📄 trainlog.txt
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.txtSecond-order Markov CRFsNumber of labels: 4Number of training sequences: 412Number of testing sequences: 86Number of unlabeled sequences: 0Number of context predicates: 10807Number of features: 17259Feature 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 = -26641.805032 Norm(log-likelihood gradient vector) = 5327.595333 Norm(lambda vector) = 0.000000 Iteration elapsed: 3 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 491 371 75.56 70.13 72.75 o 725 828 639 77.17 88.14 82.29 i-np 691 626 508 81.15 73.52 77.15 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.96 77.26 77.61 Avg2. 1945 1945 1518 78.05 78.05 78.05 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 491 269 54.79 50.85 52.75 ----- ------ ----- ----- ------- ------- ------------- Avg1. 54.79 50.85 52.75 Avg2. 529 491 269 54.79 50.85 52.75 Current max chunk-based F1: 52.75 (iteration 1) Training iteration elapsed (including testing & evaluation time): 3 secondsIteration: 2 Log-likelihood = -21582.014208 Norm(log-likelihood gradient vector) = 4677.718113 Norm(lambda vector) = 1.000000 Iteration elapsed: 2 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 486 388 79.84 73.35 76.45 o 725 786 648 82.44 89.38 85.77 i-np 691 673 551 81.87 79.74 80.79 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.38 80.82 81.10 Avg2. 1945 1945 1587 81.59 81.59 81.59 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 486 293 60.29 55.39 57.73 ----- ------ ----- ----- ------- ------- ------------- Avg1. 60.29 55.39 57.73 Avg2. 529 486 293 60.29 55.39 57.73 Current max chunk-based F1: 57.73 (iteration 2) Training iteration elapsed (including testing & evaluation time): 3 secondsIteration: 3 Log-likelihood = -9269.463734 Norm(log-likelihood gradient vector) = 3788.255409 Norm(lambda vector) = 10.401613 Iteration elapsed: 2 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 402 380 94.53 71.83 81.63 o 725 830 696 83.86 96.00 89.52 i-np 691 713 614 86.12 88.86 87.46 ----- ------ ----- ----- ------- ------- ------------- Avg1. 88.17 85.56 86.85 Avg2. 1945 1945 1690 86.89 86.89 86.89 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 402 310 77.11 58.60 66.60 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.11 58.60 66.60 Avg2. 529 402 310 77.11 58.60 66.60 Current max chunk-based F1: 66.60 (iteration 3) Training iteration elapsed (including testing & evaluation time): 2 secondsIteration: 4 Log-likelihood = -6725.988433 Norm(log-likelihood gradient vector) = 1681.165107 Norm(lambda vector) = 9.913913 Iteration elapsed: 3 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 506 459 90.71 86.77 88.70 o 725 778 707 90.87 97.52 94.08 i-np 691 661 607 91.83 87.84 89.79 ----- ------ ----- ----- ------- ------- ------------- Avg1. 91.14 90.71 90.92 Avg2. 1945 1945 1773 91.16 91.16 91.16 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 506 387 76.48 73.16 74.78 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.48 73.16 74.78 Avg2. 529 506 387 76.48 73.16 74.78 Current max chunk-based F1: 74.78 (iteration 4) Training iteration elapsed (including testing & evaluation time): 3 secondsIteration: 5 Log-likelihood = -6270.380695 Norm(log-likelihood gradient vector) = 1015.841971 Norm(lambda vector) = 9.252770 Iteration elapsed: 3 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 508 458 90.16 86.58 88.33 o 725 800 713 89.12 98.34 93.51 i-np 691 637 592 92.94 85.67 89.16 ----- ------ ----- ----- ------- ------- ------------- Avg1. 90.74 90.20 90.47 Avg2. 1945 1945 1763 90.64 90.64 90.64 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 508 388 76.38 73.35 74.83 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.38 73.35 74.83 Avg2. 529 508 388 76.38 73.35 74.83 Current max chunk-based F1: 74.83 (iteration 5) Training iteration elapsed (including testing & evaluation time): 3 secondsIteration: 6 Log-likelihood = -5969.689414 Norm(log-likelihood gradient vector) = 650.921272 Norm(lambda vector) = 9.234898 Iteration elapsed: 2 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 520 481 92.50 90.93 91.71 o 725 791 715 90.39 98.62 94.33 i-np 691 634 608 95.90 87.99 91.77 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.93 92.51 92.72 Avg2. 1945 1945 1804 92.75 92.75 92.75 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 520 431 82.88 81.47 82.17 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.88 81.47 82.17 Avg2. 529 520 431 82.88 81.47 82.17 Current max chunk-based F1: 82.17 (iteration 6) Training iteration elapsed (including testing & evaluation time): 2 secondsIteration: 7 Log-likelihood = -5516.616482 Norm(log-likelihood gradient vector) = 606.694867 Norm(lambda vector) = 9.900934 Iteration elapsed: 3 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 516 485 93.99 91.68 92.82 o 725 776 717 92.40 98.90 95.54 i-np 691 653 630 96.48 91.17 93.75 ----- ------ ----- ----- ------- ------- ------------- Avg1. 94.29 93.92 94.10 Avg2. 1945 1945 1832 94.19 94.19 94.19 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 516 447 86.63 84.50 85.55 ----- ------ ----- ----- ------- ------- ------------- Avg1. 86.63 84.50 85.55 Avg2. 529 516 447 86.63 84.50 85.55 Current max chunk-based F1: 85.55 (iteration 7) Training iteration elapsed (including testing & evaluation time): 3 secondsIteration: 8 Log-likelihood = -4872.376953 Norm(log-likelihood gradient vector) = 569.146507 Norm(lambda vector) = 11.534266 Iteration elapsed: 3 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 510 491 96.27 92.82 94.51 o 725 741 710 95.82 97.93 96.86 i-np 691 694 660 95.10 95.51 95.31 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.73 95.42 95.58 Avg2. 1945 1945 1861 95.68 95.68 95.68 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 510 458 89.80 86.58 88.16 ----- ------ ----- ----- ------- ------- ------------- Avg1. 89.80 86.58 88.16 Avg2. 529 510 458 89.80 86.58 88.16 Current max chunk-based F1: 88.16 (iteration 8) Training iteration elapsed (including testing & evaluation time): 3 secondsIteration: 9 Log-likelihood = -4079.737290 Norm(log-likelihood gradient vector) = 795.363991 Norm(lambda vector) = 15.318898 Iteration elapsed: 2 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 526 502 95.44 94.90 95.17 o 725 763 722 94.63 99.59 97.04 i-np 691 656 641 97.71 92.76 95.17 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.93 95.75 95.84 Avg2. 1945 1945 1865 95.89 95.89 95.89 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 526 471 89.54 89.04 89.29 ----- ------ ----- ----- ------- ------- ------------- Avg1. 89.54 89.04 89.29 Avg2. 529 526 471 89.54 89.04 89.29 Current max chunk-based F1: 89.29 (iteration 9) Training iteration elapsed (including testing & evaluation time): 3 secondsIteration: 10 Log-likelihood = -3616.619518 Norm(log-likelihood gradient vector) = 368.298852 Norm(lambda vector) = 17.278879 Iteration elapsed: 2 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 527 503 95.45 95.09 95.27 o 725 762 722 94.75 99.59 97.11 i-np 691 656 641 97.71 92.76 95.17 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.97 95.81 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 529 527 472 89.56 89.22 89.39 ----- ------ ----- ----- ------- ------- ------------- Avg1. 89.56 89.22 89.39 Avg2. 529 527 472 89.56 89.22 89.39 Current max chunk-based F1: 89.39 (iteration 10) Training iteration elapsed (including testing & evaluation time): 2 secondsIteration: 11 Log-likelihood = -3401.327153 Norm(log-likelihood gradient vector) = 344.623130 Norm(lambda vector) = 19.439680 Iteration elapsed: 3 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- b-np 529 529 503 95.09 95.09 95.09 o 725 762 721 94.62 99.45 96.97 i-np 691 654 640 97.86 92.62 95.17 ----- ------ ----- ----- ------- ------- ------------- Avg1. 95.85 95.72 95.79 Avg2. 1945 1945 1864 95.84 95.84 95.84 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 529 529 474 89.60 89.60 89.60 ----- ------ ----- ----- ------- ------- ------------- Avg1. 89.60 89.60 89.60 Avg2. 529 529 474 89.60 89.60 89.60 Current max chunk-based F1: 89.60 (iteration 11) Training iteration elapsed (including testing & evaluation time): 3 secondsIteration: 12 Log-likelihood = -3077.749137 Norm(log-likelihood gradient vector) = 297.771105 Norm(lambda vector) = 23.055484 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 + -