📄 trainlog.txt_first-ordercrfs
字号:
OPTION VALUES:Model directory: ./Training data file: train.taggedTesting data file: test.taggedUnlabeled data file: data.untaggedLabel representation: IOB2Model file: model.txtTraining log file (this one): trainlog.txtFirst-order Markov CRFsNumber of labels: 3Number of training sequences: 412Number of testing sequences: 86Number of unlabeled sequences: 0Number of context predicates: 17487Number of features: 20375Feature rare threshold: 1Context predicate rare threshold: 1Using multiple rare thresholds for features: 0Highlight feature: 0Number of training iterations: 60Initial lambda value: 0.0000Sigma square (for smoothing): 100.0000Epsilon for L-BFGS convergence: 0.000100Number of approximated hessian matrixes: 7Start to train ...Iteration: 1 Log-likelihood = -10556.565482 Norm(log-likelihood gradient vector) = 3704.072407 Norm(lambda vector) = 0.000000 Iteration elapsed: 1 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 444 362 81.53 68.43 74.41 o 725 923 689 74.65 95.03 83.62 i-np 691 578 505 87.37 73.08 79.59 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.18 78.85 80.00 Avg2. 1945 1945 1556 80.00 80.00 80.00 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 444 279 62.84 52.74 57.35 ----- ------ ----- ----- ------- ------- ------------- Avg1. 62.84 52.74 57.35 Avg2. 529 444 279 62.84 52.74 57.35 Current max chunk-based F1: 57.35 (iteration 1) Training iteration elapsed (including testing & evaluation time): 1 secondsIteration: 2 Log-likelihood = -7481.017477 Norm(log-likelihood gradient vector) = 2489.172777 Norm(lambda vector) = 1.000000 Iteration elapsed: 2 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 529 421 79.58 79.58 79.58 o 725 689 610 88.53 84.14 86.28 i-np 691 727 591 81.29 85.53 83.36 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.14 83.08 83.11 Avg2. 1945 1945 1622 83.39 83.39 83.39 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 529 328 62.00 62.00 62.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 62.00 62.00 62.00 Avg2. 529 529 328 62.00 62.00 62.00 Current max chunk-based F1: 62.00 (iteration 2) Training iteration elapsed (including testing & evaluation time): 2 secondsIteration: 3 Log-likelihood = -3993.656395 Norm(log-likelihood gradient vector) = 1366.205829 Norm(lambda vector) = 3.639198 Iteration elapsed: 2 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 432 397 91.90 75.05 82.62 o 725 882 714 80.95 98.48 88.86 i-np 691 631 586 92.87 84.80 88.65 ----- ------ ----- ----- ------- ------- ------------- Avg1. 88.57 86.11 87.32 Avg2. 1945 1945 1697 87.25 87.25 87.25 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 432 338 78.24 63.89 70.34 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.24 63.89 70.34 Avg2. 529 432 338 78.24 63.89 70.34 Current max chunk-based F1: 70.34 (iteration 3) Training iteration elapsed (including testing & evaluation time): 2 secondsIteration: 4 Log-likelihood = -2887.436054 Norm(log-likelihood gradient vector) = 1091.963943 Norm(lambda vector) = 5.146294 Iteration elapsed: 1 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 459 427 93.03 80.72 86.44 o 725 811 709 87.42 97.79 92.32 i-np 691 675 621 92.00 89.87 90.92 ----- ------ ----- ----- ------- ------- ------------- Avg1. 90.82 89.46 90.13 Avg2. 1945 1945 1757 90.33 90.33 90.33 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 459 367 79.96 69.38 74.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 79.96 69.38 74.29 Avg2. 529 459 367 79.96 69.38 74.29 Current max chunk-based F1: 74.29 (iteration 4) Training iteration elapsed (including testing & evaluation time): 2 secondsIteration: 5 Log-likelihood = -2412.831742 Norm(log-likelihood gradient vector) = 638.899561 Norm(lambda vector) = 6.192059 Iteration elapsed: 1 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 522 486 93.10 91.87 92.48 o 725 765 712 93.07 98.21 95.57 i-np 691 658 632 96.05 91.46 93.70 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.07 93.85 93.96 Avg2. 1945 1945 1830 94.09 94.09 94.09 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 522 447 85.63 84.50 85.06 ----- ------ ----- ----- ------- ------- ------------- Avg1. 85.63 84.50 85.06 Avg2. 529 522 447 85.63 84.50 85.06 Current max chunk-based F1: 85.06 (iteration 5) Training iteration elapsed (including testing & evaluation time): 1 secondsIteration: 6 Log-likelihood = -1849.524795 Norm(log-likelihood gradient vector) = 380.557928 Norm(lambda vector) = 8.771722 Iteration elapsed: 2 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 483 464 96.07 87.71 91.70 o 725 754 703 93.24 96.97 95.06 i-np 691 708 657 92.80 95.08 93.92 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.03 93.25 93.64 Avg2. 1945 1945 1824 93.78 93.78 93.78 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 483 425 87.99 80.34 83.99 ----- ------ ----- ----- ------- ------- ------------- Avg1. 87.99 80.34 83.99 Avg2. 529 483 425 87.99 80.34 83.99 Current max chunk-based F1: 85.06 (iteration 5) Training iteration elapsed (including testing & evaluation time): 2 secondsIteration: 7 Log-likelihood = -1700.516200 Norm(log-likelihood gradient vector) = 572.568532 Norm(lambda vector) = 12.267256 Iteration elapsed: 2 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 522 498 95.40 94.14 94.77 o 725 756 716 94.71 98.76 96.69 i-np 691 667 645 96.70 93.34 94.99 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.60 95.41 95.51 Avg2. 1945 1945 1859 95.58 95.58 95.58 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 522 465 89.08 87.90 88.49 ----- ------ ----- ----- ------- ------- ------------- Avg1. 89.08 87.90 88.49 Avg2. 529 522 465 89.08 87.90 88.49 Current max chunk-based F1: 88.49 (iteration 7) Training iteration elapsed (including testing & evaluation time): 2 secondsIteration: 8 Log-likelihood = -1359.536303 Norm(log-likelihood gradient vector) = 198.620212 Norm(lambda vector) = 14.885930 Iteration elapsed: 1 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 529 505 95.46 95.46 95.46 o 725 757 719 94.98 99.17 97.03 i-np 691 659 644 97.72 93.20 95.41 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.06 95.94 96.00 Avg2. 1945 1945 1868 96.04 96.04 96.04 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 529 473 89.41 89.41 89.41 ----- ------ ----- ----- ------- ------- ------------- Avg1. 89.41 89.41 89.41 Avg2. 529 529 473 89.41 89.41 89.41 Current max chunk-based F1: 89.41 (iteration 8) Training iteration elapsed (including testing & evaluation time): 1 secondsIteration: 9 Log-likelihood = -1243.393080 Norm(log-likelihood gradient vector) = 160.689764 Norm(lambda vector) = 17.131725 Iteration elapsed: 2 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 525 507 96.57 95.84 96.20 o 725 754 722 95.76 99.59 97.63 i-np 691 666 652 97.90 94.36 96.09 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.74 96.59 96.67 Avg2. 1945 1945 1881 96.71 96.71 96.71 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 525 478 91.05 90.36 90.70 ----- ------ ----- ----- ------- ------- ------------- Avg1. 91.05 90.36 90.70 Avg2. 529 525 478 91.05 90.36 90.70 Current max chunk-based F1: 90.70 (iteration 9) Training iteration elapsed (including testing & evaluation time): 2 secondsIteration: 10 Log-likelihood = -1058.962867 Norm(log-likelihood gradient vector) = 125.854939 Norm(lambda vector) = 20.572535 Iteration elapsed: 2 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 520 502 96.54 94.90 95.71 o 725 743 715 96.23 98.62 97.41 i-np 691 682 655 96.04 94.79 95.41 ----- ------ ----- ----- ------- ------- ------------- Avg1. 96.27 96.10 96.19 Avg2. 1945 1945 1872 96.25 96.25 96.25 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 520 470 90.38 88.85 89.61 ----- ------ ----- ----- ------- ------- ------------- Avg1. 90.38 88.85 89.61 Avg2. 529 520 470 90.38 88.85 89.61 Current max chunk-based F1: 90.70 (iteration 9) Training iteration elapsed (including testing & evaluation time): 2 secondsIteration: 11 Log-likelihood = -799.153829 Norm(log-likelihood gradient vector) = 193.296906 Norm(lambda vector) = 25.431154 Iteration elapsed: 1 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 525 503 95.81 95.09 95.45 o 725 751 717 95.47 98.90 97.15 i-np 691 669 646 96.56 93.49 95.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.95 95.82 95.89 Avg2. 1945 1945 1866 95.94 95.94 95.94 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 525 468 89.14 88.47 88.80 ----- ------ ----- ----- ------- ------- ------------- Avg1. 89.14 88.47 88.80 Avg2. 529 525 468 89.14 88.47 88.80 Current max chunk-based F1: 90.70 (iteration 9) Training iteration elapsed (including testing & evaluation time): 1 secondsIteration: 12 Log-likelihood = -614.403785 Norm(log-likelihood gradient vector) = 142.554404 Norm(lambda vector) = 31.354324 Iteration elapsed: 2 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%)⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -