Norm (lambda vector) = 78.209828 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13666 13379 97.90 97.27 97.58 i-np 16033 16058 15651 97.47 97.62 97.54 o 23216 23280 22918 98.45 98.72 98.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.94 97.87 97.90 Avg2. 53004 53004 51948 98.01 98.01 98.01 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13666 13116 95.98 95.35 95.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.98 95.35 95.66 Avg2. 13755 13666 13116 95.98 95.35 95.66 Current max chunk-based F1: 95.80 (iteration 64) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 66 Log-likelihood = -83332.811301 Norm (log-likelihood gradient vector) = 4162.966127 Norm (lambda vector) = 79.635043 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13667 13377 97.88 97.25 97.56 i-np 16033 16063 15649 97.42 97.60 97.51 o 23216 23274 22915 98.46 98.70 98.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.92 97.85 97.89 Avg2. 53004 53004 51941 97.99 97.99 97.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13667 13113 95.95 95.33 95.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.95 95.33 95.64 Avg2. 13755 13667 13113 95.95 95.33 95.64 Current max chunk-based F1: 95.80 (iteration 64) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 67 Log-likelihood = -82035.984176 Norm (log-likelihood gradient vector) = 3490.259761 Norm (lambda vector) = 80.476283 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13667 13380 97.90 97.27 97.59 i-np 16033 16067 15652 97.42 97.62 97.52 o 23216 23270 22914 98.47 98.70 98.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.93 97.87 97.90 Avg2. 53004 53004 51946 98.00 98.00 98.00 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13667 13119 95.99 95.38 95.68 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.99 95.38 95.68 Avg2. 13755 13667 13119 95.99 95.38 95.68 Current max chunk-based F1: 95.80 (iteration 64) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 68 Log-likelihood = -79323.813738 Norm (log-likelihood gradient vector) = 2955.717849 Norm (lambda vector) = 81.624840 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13705 13398 97.76 97.40 97.58 i-np 16033 15978 15593 97.59 97.26 97.42 o 23216 23321 22929 98.32 98.76 98.54 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.89 97.81 97.85 Avg2. 53004 53004 51920 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13705 13135 95.84 95.49 95.67 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.84 95.49 95.67 Avg2. 13755 13705 13135 95.84 95.49 95.67 Current max chunk-based F1: 95.80 (iteration 64) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 69 Log-likelihood = -76521.693041 Norm (log-likelihood gradient vector) = 6417.419806 Norm (lambda vector) = 82.979782 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13651 13380 98.01 97.27 97.64 i-np 16033 16122 15684 97.28 97.82 97.55 o 23216 23231 22895 98.55 98.62 98.59 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.95 97.90 97.93 Avg2. 53004 53004 51959 98.03 98.03 98.03 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13651 13130 96.18 95.46 95.82 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.18 95.46 95.82 Avg2. 13755 13651 13130 96.18 95.46 95.82 Current max chunk-based F1: 95.82 (iteration 69) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 70 Log-likelihood = -73899.126059 Norm (log-likelihood gradient vector) = 5265.066349 Norm (lambda vector) = 83.948687 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13675 13393 97.94 97.37 97.65 i-np 16033 16071 15663 97.46 97.69 97.58 o 23216 23258 22909 98.50 98.68 98.59 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.97 97.91 97.94 Avg2. 53004 53004 51965 98.04 98.04 98.04 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13675 13151 96.17 95.61 95.89 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.17 95.61 95.89 Avg2. 13755 13675 13151 96.17 95.61 95.89 Current max chunk-based F1: 95.89 (iteration 70) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 71 Log-likelihood = -73531.270057 Norm (log-likelihood gradient vector) = 2957.434715 Norm (lambda vector) = 83.619716 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13678 13398 97.95 97.40 97.68 i-np 16033 16059 15661 97.52 97.68 97.60 o 23216 23267 22917 98.50 98.71 98.60 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.99 97.93 97.96 Avg2. 53004 53004 51976 98.06 98.06 98.06 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13678 13157 96.19 95.65 95.92 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.19 95.65 95.92 Avg2. 13755 13678 13157 96.19 95.65 95.92 Current max chunk-based F1: 95.92 (iteration 71) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 72 Log-likelihood = -73217.541237 Norm (log-likelihood gradient vector) = 3158.262212 Norm (lambda vector) = 83.479767 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13684 13395 97.89 97.38 97.63 i-np 16033 16042 15644 97.52 97.57 97.55 o 23216 23278 22918 98.45 98.72 98.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.95 97.89 97.92 Avg2. 53004 53004 51957 98.02 98.02 98.02 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13684 13145 96.06 95.57 95.81 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.06 95.57 95.81 Avg2. 13755 13684 13145 96.06 95.57 95.81 Current max chunk-based F1: 95.92 (iteration 71) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 73 Log-likelihood = -71924.676170 Norm (log-likelihood gradient vector) = 2672.210841 Norm (lambda vector) = 83.753349 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13622 13377 98.20 97.25 97.72 i-np 16033 16159 15714 97.25 98.01 97.63 o 23216 23223 22898 98.60 98.63 98.62 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.02 97.96 97.99 Avg2. 53004 53004 51989 98.09 98.09 98.09 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13622 13141 96.47 95.54 96.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.47 95.54 96.00 Avg2. 13755 13622 13141 96.47 95.54 96.00 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 74 Log-likelihood = -70074.951132 Norm (log-likelihood gradient vector) = 10927.614873 Norm (lambda vector) = 85.599971 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13670 13396 98.00 97.39 97.69 i-np 16033 16050 15651 97.51 97.62 97.57 o 23216 23284 22927 98.47 98.76 98.61 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.99 97.92 97.96 Avg2. 53004 53004 51974 98.06 98.06 98.06 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13670 13146 96.17 95.57 95.87 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.17 95.57 95.87 Avg2. 13755 13670 13146 96.17 95.57 95.87 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 75 Log-likelihood = -68411.641640 Norm (log-likelihood gradient vector) = 2636.461118 Norm (lambda vector) = 86.309046 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13669 13391 97.97 97.35 97.66 i-np 16033 16052 15645 97.46 97.58 97.52 o 23216 23283 22922 98.45 98.73 98.59 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.96 97.89 97.92 Avg2. 53004 53004 51958 98.03 98.03 98.03 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13669 13135 96.09 95.49 95.79 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.09 95.49 95.79 Avg2. 13755 13669 13135 96.09 95.49 95.79 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 76 Log-likelihood = -67874.643533 Norm (log-likelihood gradient vector) = 2513.557746 Norm (lambda vector) = 86.749943 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13678 13389 97.89 97.34 97.61 i-np 16033 16050 15635 97.41 97.52 97.47 o 23216 23276 22913 98.44 98.69 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.91 97.85 97.88 Avg2. 53004 53004 51937 97.99 97.99 97.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13678 13128 95.98 95.44 95.71 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.98 95.44 95.71 Avg2. 13755 13678 13128 95.98 95.44 95.71 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 77 Log-likelihood = -66082.905367 Norm (log-likelihood gradient vector) = 2707.432703 Norm (lambda vector) = 88.415035 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13655 13378 97.97 97.26 97.61 i-np 16033 16080 15647 97.31 97.59 97.45 o 23216 23269 22910 98.46 98.68 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.91 97.84 97.88 Avg2. 53004 53004 51935 97.98 97.98 97.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13655 13110 96.01 95.31 95.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.01 95.31 95.66 Avg2. 13755 13655 13110 96.01 95.31 95.66 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 33 seconds Iteration: 78 Log-likelihood = -63233.898588 Norm (log-likelihood gradient vector) = 5648.764783 Norm (lambda vector) = 91.041623 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13696 13396 97.81 97.39 97.60 i-np 16033 15993 15604 97.57 97.32 97.45 o 23216 23315 22928 98.34 98.76 98.55 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.91 97.82 97.87 Avg2. 53004 53004 51928 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13696 13133 95.89 95.48 95.68 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.89 95.48 95.68 Avg2. 13755 13696 13133 95.89 95.48 95.68 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 79 Log-likelihood = -61189.890174 Norm (log-likelihood gradient vector) = 5223.598402 Norm (lambda vector) = 93.078551 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13658 13385 98.00 97.31 97.65 i-np 16033 16083 15669 97.43 97.73 97.58 o 23216 23263 22913 98.50 98.69 98.60 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.97 97.91 97.94 Avg2. 53004 53004 51967 98.04 98.04 98.04 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13658 13142 96.22 95.54 95.88 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.22 95.54 95.88 Avg2. 13755 13658 13142 96.22 95.54 95.88 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 80 Log-likelihood = -60670.454979 Norm (log-likelihood gradient vector) = 2426.914704 Norm (lambda vector) = 92.935101 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13653 13385 98.04 97.31 97.67 i-np 16033 16098 15678 97.39 97.79 97.59 o 23216 23253 22908 98.52 98.67 98.59 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.98 97.92 97.95 Avg2. 53004 53004 51971 98.05 98.05 98.05 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13653 13142 96.26 95.54 95.90 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.26 95.54 95.90 Avg2. 13755 13653 13142 96.26 95.54 95.90 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 81 Log-likelihood = -60255.001263 Norm (log-likelihood gradient vector) = 2116.387145 Norm (lambda vector) = 93.035763 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13648 13381 98.04 97.28 97.66 i-np 16033 16106 15681 97.36 97.80 97.58 o 23216 23250 22904 98.51 98.66 98.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.97 97.91 97.94 Avg2. 53004 53004 51966 98.04 98.04 98.04 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13648 13141 96.29 95.54 95.91 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.29 95.54 95.91 Avg2. 13755 13648 13141 96.29 95.54 95.91 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 82 Log-likelihood = -59356.223138 Norm (log-likelihood gradient vector) = 2547.735040 Norm (lambda vector) = 94.247557 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13655 13384 98.02 97.30 97.66 i-np 16033 16067 15655 97.44 97.64 97.54 o 23216 23282 22917 98.43 98.71 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.96 97.89 97.92 Avg2. 53004 53004 51956 98.02 98.02 98.02 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13655 13138 96.21 95.51 95.86 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.21 95.51 95.86 Avg2. 13755 13655 13138 96.21 95.51 95.86 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 83 Log-likelihood = -57404.327216 Norm (log-likelihood gradient vector) = 2888.211360 Norm (lambda vector) = 96.541703 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13715 13374 97.51 97.23 97.37 i-np 16033 16007 15580 97.33 97.17 97.25 o 23216 23282 22891 98.32 98.60 98.46 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.72 97.67 97.70 Avg2. 53004 53004 51845 97.81 97.81 97.81 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13715 13095 95.48 95.20 95.34 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.48 95.20 95.34 Avg2. 13755 13715 13095 95.48 95.20 95.34 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 84 Log-likelihood = -57143.386049 Norm (log-likelihood gradient vector) = 17529.689831 Norm (lambda vector) = 100.581173 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13666 13386 97.95 97.32 97.63 i-np 16033 16053 15643 97.45 97.57 97.51 o 23216 23285 22917 98.42 98.71 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.94 97.87 97.90 Avg2. 53004 53004 51946 98.00 98.00 98.00 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13666 13137 96.13 95.51 95.82 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.13 95.51 95.82 Avg2. 13755 13666 13137 96.13 95.51 95.82 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 85 Log-likelihood = -56556.471700 Norm (log-likelihood gradient vector) = 4307.155469 Norm (lambda vector) = 97.526115 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13667 13383 97.92 97.30 97.61 i-np 16033 16059 15644 97.42 97.57 97.49 o 23216 23278 22912 98.43 98.69 98.56 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.92 97.85 97.89 Avg2. 53004 53004 51939 97.99 97.99 97.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13667 13130 96.07 95.46 95.76 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.07 95.46 95.76 Avg2. 13755 13667 13130 96.07 95.46 95.76 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 33 seconds Iteration: 86 Log-likelihood = -55290.767955 Norm (log-likelihood gradient vector) = 2590.360855 Norm (lambda vector) = 98.685777 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13682 13394 97.90 97.38 97.63 i-np 16033 16031 15633 97.52 97.51 97.51 o 23216 23291 22921 98.41 98.73 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.94 97.87 97.91 Avg2. 53004 53004 51948 98.01 98.01 98.01 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13682 13143 96.06 95.55 95.80 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.06 95.55 95.80 Avg2. 13755 13682 13143 96.06 95.55 95.80 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 87 Log-likelihood = -54287.491957 Norm (log-likelihood gradient vector) = 2087.367349 Norm (lambda vector) = 99.034196 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13693 13399 97.85 97.41 97.63 i-np 16033 16038 15638 97.51 97.54 97.52 o 23216 23273 22913 98.45 98.69 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.94 97.88 97.91 Avg2. 53004 53004 51950 98.01 98.01 98.01 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13693 13153 96.06 95.62 95.84 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.06 95.62 95.84 Avg2. 13755 13693 13153 96.06 95.62 95.84 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 88 Log-likelihood = -53024.906067 Norm (log-likelihood gradient vector) = 3545.708855 Norm (lambda vector) = 99.488414 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13680 13402 97.97 97.43 97.70 i-np 16033 16054 15652 97.50 97.62 97.56 o 23216 23270 22918 98.49 98.72 98.60 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.98 97.92 97.95 Avg2. 53004 53004 51972 98.05 98.05 98.05 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13680 13155 96.16 95.64 95.90 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.16 95.64 95.90 Avg2. 13755 13680 13155 96.16 95.64 95.90 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 89 Log-likelihood = -52153.332537 Norm (log-likelihood gradient vector) = 2814.215126 Norm (lambda vector) = 99.811249 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13750 13427 97.65 97.62 97.63 i-np 16033 15914 15567 97.82 97.09 97.46 o 23216 23340 22944 98.30 98.83 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.92 97.85 97.89 Avg2. 53004 53004 51938 97.99 97.99 97.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13750 13169 95.77 95.74 95.76 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.77 95.74 95.76 Avg2. 13755 13750 13169 95.77 95.74 95.76 Current max chunk-based F1: 96.00 (iteration 73) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 90 Log-likelihood = -51674.271807 Norm (log-likelihood gradient vector) = 8108.342213 Norm (lambda vector) = 100.607568 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13683 13409 98.00 97.48 97.74 i-np 16033 16066 15678 97.58 97.79 97.69 o 23216 23255 22926 98.59 98.75 98.67 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.06 98.01 98.03 Avg2. 53004 53004 52013 98.13 98.13 98.13 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13683 13171 96.26 95.75 96.01 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.26 95.75 96.01 Avg2. 13755 13683 13171 96.26 95.75 96.01 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 91 Log-likelihood = -50795.614210 Norm (log-likelihood gradient vector) = 2333.248805 Norm (lambda vector) = 101.104157 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13674 13405 98.03 97.46 97.74 i-np 16033 16065 15671 97.55 97.74 97.64 o 23216 23265 22926 98.54 98.75 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.04 97.98 98.01 Avg2. 53004 53004 52002 98.11 98.11 98.11 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13674 13163 96.26 95.70 95.98 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.26 95.70 95.98 Avg2. 13755 13674 13163 96.26 95.70 95.98 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 92 Log-likelihood = -50539.378217 Norm (log-likelihood gradient vector) = 2038.739548 Norm (lambda vector) = 101.434430 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13668 13399 98.03 97.41 97.72 i-np 16033 16064 15668 97.53 97.72 97.63 o 23216 23272 22926 98.51 98.75 98.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.03 97.96 97.99 Avg2. 53004 53004 51993 98.09 98.09 98.09 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13668 13156 96.25 95.65 95.95 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.25 95.65 95.95 Avg2. 13755 13668 13156 96.25 95.65 95.95 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 93 Log-likelihood = -49898.480624 Norm (log-likelihood gradient vector) = 1636.371328 Norm (lambda vector) = 102.371643 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13681 13395 97.91 97.38 97.65 i-np 16033 16046 15650 97.53 97.61 97.57 o 23216 23277 22921 98.47 98.73 98.60 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.97 97.91 97.94 Avg2. 53004 53004 51966 98.04 98.04 98.04 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13681 13150 96.12 95.60 95.86 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.12 95.60 95.86 Avg2. 13755 13681 13150 96.12 95.60 95.86 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 94 Log-likelihood = -48500.974546 Norm (log-likelihood gradient vector) = 2733.061442 Norm (lambda vector) = 104.575753 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13660 13388 98.01 97.33 97.67 i-np 16033 16057 15653 97.48 97.63 97.56 o 23216 23287 22922 98.43 98.73 98.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.98 97.90 97.94 Avg2. 53004 53004 51963 98.04 98.04 98.04 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13660 13141 96.20 95.54 95.87 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.20 95.54 95.87 Avg2. 13755 13660 13141 96.20 95.54 95.87 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 95 Log-likelihood = -47528.617392 Norm (log-likelihood gradient vector) = 3698.084807 Norm (lambda vector) = 106.478010 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13674 13389 97.92 97.34 97.63 i-np 16033 16040 15642 97.52 97.56 97.54 o 23216 23290 22922 98.42 98.73 98.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.95 97.88 97.91 Avg2. 53004 53004 51953 98.02 98.02 98.02 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13674 13144 96.12 95.56 95.84 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.12 95.56 95.84 Avg2. 13755 13674 13144 96.12 95.56 95.84 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 96 Log-likelihood = -46524.111263 Norm (log-likelihood gradient vector) = 2410.768021 Norm (lambda vector) = 107.653097 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13685 13388 97.83 97.33 97.58 i-np 16033 16063 15637 97.35 97.53 97.44 o 23216 23256 22899 98.46 98.63 98.55 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.88 97.83 97.86 Avg2. 53004 53004 51924 97.96 97.96 97.96 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13685 13125 95.91 95.42 95.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.91 95.42 95.66 Avg2. 13755 13685 13125 95.91 95.42 95.66 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 97 Log-likelihood = -45626.778056 Norm (log-likelihood gradient vector) = 1807.915825 Norm (lambda vector) = 108.823262 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13690 13388 97.79 97.33 97.56 i-np 16033 16054 15627 97.34 97.47 97.40 o 23216 23260 22898 98.44 98.63 98.54 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.81 97.83 Avg2. 53004 53004 51913 97.94 97.94 97.94 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13690 13122 95.85 95.40 95.62 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.85 95.40 95.62 Avg2. 13755 13690 13122 95.85 95.40 95.62 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 98 Log-likelihood = -44930.062340 Norm (log-likelihood gradient vector) = 1885.256621 Norm (lambda vector) = 110.217441 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13693 13392 97.80 97.36 97.58 i-np 16033 16099 15664 97.30 97.70 97.50 o 23216 23212 22882 98.58 98.56 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.89 97.87 97.88 Avg2. 53004 53004 51938 97.99 97.99 97.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13693 13140 95.96 95.53 95.74 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.96 95.53 95.74 Avg2. 13755 13693 13140 95.96 95.53 95.74 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 99 Log-likelihood = -43717.953735 Norm (log-likelihood gradient vector) = 3017.603583 Norm (lambda vector) = 113.538560 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13696 13386 97.74 97.32 97.53 i-np 16033 16036 15622 97.42 97.44 97.43 o 23216 23272 22905 98.42 98.66 98.54 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.80 97.83 Avg2. 53004 53004 51913 97.94 97.94 97.94 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13696 13128 95.85 95.44 95.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.85 95.44 95.65 Avg2. 13755 13696 13128 95.85 95.44 95.65 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 100 Log-likelihood = -43031.357659 Norm (log-likelihood gradient vector) = 2135.981144 Norm (lambda vector) = 115.631463 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13656 13377 97.96 97.25 97.60 i-np 16033 16105 15673 97.32 97.75 97.54 o 23216 23243 22897 98.51 98.63 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.93 97.88 97.90 Avg2. 53004 53004 51947 98.01 98.01 98.01 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13656 13131 96.16 95.46 95.81 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.16 95.46 95.81 Avg2. 13755 13656 13131 96.16 95.46 95.81 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 101 Log-likelihood = -42954.913300 Norm (log-likelihood gradient vector) = 3587.201643 Norm (lambda vector) = 116.368842 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13688 13391 97.83 97.35 97.59 i-np 16033 16028 15623 97.47 97.44 97.46 o 23216 23288 22912 98.39 98.69 98.54 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.90 97.83 97.86 Avg2. 53004 53004 51926 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13688 13136 95.97 95.50 95.73 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.97 95.50 95.73 Avg2. 13755 13688 13136 95.97 95.50 95.73 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 33 seconds Iteration: 102 Log-likelihood = -42872.359059 Norm (log-likelihood gradient vector) = 1975.260123 Norm (lambda vector) = 116.744408 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13689 13389 97.81 97.34 97.57 i-np 16033 16044 15631 97.43 97.49 97.46 o 23216 23271 22908 98.44 98.67 98.56 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.89 97.84 97.86 Avg2. 53004 53004 51928 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13689 13130 95.92 95.46 95.69 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.92 95.46 95.69 Avg2. 13755 13689 13130 95.92 95.46 95.69 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 103 Log-likelihood = -42696.051569 Norm (log-likelihood gradient vector) = 1809.977577 Norm (lambda vector) = 117.246385 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13691 13392 97.82 97.36 97.59 i-np 16033 16059 15643 97.41 97.57 97.49 o 23216 23254 22904 98.49 98.66 98.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.91 97.86 97.88 Avg2. 53004 53004 51939 97.99 97.99 97.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13691 13137 95.95 95.51 95.73 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.95 95.51 95.73 Avg2. 13755 13691 13137 95.95 95.51 95.73 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 104 Log-likelihood = -41926.947474 Norm (log-likelihood gradient vector) = 1788.609783 Norm (lambda vector) = 119.284429 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13679 13383 97.84 97.30 97.57 i-np 16033 16072 15650 97.37 97.61 97.49 o 23216 23253 22901 98.49 98.64 98.56 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.90 97.85 97.87 Avg2. 53004 53004 51934 97.98 97.98 97.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13679 13134 96.02 95.49 95.75 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.02 95.49 95.75 Avg2. 13755 13679 13134 96.02 95.49 95.75 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 105 Log-likelihood = -41301.713153 Norm (log-likelihood gradient vector) = 7516.673241 Norm (lambda vector) = 122.488413 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13683 13388 97.84 97.33 97.59 i-np 16033 16070 15650 97.39 97.61 97.50 o 23216 23251 22902 98.50 98.65 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.91 97.86 97.89 Avg2. 53004 53004 51940 97.99 97.99 97.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13683 13138 96.02 95.51 95.76 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.02 95.51 95.76 Avg2. 13755 13683 13138 96.02 95.51 95.76 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 106 Log-likelihood = -41337.568168 Norm (log-likelihood gradient vector) = 4053.502563 Norm (lambda vector) = 120.953307 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13688 13382 97.76 97.29 97.53 i-np 16033 16056 15633 97.37 97.51 97.44 o 23216 23260 22901 98.46 98.64 98.55 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.86 97.81 97.84 Avg2. 53004 53004 51916 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13688 13124 95.88 95.41 95.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.88 95.41 95.65 Avg2. 13755 13688 13124 95.88 95.41 95.65 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 107 Log-likelihood = -40442.985778 Norm (log-likelihood gradient vector) = 2453.537342 Norm (lambda vector) = 122.953974 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13688 13383 97.77 97.30 97.53 i-np 16033 16055 15633 97.37 97.51 97.44 o 23216 23261 22902 98.46 98.65 98.55 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.87 97.82 97.84 Avg2. 53004 53004 51918 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13688 13124 95.88 95.41 95.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.88 95.41 95.65 Avg2. 13755 13688 13124 95.88 95.41 95.65 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 108 Log-likelihood = -39350.681973 Norm (log-likelihood gradient vector) = 1904.847246 Norm (lambda vector) = 124.939033 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13685 13393 97.87 97.37 97.62 i-np 16033 16053 15641 97.43 97.56 97.49 o 23216 23266 22910 98.47 98.68 98.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.92 97.87 97.90 Avg2. 53004 53004 51944 98.00 98.00 98.00 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13685 13141 96.02 95.54 95.78 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.02 95.54 95.78 Avg2. 13755 13685 13141 96.02 95.54 95.78 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 33 seconds Iteration: 109 Log-likelihood = -37772.353878 Norm (log-likelihood gradient vector) = 1787.614367 Norm (lambda vector) = 127.545258 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13679 13389 97.88 97.34 97.61 i-np 16033 16052 15644 97.46 97.57 97.52 o 23216 23273 22913 98.45 98.69 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.93 97.87 97.90 Avg2. 53004 53004 51946 98.00 98.00 98.00 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13679 13144 96.09 95.56 95.82 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.09 95.56 95.82 Avg2. 13755 13679 13144 96.09 95.56 95.82 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 110 Log-likelihood = -36394.283453 Norm (log-likelihood gradient vector) = 3236.752414 Norm (lambda vector) = 130.654515 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13733 13410 97.65 97.49 97.57 i-np 16033 15955 15589 97.71 97.23 97.47 o 23216 23316 22931 98.35 98.77 98.56 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.90 97.83 97.87 Avg2. 53004 53004 51930 97.97 97.97 97.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13733 13151 95.76 95.61 95.69 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.76 95.61 95.69 Avg2. 13755 13733 13151 95.76 95.61 95.69 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 111 Log-likelihood = -35639.151443 Norm (log-likelihood gradient vector) = 5563.157757 Norm (lambda vector) = 132.357106 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13704 13399 97.77 97.41 97.59 i-np 16033 16022 15626 97.53 97.46 97.49 o 23216 23278 22914 98.44 98.70 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.91 97.86 97.89 Avg2. 53004 53004 51939 97.99 97.99 97.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13704 13146 95.93 95.57 95.75 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.93 95.57 95.75 Avg2. 13755 13704 13146 95.93 95.57 95.75 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 112 Log-likelihood = -35504.198308 Norm (log-likelihood gradient vector) = 2114.499863 Norm (lambda vector) = 131.484479 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13690 13394 97.84 97.38 97.61 i-np 16033 16059 15650 97.45 97.61 97.53 o 23216 23255 22907 98.50 98.67 98.59 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.93 97.89 97.91 Avg2. 53004 53004 51951 98.01 98.01 98.01 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13690 13146 96.03 95.57 95.80 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.03 95.57 95.80 Avg2. 13755 13690 13146 96.03 95.57 95.80 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 113 Log-likelihood = -35302.280364 Norm (log-likelihood gradient vector) = 1293.133734 Norm (lambda vector) = 131.562761 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13686 13390 97.84 97.35 97.59 i-np 16033 16078 15660 97.40 97.67 97.54 o 23216 23240 22901 98.54 98.64 98.59 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.93 97.89 97.91 Avg2. 53004 53004 51951 98.01 98.01 98.01 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13686 13143 96.03 95.55 95.79 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.03 95.55 95.79 Avg2. 13755 13686 13143 96.03 95.55 95.79 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 114 Log-likelihood = -34927.510170 Norm (log-likelihood gradient vector) = 1796.235032 Norm (lambda vector) = 132.361857 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13680 13390 97.88 97.35 97.61 i-np 16033 16090 15670 97.39 97.74 97.56 o 23216 23234 22900 98.56 98.64 98.60 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.94 97.91 97.93 Avg2. 53004 53004 51960 98.03 98.03 98.03 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13680 13144 96.08 95.56 95.82 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.08 95.56 95.82 Avg2. 13755 13680 13144 96.08 95.56 95.82 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 115 Log-likelihood = -34128.029599 Norm (log-likelihood gradient vector) = 2136.566025 Norm (lambda vector) = 134.051414 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13689 13395 97.85 97.38 97.62 i-np 16033 16080 15660 97.39 97.67 97.53 o 23216 23235 22896 98.54 98.62 98.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.93 97.89 97.91 Avg2. 53004 53004 51951 98.01 98.01 98.01 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13689 13146 96.03 95.57 95.80 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.03 95.57 95.80 Avg2. 13755 13689 13146 96.03 95.57 95.80 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 116 Log-likelihood = -32746.393129 Norm (log-likelihood gradient vector) = 5505.023288 Norm (lambda vector) = 137.779411 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13697 13395 97.80 97.38 97.59 i-np 16033 16060 15646 97.42 97.59 97.50 o 23216 23247 22900 98.51 98.64 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.91 97.87 97.89 Avg2. 53004 53004 51941 97.99 97.99 97.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13697 13142 95.95 95.54 95.75 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.95 95.54 95.75 Avg2. 13755 13697 13142 95.95 95.54 95.75 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 117 Log-likelihood = -31870.916449 Norm (log-likelihood gradient vector) = 2192.729707 Norm (lambda vector) = 138.852991 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13706 13400 97.77 97.42 97.59 i-np 16033 16037 15634 97.49 97.51 97.50 o 23216 23261 22909 98.49 98.68 98.58 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.91 97.87 97.89 Avg2. 53004 53004 51943 98.00 98.00 98.00 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13706 13147 95.92 95.58 95.75 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.92 95.58 95.75 Avg2. 13755 13706 13147 95.92 95.58 95.75 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 118 Log-likelihood = -31645.588172 Norm (log-likelihood gradient vector) = 1470.110859 Norm (lambda vector) = 138.792126 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13714 13407 97.76 97.47 97.62 i-np 16033 16015 15630 97.60 97.49 97.54 o 23216 23275 22920 98.47 98.73 98.60 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.94 97.89 97.92 Avg2. 53004 53004 51957 98.02 98.02 98.02 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13714 13157 95.94 95.65 95.80 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.94 95.65 95.80 Avg2. 13755 13714 13157 95.94 95.65 95.80 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 119 Log-likelihood = -31333.352396 Norm (log-likelihood gradient vector) = 1633.565951 Norm (lambda vector) = 138.896597 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13716 13409 97.76 97.48 97.62 i-np 16033 16004 15624 97.63 97.45 97.54 o 23216 23284 22926 98.46 98.75 98.61 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.95 97.89 97.92 Avg2. 53004 53004 51959 98.03 98.03 98.03 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13716 13157 95.92 95.65 95.79 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.92 95.65 95.79 Avg2. 13755 13716 13157 95.92 95.65 95.79 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 120 Log-likelihood = -30897.459996 Norm (log-likelihood gradient vector) = 2111.885608 Norm (lambda vector) = 139.444133 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13681 13396 97.92 97.39 97.65 i-np 16033 16073 15659 97.42 97.67 97.55 o 23216 23250 22908 98.53 98.67 98.60 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.96 97.91 97.93 Avg2. 53004 53004 51963 98.04 98.04 98.04 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13681 13147 96.10 95.58 95.84 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.10 95.58 95.84 Avg2. 13755 13681 13147 96.10 95.58 95.84 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 121 Log-likelihood = -30374.180256 Norm (log-likelihood gradient vector) = 2990.696597 Norm (lambda vector) = 140.727889 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13718 13410 97.75 97.49 97.62 i-np 16033 16002 15621 97.62 97.43 97.52 o 23216 23284 22924 98.45 98.74 98.60 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.94 97.89 97.92 Avg2. 53004 53004 51955 98.02 98.02 98.02 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13718 13158 95.92 95.66 95.79 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.92 95.66 95.79 Avg2. 13755 13718 13158 95.92 95.66 95.79 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 122 Log-likelihood = -30121.612892 Norm (log-likelihood gradient vector) = 2171.633589 Norm (lambda vector) = 141.444608 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13709 13410 97.82 97.49 97.66 i-np 16033 16031 15640 97.56 97.55 97.55 o 23216 23264 22918 98.51 98.72 98.61 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.96 97.92 97.94 Avg2. 53004 53004 51968 98.05 98.05 98.05 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13709 13158 95.98 95.66 95.82 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.98 95.66 95.82 Avg2. 13755 13709 13158 95.98 95.66 95.82 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 123 Log-likelihood = -30096.135351 Norm (log-likelihood gradient vector) = 1546.450975 Norm (lambda vector) = 141.581115 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13695 13402 97.86 97.43 97.65 i-np 16033 16064 15653 97.44 97.63 97.54 o 23216 23245 22907 98.55 98.67 98.61 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.95 97.91 97.93 Avg2. 53004 53004 51962 98.03 98.03 98.03 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13695 13150 96.02 95.60 95.81 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.02 95.60 95.81 Avg2. 13755 13695 13150 96.02 95.60 95.81 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 124 Log-likelihood = -30081.262434 Norm (log-likelihood gradient vector) = 1353.619751 Norm (lambda vector) = 142.159962 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13694 13403 97.87 97.44 97.66 i-np 16033 16083 15666 97.41 97.71 97.56 o 23216 23227 22900 98.59 98.64 98.62 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.96 97.93 97.94 Avg2. 53004 53004 51969 98.05 98.05 98.05 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13694 13152 96.04 95.62 95.83 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.04 95.62 95.83 Avg2. 13755 13694 13152 96.04 95.62 95.83 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 125 Log-likelihood = -30038.497181 Norm (log-likelihood gradient vector) = 1812.584355 Norm (lambda vector) = 142.758090 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13691 13399 97.87 97.41 97.64 i-np 16033 16087 15669 97.40 97.73 97.57 o 23216 23226 22901 98.60 98.64 98.62 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.96 97.93 97.94 Avg2. 53004 53004 51969 98.05 98.05 98.05 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13691 13151 96.06 95.61 95.83 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.06 95.61 95.83 Avg2. 13755 13691 13151 96.06 95.61 95.83 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 33 seconds Iteration: 126 Log-likelihood = -29888.836475 Norm (log-likelihood gradient vector) = 2453.740477 Norm (lambda vector) = 143.828606 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13738 13408 97.60 97.48 97.54 i-np 16033 16001 15600 97.49 97.30 97.40 o 23216 23265 22908 98.47 98.67 98.57 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.85 97.82 97.83 Avg2. 53004 53004 51916 97.95 97.95 97.95 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13738 13132 95.59 95.47 95.53 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.59 95.47 95.53 Avg2. 13755 13738 13132 95.59 95.47 95.53 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 127 Log-likelihood = -30391.617749 Norm (log-likelihood gradient vector) = 8017.071354 Norm (lambda vector) = 146.182557 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13700 13398 97.80 97.40 97.60 i-np 16033 16077 15656 97.38 97.65 97.51 o 23216 23227 22895 98.57 98.62 98.59 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.92 97.89 97.90 Avg2. 53004 53004 51949 98.01 98.01 98.01 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13700 13143 95.93 95.55 95.74 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.93 95.55 95.74 Avg2. 13755 13700 13143 95.93 95.55 95.74 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 128 Log-likelihood = -29706.101275 Norm (log-likelihood gradient vector) = 2740.003388 Norm (lambda vector) = 144.638910 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13699 13401 97.82 97.43 97.63 i-np 16033 16067 15653 97.42 97.63 97.53 o 23216 23238 22905 98.57 98.66 98.61 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.94 97.91 97.92 Avg2. 53004 53004 51959 98.03 98.03 98.03 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13699 13147 95.97 95.58 95.77 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.97 95.58 95.77 Avg2. 13755 13699 13147 95.97 95.58 95.77 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 129 Log-likelihood = -29402.949309 Norm (log-likelihood gradient vector) = 1590.672116 Norm (lambda vector) = 145.134737 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13706 13407 97.82 97.47 97.64 i-np 16033 16038 15643 97.54 97.57 97.55 o 23216 23260 22920 98.54 98.73 98.63 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.96 97.92 97.94 Avg2. 53004 53004 51970 98.05 98.05 98.05 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13706 13153 95.97 95.62 95.79 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.97 95.62 95.79 Avg2. 13755 13706 13153 95.97 95.62 95.79 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 130 Log-likelihood = -28925.776455 Norm (log-likelihood gradient vector) = 1103.733870 Norm (lambda vector) = 145.506591 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13707 13413 97.86 97.51 97.68 i-np 16033 16031 15646 97.60 97.59 97.59 o 23216 23266 22926 98.54 98.75 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.00 97.95 97.97 Avg2. 53004 53004 51985 98.08 98.08 98.08 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13707 13162 96.02 95.69 95.86 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.02 95.69 95.86 Avg2. 13755 13707 13162 96.02 95.69 95.86 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 131 Log-likelihood = -28350.313622 Norm (log-likelihood gradient vector) = 1268.333021 Norm (lambda vector) = 145.891222 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13726 13422 97.79 97.58 97.68 i-np 16033 16012 15632 97.63 97.50 97.56 o 23216 23266 22925 98.53 98.75 98.64 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.98 97.94 97.96 Avg2. 53004 53004 51979 98.07 98.07 98.07 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13726 13166 95.92 95.72 95.82 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.92 95.72 95.82 Avg2. 13755 13726 13166 95.92 95.72 95.82 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 132 Log-likelihood = -27481.180667 Norm (log-likelihood gradient vector) = 2023.830394 Norm (lambda vector) = 146.857595 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13646 13389 98.12 97.34 97.73 i-np 16033 16154 15711 97.26 97.99 97.62 o 23216 23204 22898 98.68 98.63 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.02 97.99 98.00 Avg2. 53004 53004 51998 98.10 98.10 98.10 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13646 13149 96.36 95.59 95.97 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.36 95.59 95.97 Avg2. 13755 13646 13149 96.36 95.59 95.97 Current max chunk-based F1: 96.01 (iteration 90) Training iteration elapsed (including evaluation time): 33 seconds Iteration: 133 Log-likelihood = -26451.179889 Norm (log-likelihood gradient vector) = 5653.770011 Norm (lambda vector) = 148.945670 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13684 13411 98.00 97.50 97.75 i-np 16033 16083 15684 97.52 97.82 97.67 o 23216 23237 22921 98.64 98.73 98.68 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.05 98.02 98.04 Avg2. 53004 53004 52016 98.14 98.14 98.14 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13684 13172 96.26 95.76 96.01 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.26 95.76 96.01 Avg2. 13755 13684 13172 96.26 95.76 96.01 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 31 seconds Iteration: 134 Log-likelihood = -26783.739866 Norm (log-likelihood gradient vector) = 1877.676100 Norm (lambda vector) = 147.750099 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13701 13416 97.92 97.54 97.73 i-np 16033 16041 15653 97.58 97.63 97.61 o 23216 23262 22929 98.57 98.76 98.67 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.02 97.98 98.00 Avg2. 53004 53004 51998 98.10 98.10 98.10 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13701 13167 96.10 95.73 95.91 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.10 95.73 95.91 Avg2. 13755 13701 13167 96.10 95.73 95.91 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 135 Log-likelihood = -26292.178193 Norm (log-likelihood gradient vector) = 1100.390179 Norm (lambda vector) = 148.482497 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13714 13423 97.88 97.59 97.73 i-np 16033 16024 15646 97.64 97.59 97.61 o 23216 23266 22934 98.57 98.79 98.68 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.03 97.99 98.01 Avg2. 53004 53004 52003 98.11 98.11 98.11 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13714 13173 96.06 95.77 95.91 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.06 95.77 95.91 Avg2. 13755 13714 13173 96.06 95.77 95.91 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 136 Log-likelihood = -25934.405985 Norm (log-likelihood gradient vector) = 1137.852606 Norm (lambda vector) = 149.167336 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13716 13421 97.85 97.57 97.71 i-np 16033 16003 15631 97.68 97.49 97.58 o 23216 23285 22941 98.52 98.82 98.67 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.02 97.96 97.99 Avg2. 53004 53004 51993 98.09 98.09 98.09 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13716 13165 95.98 95.71 95.85 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.98 95.71 95.85 Avg2. 13755 13716 13165 95.98 95.71 95.85 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 137 Log-likelihood = -25303.260087 Norm (log-likelihood gradient vector) = 2718.719539 Norm (lambda vector) = 150.622231 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13708 13415 97.86 97.53 97.70 i-np 16033 16029 15641 97.58 97.56 97.57 o 23216 23267 22930 98.55 98.77 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.00 97.95 97.97 Avg2. 53004 53004 51986 98.08 98.08 98.08 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13708 13162 96.02 95.69 95.85 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.02 95.69 95.85 Avg2. 13755 13708 13162 96.02 95.69 95.85 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 138 Log-likelihood = -24905.097823 Norm (log-likelihood gradient vector) = 1133.411491 Norm (lambda vector) = 151.505892 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13703 13414 97.89 97.52 97.71 i-np 16033 16046 15652 97.54 97.62 97.58 o 23216 23255 22925 98.58 98.75 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.01 97.96 97.98 Avg2. 53004 53004 51991 98.09 98.09 98.09 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13703 13159 96.03 95.67 95.85 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.03 95.67 95.85 Avg2. 13755 13703 13159 96.03 95.67 95.85 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 139 Log-likelihood = -24717.920886 Norm (log-likelihood gradient vector) = 996.017290 Norm (lambda vector) = 151.839866 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13705 13413 97.87 97.51 97.69 i-np 16033 16034 15646 97.58 97.59 97.58 o 23216 23265 22930 98.56 98.77 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.00 97.96 97.98 Avg2. 53004 53004 51989 98.09 98.09 98.09 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13705 13161 96.03 95.68 95.86 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.03 95.68 95.86 Avg2. 13755 13705 13161 96.03 95.68 95.86 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 33 seconds Iteration: 140 Log-likelihood = -24337.182119 Norm (log-likelihood gradient vector) = 1467.624131 Norm (lambda vector) = 152.722997 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13698 13412 97.91 97.51 97.71 i-np 16033 16074 15666 97.46 97.71 97.59 o 23216 23232 22915 98.64 98.70 98.67 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.00 97.97 97.99 Avg2. 53004 53004 51993 98.09 98.09 98.09 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13698 13157 96.05 95.65 95.85 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.05 95.65 95.85 Avg2. 13755 13698 13157 96.05 95.65 95.85 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 141 Log-likelihood = -24070.371000 Norm (log-likelihood gradient vector) = 1641.517005 Norm (lambda vector) = 153.499400 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13695 13416 97.96 97.54 97.75 i-np 16033 16055 15666 97.58 97.71 97.64 o 23216 23254 22929 98.60 98.76 98.68 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.05 98.00 98.03 Avg2. 53004 53004 52011 98.13 98.13 98.13 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13695 13166 96.14 95.72 95.93 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.14 95.72 95.93 Avg2. 13755 13695 13166 96.14 95.72 95.93 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 142 Log-likelihood = -23754.439351 Norm (log-likelihood gradient vector) = 1079.789796 Norm (lambda vector) = 154.297217 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 31 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13677 13403 98.00 97.44 97.72 i-np 16033 16080 15676 97.49 97.77 97.63 o 23216 23247 22920 98.59 98.73 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.03 97.98 98.00 Avg2. 53004 53004 51999 98.10 98.10 98.10 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13677 13162 96.23 95.69 95.96 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.23 95.69 95.96 Avg2. 13755 13677 13162 96.23 95.69 95.96 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 143 Log-likelihood = -23359.789255 Norm (log-likelihood gradient vector) = 2368.274760 Norm (lambda vector) = 155.775486 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13725 13422 97.79 97.58 97.69 i-np 16033 15990 15620 97.69 97.42 97.55 o 23216 23289 22938 98.49 98.80 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.99 97.94 97.96 Avg2. 53004 53004 51980 98.07 98.07 98.07 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13725 13165 95.92 95.71 95.82 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.92 95.71 95.82 Avg2. 13755 13725 13165 95.92 95.71 95.82 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 144 Log-likelihood = -23231.778780 Norm (log-likelihood gradient vector) = 2423.906838 Norm (lambda vector) = 156.388007 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13706 13416 97.88 97.54 97.71 i-np 16033 16030 15646 97.60 97.59 97.60 o 23216 23268 22931 98.55 98.77 98.66 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.01 97.96 97.99 Avg2. 53004 53004 51993 98.09 98.09 98.09 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13706 13164 96.05 95.70 95.87 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.05 95.70 95.87 Avg2. 13755 13706 13164 96.05 95.70 95.87 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 145 Log-likelihood = -23178.117454 Norm (log-likelihood gradient vector) = 1097.506850 Norm (lambda vector) = 156.074473 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13693 13416 97.98 97.54 97.76 i-np 16033 16051 15666 97.60 97.71 97.66 o 23216 23260 22933 98.59 98.78 98.69 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.06 98.01 98.03 Avg2. 53004 53004 52015 98.13 98.13 98.13 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13693 13172 96.20 95.76 95.98 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.20 95.76 95.98 Avg2. 13755 13693 13172 96.20 95.76 95.98 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 146 Log-likelihood = -23093.774826 Norm (log-likelihood gradient vector) = 801.116663 Norm (lambda vector) = 156.123142 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13678 13406 98.01 97.46 97.74 i-np 16033 16073 15677 97.54 97.78 97.66 o 23216 23253 22928 98.60 98.76 98.68 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.05 98.00 98.03 Avg2. 53004 53004 52011 98.13 98.13 98.13 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13678 13164 96.24 95.70 95.97 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.24 95.70 95.97 Avg2. 13755 13678 13164 96.24 95.70 95.97 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 147 Log-likelihood = -22911.317470 Norm (log-likelihood gradient vector) = 1376.791560 Norm (lambda vector) = 156.710131 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13671 13401 98.03 97.43 97.72 i-np 16033 16092 15684 97.46 97.82 97.64 o 23216 23241 22921 98.62 98.73 98.68 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.04 97.99 98.02 Avg2. 53004 53004 52006 98.12 98.12 98.12 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13671 13158 96.25 95.66 95.95 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.25 95.66 95.95 Avg2. 13755 13671 13158 96.25 95.66 95.95 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 148 Log-likelihood = -22715.893435 Norm (log-likelihood gradient vector) = 1426.126026 Norm (lambda vector) = 157.688164 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13671 13397 98.00 97.40 97.70 i-np 16033 16106 15678 97.34 97.79 97.56 o 23216 23227 22909 98.63 98.68 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.99 97.95 97.97 Avg2. 53004 53004 51984 98.08 98.08 98.08 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13671 13144 96.15 95.56 95.85 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.15 95.56 95.85 Avg2. 13755 13671 13144 96.15 95.56 95.85 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 149 Log-likelihood = -22314.018839 Norm (log-likelihood gradient vector) = 2716.274548 Norm (lambda vector) = 160.102747 Log-likelihood and gradient computational time: 31 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13654 13388 98.05 97.33 97.69 i-np 16033 16140 15693 97.23 97.88 97.55 o 23216 23210 22899 98.66 98.63 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 97.98 97.95 97.96 Avg2. 53004 53004 51980 98.07 98.07 98.07 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13654 13138 96.22 95.51 95.87 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.22 95.51 95.87 Avg2. 13755 13654 13138 96.22 95.51 95.87 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 32 seconds Iteration: 150 Log-likelihood = -22188.196635 Norm (log-likelihood gradient vector) = 3394.614622 Norm (lambda vector) = 163.259555 Log-likelihood and gradient computational time: 32 seconds Training iteration elapsed: 32 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 13755 13670 13396 98.00 97.39 97.69 i-np 16033 16089 15673 97.41 97.75 97.58 o 23216 23245 22917 98.59 98.71 98.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 98.00 97.95 97.98 Avg2. 53004 53004 51986 98.08 98.08 98.08 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 13755 13670 13152 96.21 95.62 95.91 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.21 95.62 95.91 Avg2. 13755 13670 13152 96.21 95.62 95.91 Current max chunk-based F1: 96.01 (iteration 133) Training iteration elapsed (including evaluation time): 32 seconds The training process elapsed: 4804 seconds