Hugging Face's logo

silmi224
/
finetune-led-thousanddata

Summarization
Transformers
TensorBoard
Safetensors
led
text2text-generation
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
finetune-led-thousanddata / README.md
silmi224's picture
silmi224
Training complete
6df9ce9 verified
preview code
|
raw
history blame contribute delete
No virus
14.3 kB
metadata
tags:
  - summarization
  - generated_from_trainer
model-index:
  - name: finetune-led-thousanddata
    results: []

finetune-led-thousanddata

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.9539
  • Rouge1 Precision: 0.2722
  • Rouge1 Recall: 0.3458
  • Rouge1 Fmeasure: 0.3011

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: 5e-05
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 4
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Rouge1 Fmeasure Rouge1 Precision Rouge1 Recall
2.0529 0.13 10 2.6191 0.3014 0.2948 0.324
1.778 0.26 20 2.4690 0.2947 0.2802 0.3213
1.7425 0.38 30 2.3989 0.3037 0.2734 0.3524
1.7006 0.51 40 2.3216 0.2941 0.2665 0.3386
1.6751 0.64 50 2.3027 0.3101 0.282 0.3551
1.6887 0.77 60 2.2911 0.3058 0.2731 0.3577
1.6008 0.89 70 2.2476 0.3016 0.272 0.3487
1.5767 1.02 80 2.2167 0.3043 0.2775 0.3465
1.5046 1.15 90 2.2185 0.3004 0.2721 0.3458
1.5394 1.28 100 2.1977 0.2991 0.2696 0.3463
1.5449 1.41 110 2.1823 0.2978 0.2704 0.341
1.5073 1.53 120 2.1832 0.3057 0.276 0.3527
1.5232 0.42 130 2.2091 0.2955 0.2664 0.3424
1.4896 0.45 140 2.2069 0.2905 0.2574 0.3424
1.4848 0.48 150 2.1913 0.2868 0.2567 0.3356
1.5084 0.51 160 2.1826 0.3006 0.2755 0.3406
1.4322 0.54 170 2.2525 0.3049 0.2716 0.3582
1.4672 0.58 180 2.1890 0.2919 0.2663 0.3322
1.4543 0.61 190 2.1487 0.3022 0.276 0.344
1.5446 0.64 200 2.1496 0.2993 0.273 0.3418
1.412 0.67 210 2.1837 0.2976 0.268 0.3439
1.5241 0.7 220 2.1423 0.2913 0.2665 0.3305
1.4806 0.74 230 2.1303 0.2997 0.2736 0.3411
1.5405 0.77 240 2.1205 0.2966 0.2668 0.3428
1.4287 0.8 250 2.1322 0.2976 0.268 0.3442
1.4977 0.83 260 2.1334 0.2979 0.2665 0.3477
1.4171 0.86 270 2.1184 0.3043 0.2741 0.3509
1.4491 0.9 280 2.1038 0.2868 0.2628 0.3253
1.4316 0.93 290 2.1254 0.2958 0.2678 0.3393
1.4689 0.96 300 2.1052 0.299 0.2685 0.3471
1.4347 0.99 310 2.0815 0.3019 0.273 0.3476
1.3285 1.02 320 2.0877 0.2981 0.2695 0.3427
1.2636 1.06 330 2.0740 0.2933 0.2645 0.3382
1.32 1.09 340 2.0755 0.2997 0.2689 0.3487
1.357 1.12 350 2.0594 0.301 0.2743 0.3434
1.3412 1.15 360 2.0660 0.2961 0.2677 0.3405
1.327 1.18 370 2.0649 0.2912 0.263 0.335
1.3193 1.22 380 2.0842 0.2952 0.2673 0.3392
1.2961 1.25 390 2.0749 0.2957 0.2705 0.3342
1.3093 1.28 400 2.0715 0.2997 0.272 0.3441
1.3403 1.31 410 2.0671 0.3119 0.2823 0.3584
1.3685 1.34 420 2.0580 0.2973 0.2695 0.3409
1.2913 1.38 430 2.0685 0.2926 0.2632 0.339
1.3796 1.41 440 2.0339 0.2962 0.2697 0.3387
1.354 1.44 450 2.0371 0.2953 0.2665 0.3412
1.3268 1.47 460 2.0309 0.2957 0.2681 0.3395
1.3706 1.5 470 2.0215 0.2932 0.2685 0.3315
1.3288 1.54 480 2.0044 0.2948 0.2674 0.3374
1.4102 1.57 490 2.0046 0.2998 0.271 0.3446
1.3952 1.6 500 2.0044 0.3063 0.2794 0.3487
1.2994 1.63 510 1.9993 0.3052 0.2787 0.3461
1.2948 1.66 520 2.0168 0.3 0.2743 0.3406
1.2972 1.7 530 2.0290 0.3003 0.2734 0.342
1.3181 1.73 540 2.0234 0.2949 0.2676 0.338
1.3505 1.76 550 1.9942 0.301 0.2737 0.3436
1.3163 1.79 560 1.9983 0.2963 0.2705 0.3366
1.2876 1.82 570 2.0206 0.303 0.2739 0.3486
1.2895 1.86 580 2.0131 0.2958 0.2652 0.3443
1.3257 1.89 590 1.9888 0.3022 0.2743 0.3455
1.2891 1.92 600 1.9928 0.2972 0.2694 0.3408
1.3152 1.95 610 1.9785 0.292 0.2653 0.334
1.2834 1.98 620 2.0105 0.3039 0.2735 0.3511
1.2373 2.02 630 2.0023 0.3019 0.2735 0.346
1.2569 2.05 640 2.0006 0.3029 0.2753 0.3463
1.2337 2.08 650 1.9919 0.3006 0.2746 0.3416
1.1274 2.11 660 2.0095 0.3015 0.2732 0.3457
1.2178 2.14 670 1.9974 0.3031 0.275 0.3475
1.22 2.18 680 1.9924 0.3059 0.2777 0.3501
1.2913 2.21 690 1.9880 0.3044 0.2745 0.351
1.2441 2.24 700 1.9886 0.299 0.2721 0.3412
1.3258 2.27 710 1.9772 0.2956 0.2686 0.3377
1.158 2.3 720 2.0003 0.2983 0.2702 0.3424
1.1908 2.34 730 1.9845 0.2975 0.2705 0.3398
1.2411 2.37 740 1.9768 0.304 0.275 0.3493
1.1936 2.4 750 2.0065 0.293 0.2628 0.3403
1.1578 2.44 760 2.0199 0.301 0.2713 0.3473
1.2086 2.47 770 1.9949 0.2921 0.2664 0.3323
1.2574 2.5 780 1.9806 0.297 0.2693 0.3405
1.2331 2.53 790 2.0100 0.3012 0.2733 0.3446
1.2522 2.56 800 1.9969 0.301 0.2716 0.3468
1.2508 2.6 810 1.9931 0.3016 0.2719 0.3471
1.1558 2.63 820 1.9873 0.2986 0.2725 0.3402
1.2721 2.66 830 1.9763 0.2988 0.2671 0.348
1.2817 2.69 840 1.9713 0.2961 0.2688 0.3388
1.2183 2.72 850 1.9783 0.2985 0.2709 0.3416
1.2278 2.76 860 1.9757 0.2964 0.2681 0.3402
1.2087 2.79 870 1.9818 0.304 0.2735 0.3516
1.1838 2.82 880 1.9845 0.2916 0.2659 0.3312
1.1185 2.85 890 1.9912 0.3044 0.2759 0.3492
1.1214 2.88 900 1.9838 0.2995 0.2692 0.3468
1.2341 2.92 910 1.9685 0.296 0.2713 0.3344
1.1808 2.95 920 1.9803 0.3008 0.2725 0.345
1.2843 2.98 930 1.9645 0.3041 0.2745 0.3504
1.1824 3.01 940 1.9750 0.2985 0.2713 0.3412
1.1399 3.04 950 1.9762 0.2943 0.264 0.3416
1.1347 3.08 960 1.9841 0.2971 0.2685 0.3419
1.2298 3.11 970 1.9526 0.2993 0.2701 0.3448
1.1731 3.14 980 1.9787 0.304 0.2726 0.3531
1.1819 3.17 990 1.9570 0.2995 0.2715 0.3437
1.2072 3.2 1000 1.9613 0.3004 0.2705 0.3472
1.1214 3.24 1010 1.9670 0.3 0.2723 0.3432
1.226 3.27 1020 1.9676 0.2945 0.2639 0.3422
1.1956 3.3 1030 1.9721 0.2949 0.2657 0.3406
1.2286 3.33 1040 1.9572 0.3046 0.2759 0.3489
1.1786 3.36 1050 1.9549 0.3009 0.2728 0.3448
1.1512 3.4 1060 1.9609 0.2989 0.2699 0.3441
1.1897 3.43 1070 1.9626 0.2983 0.2697 0.3427
1.187 3.46 1080 1.9612 0.3016 0.2731 0.3457
1.1394 3.49 1090 1.9519 0.3015 0.2746 0.3431
1.1088 3.52 1100 1.9674 0.301 0.2709 0.3477
1.1787 3.56 1110 1.9549 0.3009 0.2728 0.3449
1.1961 3.59 1120 1.9545 0.3016 0.2722 0.3476
1.1194 3.62 1130 1.9693 0.3028 0.2735 0.3484
1.1991 3.65 1140 1.9538 0.3002 0.2706 0.3461
1.2109 3.68 1150 1.9428 0.3018 0.2729 0.3465
1.1389 3.72 1160 1.9578 0.3008 0.2723 0.3452
1.1922 3.75 1170 1.9576 0.2992 0.2701 0.3446
1.1002 3.78 1180 1.9571 0.299 0.2696 0.3445
1.1407 3.81 1190 1.9530 0.2979 0.2692 0.3422
1.1882 3.84 1200 1.9491 0.3009 0.2725 0.345
1.1755 3.88 1210 1.9562 0.3024 0.2735 0.3468
1.062 3.91 1220 1.9577 0.302 0.2722 0.3478
1.1965 3.94 1230 1.9575 0.3013 0.2716 0.3472
1.1255 3.97 1240 1.9550 0.3014 0.272 0.3466

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.1.1+cu121
  • Datasets 2.14.5
  • Tokenizers 0.15.1