Hugging Face's logo

IrwinD
/
flan-log-sage

Summarization
Transformers
Safetensors
t5
text2text-generation
Generated from Trainer
Eval Results
text-generation-inference
Inference Endpoints
Model card Files Community
flan-log-sage / README.md
IrwinD's picture
IrwinD
End of training
467d2d3 verified
preview code
|
raw
history blame
21.8 kB
metadata
license: apache-2.0
base_model: google/flan-t5-small
tags:
  - generated_from_trainer
datasets:
  - hdfs_log_summary_dataset
metrics:
  - rouge
model-index:
  - name: flan-log-sage
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: hdfs_log_summary_dataset
          type: hdfs_log_summary_dataset
          config: default
          split: test
          args: default
        metrics:
          - name: Rouge1
            type: rouge
            value: 0.3738

flan-log-sage

This model is a fine-tuned version of google/flan-t5-small on the hdfs_log_summary_dataset dataset. It achieves the following results on the evaluation set:

  • Loss: 2.0636
  • Rouge1: 0.3738
  • Rouge2: 0.1028
  • Rougel: 0.2953
  • Rougelsum: 0.2962
  • Gen Len: 19.0

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 2e-05
  • train_batch_size: 6
  • eval_batch_size: 6
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 200
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum Gen Len
No log 1.0 7 3.9522 0.1033 0.034 0.1025 0.1023 19.0
No log 2.0 14 3.7383 0.0895 0.0152 0.0895 0.0903 19.0
No log 3.0 21 3.5753 0.184 0.0043 0.1365 0.1364 19.0
No log 4.0 28 3.4379 0.1978 0.0043 0.1442 0.1437 19.0
No log 5.0 35 3.3177 0.1967 0.0043 0.1479 0.1473 19.0
No log 6.0 42 3.2163 0.2099 0.0157 0.1487 0.1482 19.0
No log 7.0 49 3.1227 0.2061 0.0115 0.1533 0.153 18.6
No log 8.0 56 3.0415 0.1742 0.0165 0.1209 0.1187 16.8
No log 9.0 63 2.9693 0.2544 0.0452 0.2011 0.2013 18.1
No log 10.0 70 2.8964 0.2847 0.0431 0.2125 0.2127 18.4
No log 11.0 77 2.8268 0.292 0.041 0.2144 0.2139 18.9
No log 12.0 84 2.7622 0.316 0.0413 0.2289 0.2286 19.0
No log 13.0 91 2.7027 0.316 0.0413 0.2289 0.2286 19.0
No log 14.0 98 2.6504 0.3384 0.0535 0.2504 0.2498 19.0
No log 15.0 105 2.5991 0.332 0.0657 0.2351 0.2345 17.8
No log 16.0 112 2.5532 0.323 0.0774 0.2514 0.2507 17.9
No log 17.0 119 2.5090 0.3752 0.0989 0.3083 0.3067 18.8
No log 18.0 126 2.4714 0.4073 0.1829 0.363 0.3602 19.0
No log 19.0 133 2.4451 0.4031 0.1918 0.3644 0.3628 19.0
No log 20.0 140 2.4147 0.3988 0.1825 0.3444 0.3428 19.0
No log 21.0 147 2.3924 0.4112 0.1871 0.3496 0.3479 19.0
No log 22.0 154 2.3742 0.4413 0.1989 0.3648 0.3621 19.0
No log 23.0 161 2.3566 0.469 0.2249 0.3902 0.386 19.0
No log 24.0 168 2.3394 0.4573 0.2049 0.3761 0.371 19.0
No log 25.0 175 2.3192 0.4612 0.2045 0.3751 0.3695 19.0
No log 26.0 182 2.3011 0.4606 0.2153 0.3791 0.3741 19.0
No log 27.0 189 2.2876 0.4568 0.1974 0.3697 0.3657 19.0
No log 28.0 196 2.2773 0.452 0.1944 0.3694 0.3655 19.0
No log 29.0 203 2.2647 0.4392 0.1819 0.3649 0.3619 19.0
No log 30.0 210 2.2535 0.4234 0.1683 0.359 0.3567 19.0
No log 31.0 217 2.2380 0.4234 0.1614 0.359 0.3567 19.0
No log 32.0 224 2.2316 0.4189 0.1645 0.3464 0.3425 19.0
No log 33.0 231 2.2243 0.4239 0.1696 0.352 0.348 19.0
No log 34.0 238 2.2169 0.4227 0.1653 0.359 0.3567 19.0
No log 35.0 245 2.2017 0.4227 0.1653 0.359 0.3567 19.0
No log 36.0 252 2.1875 0.4176 0.1653 0.3542 0.3511 19.0
No log 37.0 259 2.1828 0.4209 0.169 0.3542 0.3511 19.0
No log 38.0 266 2.1778 0.431 0.1746 0.3555 0.3503 19.0
No log 39.0 273 2.1683 0.4408 0.1847 0.3555 0.3503 19.0
No log 40.0 280 2.1626 0.4387 0.1836 0.3589 0.3537 19.0
No log 41.0 287 2.1544 0.4297 0.1732 0.3537 0.3495 19.0
No log 42.0 294 2.1496 0.4393 0.1771 0.3566 0.3525 19.0
No log 43.0 301 2.1433 0.4345 0.1853 0.3656 0.3626 19.0
No log 44.0 308 2.1347 0.426 0.1718 0.3613 0.3581 19.0
No log 45.0 315 2.1235 0.426 0.1718 0.3576 0.3546 19.0
No log 46.0 322 2.1172 0.4188 0.1682 0.3621 0.3571 19.0
No log 47.0 329 2.1149 0.4188 0.1682 0.3621 0.3571 19.0
No log 48.0 336 2.1124 0.4172 0.1643 0.3539 0.3484 19.0
No log 49.0 343 2.1091 0.4465 0.19 0.3659 0.3609 19.0
No log 50.0 350 2.1041 0.449 0.2075 0.3769 0.3742 19.0
No log 51.0 357 2.0955 0.449 0.2075 0.3769 0.3742 19.0
No log 52.0 364 2.0906 0.4449 0.2077 0.3819 0.3791 19.0
No log 53.0 371 2.0858 0.4332 0.1597 0.3566 0.3535 19.0
No log 54.0 378 2.0800 0.4373 0.1878 0.375 0.3711 19.0
No log 55.0 385 2.0839 0.4216 0.1719 0.3652 0.3612 19.0
No log 56.0 392 2.0877 0.4216 0.1719 0.3652 0.3612 19.0
No log 57.0 399 2.0862 0.4216 0.1719 0.3652 0.3612 19.0
No log 58.0 406 2.0817 0.4472 0.1833 0.3689 0.3675 19.0
No log 59.0 413 2.0783 0.4564 0.1967 0.3732 0.3679 19.0
No log 60.0 420 2.0704 0.4564 0.1934 0.3732 0.3679 19.0
No log 61.0 427 2.0649 0.4566 0.2001 0.381 0.3769 19.0
No log 62.0 434 2.0618 0.4446 0.1931 0.3691 0.3643 19.0
No log 63.0 441 2.0566 0.4418 0.1931 0.3662 0.3611 19.0
No log 64.0 448 2.0469 0.4494 0.2075 0.3772 0.3742 19.0
No log 65.0 455 2.0500 0.4494 0.2075 0.3772 0.3742 19.0
No log 66.0 462 2.0504 0.4494 0.2075 0.3772 0.3742 19.0
No log 67.0 469 2.0531 0.4494 0.2075 0.3772 0.3742 19.0
No log 68.0 476 2.0540 0.4418 0.19 0.3662 0.3611 19.0
No log 69.0 483 2.0494 0.4418 0.1931 0.3662 0.3611 19.0
No log 70.0 490 2.0578 0.4258 0.1769 0.3653 0.3611 19.0
No log 71.0 497 2.0632 0.4293 0.1769 0.3685 0.3645 19.0
1.9892 72.0 504 2.0639 0.4293 0.1769 0.3685 0.3645 19.0
1.9892 73.0 511 2.0627 0.4258 0.1682 0.3646 0.361 19.0
1.9892 74.0 518 2.0551 0.421 0.1643 0.3539 0.3483 19.0
1.9892 75.0 525 2.0444 0.421 0.1643 0.3539 0.3483 19.0
1.9892 76.0 532 2.0428 0.4258 0.1731 0.3653 0.3611 19.0
1.9892 77.0 539 2.0509 0.4255 0.1731 0.3652 0.3609 19.0
1.9892 78.0 546 2.0566 0.4207 0.1643 0.3537 0.3481 19.0
1.9892 79.0 553 2.0575 0.4359 0.1876 0.37 0.3654 19.0
1.9892 80.0 560 2.0479 0.4331 0.1566 0.3432 0.3394 19.0
1.9892 81.0 567 2.0430 0.4334 0.1566 0.3434 0.3396 19.0
1.9892 82.0 574 2.0377 0.4334 0.1566 0.3434 0.3396 19.0
1.9892 83.0 581 2.0349 0.4331 0.1566 0.3432 0.3394 19.0
1.9892 84.0 588 2.0338 0.4331 0.1566 0.3432 0.3394 19.0
1.9892 85.0 595 2.0345 0.4331 0.1566 0.3432 0.3394 19.0
1.9892 86.0 602 2.0359 0.4535 0.1962 0.3792 0.3735 19.0
1.9892 87.0 609 2.0338 0.4535 0.1962 0.3792 0.3735 19.0
1.9892 88.0 616 2.0456 0.4249 0.1728 0.3563 0.3512 19.0
1.9892 89.0 623 2.0556 0.4255 0.1731 0.3652 0.3609 19.0
1.9892 90.0 630 2.0532 0.4249 0.1728 0.3563 0.3512 19.0
1.9892 91.0 637 2.0461 0.4535 0.1962 0.3792 0.3735 19.0
1.9892 92.0 644 2.0395 0.4507 0.1962 0.3794 0.3736 19.0
1.9892 93.0 651 2.0314 0.4507 0.1962 0.3794 0.3736 19.0
1.9892 94.0 658 2.0308 0.4212 0.1728 0.3565 0.3513 19.0
1.9892 95.0 665 2.0359 0.4264 0.1826 0.3691 0.3647 19.0
1.9892 96.0 672 2.0418 0.4223 0.1731 0.3653 0.3611 19.0
1.9892 97.0 679 2.0400 0.4155 0.1682 0.3626 0.3576 19.0
1.9892 98.0 686 2.0303 0.4155 0.1682 0.3626 0.3576 19.0
1.9892 99.0 693 2.0311 0.4155 0.1682 0.3626 0.3576 19.0
1.9892 100.0 700 2.0339 0.4155 0.1682 0.3626 0.3576 19.0
1.9892 101.0 707 2.0359 0.4058 0.1585 0.3385 0.3378 19.0
1.9892 102.0 714 2.0352 0.4085 0.1583 0.3488 0.3475 19.0
1.9892 103.0 721 2.0339 0.4264 0.1643 0.36 0.3578 19.0
1.9892 104.0 728 2.0419 0.4204 0.1603 0.3529 0.3516 19.0
1.9892 105.0 735 2.0433 0.4031 0.1568 0.3427 0.3377 19.0
1.9892 106.0 742 2.0400 0.4137 0.1643 0.3578 0.3524 19.0
1.9892 107.0 749 2.0400 0.4226 0.1692 0.3641 0.361 19.0
1.9892 108.0 756 2.0401 0.4226 0.1692 0.3641 0.361 19.0
1.9892 109.0 763 2.0416 0.4054 0.1564 0.3491 0.3437 19.0
1.9892 110.0 770 2.0372 0.4137 0.1643 0.3578 0.3524 19.0
1.9892 111.0 777 2.0403 0.4211 0.1692 0.3658 0.3597 19.0
1.9892 112.0 784 2.0379 0.4156 0.1612 0.3586 0.3549 19.0
1.9892 113.0 791 2.0369 0.4266 0.1739 0.3683 0.3655 19.0
1.9892 114.0 798 2.0342 0.4296 0.1373 0.3372 0.3367 19.0
1.9892 115.0 805 2.0294 0.4278 0.1376 0.3316 0.3311 19.0
1.9892 116.0 812 2.0272 0.4495 0.1845 0.3704 0.369 19.0
1.9892 117.0 819 2.0245 0.4042 0.1545 0.3464 0.3446 19.0
1.9892 118.0 826 2.0214 0.4149 0.1689 0.3566 0.3539 19.0
1.9892 119.0 833 2.0177 0.4181 0.1373 0.3326 0.3318 19.0
1.9892 120.0 840 2.0206 0.4403 0.1798 0.3627 0.3644 19.0
1.9892 121.0 847 2.0258 0.4042 0.1545 0.3464 0.3446 19.0
1.9892 122.0 854 2.0264 0.4165 0.1373 0.3265 0.3256 19.0
1.9892 123.0 861 2.0250 0.4104 0.1231 0.3161 0.3158 19.0
1.9892 124.0 868 2.0268 0.389 0.1097 0.3055 0.3064 19.0
1.9892 125.0 875 2.0286 0.3945 0.1099 0.3067 0.307 19.0
1.9892 126.0 882 2.0300 0.3893 0.1105 0.3011 0.3011 19.0
1.9892 127.0 889 2.0358 0.4174 0.1562 0.3385 0.3401 19.0
1.9892 128.0 896 2.0387 0.3873 0.1473 0.3284 0.3286 19.0
1.9892 129.0 903 2.0426 0.3869 0.1493 0.3263 0.3267 19.0
1.9892 130.0 910 2.0440 0.4052 0.1616 0.3422 0.3444 19.0
1.9892 131.0 917 2.0471 0.4255 0.1719 0.3685 0.3652 19.0
1.9892 132.0 924 2.0479 0.4255 0.1719 0.3685 0.3652 19.0
1.9892 133.0 931 2.0469 0.4255 0.1719 0.3685 0.3652 19.0
1.9892 134.0 938 2.0428 0.4255 0.1719 0.3685 0.3652 19.0
1.9892 135.0 945 2.0383 0.4052 0.1616 0.3422 0.3444 19.0
1.9892 136.0 952 2.0322 0.4052 0.1616 0.3422 0.3444 19.0
1.9892 137.0 959 2.0371 0.4255 0.1719 0.3685 0.3652 19.0
1.9892 138.0 966 2.0406 0.4255 0.1719 0.3685 0.3652 19.0
1.9892 139.0 973 2.0466 0.4255 0.1719 0.3685 0.3652 19.0
1.9892 140.0 980 2.0515 0.4255 0.1719 0.3685 0.3652 19.0
1.9892 141.0 987 2.0546 0.4255 0.1719 0.3685 0.3652 19.0
1.9892 142.0 994 2.0581 0.4255 0.1719 0.3685 0.3652 19.0
1.0228 143.0 1001 2.0591 0.4255 0.1719 0.3685 0.3652 19.0
1.0228 144.0 1008 2.0599 0.4255 0.1719 0.3685 0.3652 19.0
1.0228 145.0 1015 2.0608 0.4016 0.1573 0.3387 0.3408 19.0
1.0228 146.0 1022 2.0583 0.4016 0.1573 0.3387 0.3408 19.0
1.0228 147.0 1029 2.0512 0.3814 0.1426 0.3263 0.328 19.0
1.0228 148.0 1036 2.0477 0.3844 0.1426 0.3225 0.3238 19.0
1.0228 149.0 1043 2.0428 0.3944 0.1441 0.3266 0.3287 19.0
1.0228 150.0 1050 2.0393 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 151.0 1057 2.0412 0.3944 0.1441 0.3266 0.3287 19.0
1.0228 152.0 1064 2.0437 0.3844 0.1426 0.3225 0.3238 19.0
1.0228 153.0 1071 2.0433 0.3814 0.1426 0.3263 0.328 19.0
1.0228 154.0 1078 2.0468 0.3829 0.1448 0.3232 0.3223 19.0
1.0228 155.0 1085 2.0512 0.3814 0.1426 0.3263 0.328 19.0
1.0228 156.0 1092 2.0545 0.3814 0.1426 0.3263 0.328 19.0
1.0228 157.0 1099 2.0582 0.3814 0.1426 0.3263 0.328 19.0
1.0228 158.0 1106 2.0624 0.3844 0.1426 0.3225 0.3238 19.0
1.0228 159.0 1113 2.0659 0.3844 0.1426 0.3225 0.3238 19.0
1.0228 160.0 1120 2.0690 0.3814 0.1426 0.3263 0.328 19.0
1.0228 161.0 1127 2.0691 0.3814 0.1426 0.3263 0.328 19.0
1.0228 162.0 1134 2.0675 0.3847 0.1464 0.3303 0.3315 19.0
1.0228 163.0 1141 2.0653 0.3847 0.1464 0.3303 0.3315 19.0
1.0228 164.0 1148 2.0641 0.3847 0.1464 0.3303 0.3315 19.0
1.0228 165.0 1155 2.0649 0.3847 0.1464 0.3303 0.3315 19.0
1.0228 166.0 1162 2.0648 0.3869 0.1493 0.3263 0.3267 19.0
1.0228 167.0 1169 2.0627 0.3869 0.1493 0.3263 0.3267 19.0
1.0228 168.0 1176 2.0610 0.3844 0.1426 0.3225 0.3238 19.0
1.0228 169.0 1183 2.0617 0.3844 0.1426 0.3225 0.3238 19.0
1.0228 170.0 1190 2.0617 0.3844 0.1426 0.3225 0.3238 19.0
1.0228 171.0 1197 2.0594 0.3844 0.1426 0.3225 0.3238 19.0
1.0228 172.0 1204 2.0579 0.3844 0.1426 0.3225 0.3238 19.0
1.0228 173.0 1211 2.0578 0.3844 0.1426 0.3225 0.3238 19.0
1.0228 174.0 1218 2.0584 0.3844 0.1426 0.3225 0.3238 19.0
1.0228 175.0 1225 2.0588 0.3847 0.1464 0.3303 0.3315 19.0
1.0228 176.0 1232 2.0585 0.3847 0.1464 0.3303 0.3315 19.0
1.0228 177.0 1239 2.0579 0.3847 0.1464 0.3303 0.3315 19.0
1.0228 178.0 1246 2.0568 0.3844 0.1426 0.3225 0.3238 19.0
1.0228 179.0 1253 2.0549 0.3944 0.1441 0.3266 0.3287 19.0
1.0228 180.0 1260 2.0553 0.3944 0.1441 0.3266 0.3287 19.0
1.0228 181.0 1267 2.0553 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 182.0 1274 2.0557 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 183.0 1281 2.0565 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 184.0 1288 2.0565 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 185.0 1295 2.0569 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 186.0 1302 2.0585 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 187.0 1309 2.0590 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 188.0 1316 2.0597 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 189.0 1323 2.0610 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 190.0 1330 2.0615 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 191.0 1337 2.0621 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 192.0 1344 2.0625 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 193.0 1351 2.0628 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 194.0 1358 2.0632 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 195.0 1365 2.0634 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 196.0 1372 2.0634 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 197.0 1379 2.0633 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 198.0 1386 2.0634 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 199.0 1393 2.0635 0.3738 0.1028 0.2953 0.2962 19.0
1.0228 200.0 1400 2.0636 0.3738 0.1028 0.2953 0.2962 19.0

Framework versions

  • Transformers 4.39.0
  • Pytorch 2.2.1+cu121
  • Datasets 2.18.0
  • Tokenizers 0.15.2