GuysTrans's picture
update model card README.md
4e21891
|
raw
history blame
11.8 kB
metadata
license: apache-2.0
base_model: facebook/bart-base
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: bart-base-finetuned-xsum
    results: []

bart-base-finetuned-xsum

This model is a fine-tuned version of facebook/bart-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0568
  • Rouge1: 30.2563
  • Rouge2: 28.8168
  • Rougel: 30.2467
  • Rougelsum: 30.2569
  • Gen Len: 19.9298

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: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 100

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum Gen Len
No log 1.0 22 3.5383 9.5779 1.3552 7.1347 8.7496 19.5556
No log 2.0 44 3.2542 10.1333 1.5586 7.707 9.2027 19.4561
No log 3.0 66 3.0487 11.5328 2.0714 8.6351 10.4996 19.462
No log 4.0 88 2.8741 11.3918 2.2439 8.7008 10.3865 19.1754
No log 5.0 110 2.7390 11.6082 2.4409 8.9366 10.6473 19.0351
No log 6.0 132 2.6105 12.3196 2.8152 9.3727 11.2198 19.3099
No log 7.0 154 2.4855 12.9853 3.0374 9.8049 11.5607 19.5322
No log 8.0 176 2.3759 14.0849 3.4421 10.7915 12.6688 19.7836
No log 9.0 198 2.2703 14.21 3.62 10.7563 12.8026 19.8129
No log 10.0 220 2.1689 13.587 3.08 10.2941 12.2536 19.7836
No log 11.0 242 2.0763 13.8846 3.253 10.6637 12.4641 19.8596
No log 12.0 264 1.9905 14.2798 4.3509 11.2371 13.2956 19.7836
No log 13.0 286 1.8956 14.6906 4.0982 11.3513 13.2757 19.8421
No log 14.0 308 1.8114 14.5806 4.0165 11.2885 13.2862 19.8596
No log 15.0 330 1.7337 14.3077 3.783 10.9654 13.1235 19.8538
No log 16.0 352 1.6598 15.269 4.8164 11.7211 13.8863 19.8596
No log 17.0 374 1.5849 15.1094 4.3627 11.5927 13.6904 19.8363
No log 18.0 396 1.5092 15.6749 5.2096 12.4685 14.5261 19.8596
No log 19.0 418 1.4288 15.0926 4.8751 11.7242 13.9716 19.8772
No log 20.0 440 1.3674 16.0192 5.5445 12.62 14.7367 19.807
No log 21.0 462 1.2949 16.0017 5.6049 12.6056 14.804 19.9123
No log 22.0 484 1.2363 16.1471 5.6893 12.6893 14.8879 19.8421
2.7001 23.0 506 1.1718 15.9206 5.9164 12.5858 14.7608 19.8187
2.7001 24.0 528 1.1139 17.3265 7.7984 14.3624 16.1346 19.8889
2.7001 25.0 550 1.0572 17.4553 7.8986 14.5663 16.2582 19.8246
2.7001 26.0 572 1.0005 17.8156 7.6528 14.6735 16.5001 19.9006
2.7001 27.0 594 0.9436 18.4062 8.656 15.1709 17.1584 19.8947
2.7001 28.0 616 0.9016 18.6358 9.2967 15.7687 17.5333 19.9532
2.7001 29.0 638 0.8521 19.509 10.2714 16.7276 18.3507 19.9123
2.7001 30.0 660 0.8054 19.6085 10.6017 16.7532 18.5697 19.9123
2.7001 31.0 682 0.7572 19.9368 10.9483 17.2141 18.919 19.9123
2.7001 32.0 704 0.7174 20.6909 12.3541 18.3001 19.7958 19.9532
2.7001 33.0 726 0.6796 22.126 14.5214 19.8644 21.3789 19.8772
2.7001 34.0 748 0.6393 22.1752 14.8152 20.0276 21.3231 19.8596
2.7001 35.0 770 0.6060 22.2251 15.2014 20.2721 21.3932 19.9123
2.7001 36.0 792 0.5731 22.6914 15.7427 20.5852 21.9249 19.8772
2.7001 37.0 814 0.5386 23.93 17.4114 22.021 23.1866 19.9181
2.7001 38.0 836 0.5138 23.7373 17.32 21.9715 23.0144 19.9123
2.7001 39.0 858 0.4774 24.2258 18.2839 22.6412 23.6068 19.9181
2.7001 40.0 880 0.4573 24.997 19.2868 23.4309 24.389 19.9123
2.7001 41.0 902 0.4280 24.7499 19.2674 23.382 24.1711 19.8713
2.7001 42.0 924 0.4049 25.7943 21.0893 24.7284 25.361 19.9123
2.7001 43.0 946 0.3822 26.3463 21.9698 25.4544 25.9681 19.8655
2.7001 44.0 968 0.3601 26.7457 22.6408 25.9031 26.4579 19.9123
2.7001 45.0 990 0.3420 26.7588 22.6918 25.9063 26.5082 19.9123
1.2604 46.0 1012 0.3243 27.4421 23.8278 26.8448 27.2093 19.9123
1.2604 47.0 1034 0.3083 27.9994 24.5363 27.3867 27.7983 19.9123
1.2604 48.0 1056 0.2874 28.4342 25.4142 28.0441 28.3455 19.9123
1.2604 49.0 1078 0.2726 28.4356 25.4477 28.0358 28.384 19.9123
1.2604 50.0 1100 0.2592 28.3333 25.2923 27.9377 28.195 19.9123
1.2604 51.0 1122 0.2436 29.2428 26.6115 28.8516 29.1073 19.9298
1.2604 52.0 1144 0.2304 28.9631 26.2987 28.61 28.8463 19.9006
1.2604 53.0 1166 0.2207 29.2598 26.7471 29.0594 29.2673 19.9474
1.2604 54.0 1188 0.2123 29.4739 27.3109 29.4082 29.5044 19.9298
1.2604 55.0 1210 0.1975 28.9596 26.6549 28.8391 28.9865 19.9006
1.2604 56.0 1232 0.1899 29.6305 27.5357 29.4856 29.5378 19.9006
1.2604 57.0 1254 0.1785 29.891 27.7983 29.6302 29.7995 19.9298
1.2604 58.0 1276 0.1697 30.072 28.1443 29.8833 30.002 19.9298
1.2604 59.0 1298 0.1631 29.9935 28.0824 29.813 29.9039 19.9298
1.2604 60.0 1320 0.1561 29.8404 27.955 29.7251 29.7914 19.9006
1.2604 61.0 1342 0.1494 29.9133 28.1012 29.7574 29.8523 19.9006
1.2604 62.0 1364 0.1411 30.013 28.1783 29.8939 30.0162 19.9298
1.2604 63.0 1386 0.1363 29.7979 28.0003 29.666 29.8143 19.9006
1.2604 64.0 1408 0.1302 30.07 28.3675 30.015 30.0808 19.9298
1.2604 65.0 1430 0.1250 30.1907 28.5083 30.1438 30.1837 19.9298
1.2604 66.0 1452 0.1189 29.8624 28.1503 29.7559 29.8928 19.9006
1.2604 67.0 1474 0.1130 30.187 28.5748 30.135 30.1947 19.9474
1.2604 68.0 1496 0.1086 30.1976 28.6341 30.1643 30.2108 19.9298
0.6549 69.0 1518 0.1072 30.2537 28.7763 30.2247 30.2813 19.9474
0.6549 70.0 1540 0.1053 30.1574 28.4808 30.0796 30.1488 19.924
0.6549 71.0 1562 0.1002 30.2294 28.749 30.2237 30.2444 19.9474
0.6549 72.0 1584 0.0958 30.2275 28.6767 30.2092 30.2327 19.9298
0.6549 73.0 1606 0.0926 30.204 28.7073 30.1788 30.2204 19.9298
0.6549 74.0 1628 0.0898 30.2236 28.7646 30.2168 30.2423 19.9298
0.6549 75.0 1650 0.0848 30.2375 28.8277 30.2291 30.2674 19.9474
0.6549 76.0 1672 0.0838 30.2383 28.833 30.2295 30.2596 19.9474
0.6549 77.0 1694 0.0814 30.2612 28.8227 30.253 30.2757 19.9474
0.6549 78.0 1716 0.0789 30.2242 28.7884 30.2173 30.2367 19.924
0.6549 79.0 1738 0.0778 30.2501 28.8825 30.2431 30.2649 19.9474
0.6549 80.0 1760 0.0746 30.2242 28.8027 30.2173 30.2367 19.924
0.6549 81.0 1782 0.0738 30.2184 28.7956 30.2107 30.2268 19.9298
0.6549 82.0 1804 0.0712 30.2184 28.8063 30.2107 30.2268 19.924
0.6549 83.0 1826 0.0695 30.2184 28.8063 30.2107 30.2268 19.924
0.6549 84.0 1848 0.0683 30.2184 28.8063 30.2107 30.2268 19.924
0.6549 85.0 1870 0.0664 30.1396 28.7072 30.1389 30.1451 19.9064
0.6549 86.0 1892 0.0657 30.1239 28.688 30.1254 30.1301 19.9064
0.6549 87.0 1914 0.0641 30.2184 28.792 30.2107 30.2268 19.924
0.6549 88.0 1936 0.0630 30.2357 28.8062 30.2215 30.2395 19.9298
0.6549 89.0 1958 0.0618 30.2184 28.7851 30.2107 30.2268 19.924
0.6549 90.0 1980 0.0606 30.2184 28.7898 30.2107 30.2268 19.9298
0.4294 91.0 2002 0.0601 30.2184 28.792 30.2107 30.2268 19.9298
0.4294 92.0 2024 0.0594 30.2184 28.7851 30.2107 30.2268 19.9298
0.4294 93.0 2046 0.0591 30.2184 28.7851 30.2107 30.2268 19.9298
0.4294 94.0 2068 0.0583 30.2184 28.792 30.2107 30.2268 19.9298
0.4294 95.0 2090 0.0578 30.2357 28.8062 30.2215 30.2395 19.9298
0.4294 96.0 2112 0.0574 30.2357 28.8296 30.2215 30.2395 19.9298
0.4294 97.0 2134 0.0572 30.2563 28.8382 30.2467 30.2569 19.9298
0.4294 98.0 2156 0.0572 30.2563 28.8382 30.2467 30.2569 19.9298
0.4294 99.0 2178 0.0569 30.2563 28.8382 30.2467 30.2569 19.9298
0.4294 100.0 2200 0.0568 30.2563 28.8168 30.2467 30.2569 19.9298

Framework versions

  • Transformers 4.31.0
  • Pytorch 2.0.1+cu118
  • Datasets 2.14.0
  • Tokenizers 0.13.3