scenario-KD-PO-MSV-EN-EN-D2_data-en-massive_all_1_166
This model is a fine-tuned version of haryoaw/scenario-MDBT-TCR_data-en-massive_all_1_1 on the massive dataset. It achieves the following results on the evaluation set:
- Loss: 18.6035
- Accuracy: 0.2922
- F1: 0.2796
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: 32
- eval_batch_size: 32
- seed: 66
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 30
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
---|---|---|---|---|---|
No log | 0.28 | 100 | 16.0585 | 0.0733 | 0.0068 |
No log | 0.56 | 200 | 14.9516 | 0.1179 | 0.0282 |
No log | 0.83 | 300 | 15.2519 | 0.1360 | 0.0631 |
No log | 1.11 | 400 | 14.4169 | 0.1781 | 0.0989 |
11.7852 | 1.39 | 500 | 13.7629 | 0.2084 | 0.1253 |
11.7852 | 1.67 | 600 | 14.0237 | 0.2160 | 0.1496 |
11.7852 | 1.94 | 700 | 13.4321 | 0.2267 | 0.1644 |
11.7852 | 2.22 | 800 | 13.4657 | 0.2486 | 0.1846 |
11.7852 | 2.5 | 900 | 13.5450 | 0.2570 | 0.1981 |
5.8982 | 2.78 | 1000 | 13.3906 | 0.2568 | 0.2051 |
5.8982 | 3.06 | 1100 | 13.6663 | 0.2642 | 0.2024 |
5.8982 | 3.33 | 1200 | 14.2607 | 0.2513 | 0.1952 |
5.8982 | 3.61 | 1300 | 14.3919 | 0.2542 | 0.2029 |
5.8982 | 3.89 | 1400 | 14.5446 | 0.2510 | 0.2007 |
4.048 | 4.17 | 1500 | 14.1289 | 0.2602 | 0.2153 |
4.048 | 4.44 | 1600 | 13.3403 | 0.2956 | 0.2267 |
4.048 | 4.72 | 1700 | 14.8539 | 0.2622 | 0.2131 |
4.048 | 5.0 | 1800 | 13.5653 | 0.2939 | 0.2389 |
4.048 | 5.28 | 1900 | 14.6566 | 0.2792 | 0.2383 |
3.0561 | 5.56 | 2000 | 14.7909 | 0.2676 | 0.2348 |
3.0561 | 5.83 | 2100 | 13.9744 | 0.3007 | 0.2483 |
3.0561 | 6.11 | 2200 | 15.5685 | 0.2777 | 0.2348 |
3.0561 | 6.39 | 2300 | 15.5535 | 0.2727 | 0.2297 |
3.0561 | 6.67 | 2400 | 15.1022 | 0.2858 | 0.2416 |
2.3857 | 6.94 | 2500 | 15.8558 | 0.2734 | 0.2361 |
2.3857 | 7.22 | 2600 | 16.0300 | 0.2645 | 0.2272 |
2.3857 | 7.5 | 2700 | 16.0818 | 0.2734 | 0.2460 |
2.3857 | 7.78 | 2800 | 16.2737 | 0.2782 | 0.2445 |
2.3857 | 8.06 | 2900 | 16.3877 | 0.2587 | 0.2262 |
1.9693 | 8.33 | 3000 | 17.3819 | 0.2583 | 0.2331 |
1.9693 | 8.61 | 3100 | 17.1412 | 0.2636 | 0.2366 |
1.9693 | 8.89 | 3200 | 17.3173 | 0.2611 | 0.2337 |
1.9693 | 9.17 | 3300 | 16.1257 | 0.2785 | 0.2468 |
1.9693 | 9.44 | 3400 | 17.4479 | 0.2671 | 0.2453 |
1.6624 | 9.72 | 3500 | 15.9842 | 0.2959 | 0.2595 |
1.6624 | 10.0 | 3600 | 16.6481 | 0.2764 | 0.2454 |
1.6624 | 10.28 | 3700 | 16.0613 | 0.2952 | 0.2496 |
1.6624 | 10.56 | 3800 | 17.3130 | 0.2796 | 0.2483 |
1.6624 | 10.83 | 3900 | 17.9793 | 0.2768 | 0.2415 |
1.4248 | 11.11 | 4000 | 17.9004 | 0.2768 | 0.2508 |
1.4248 | 11.39 | 4100 | 17.6532 | 0.2776 | 0.2549 |
1.4248 | 11.67 | 4200 | 17.9802 | 0.2763 | 0.2512 |
1.4248 | 11.94 | 4300 | 19.2692 | 0.2543 | 0.2468 |
1.4248 | 12.22 | 4400 | 18.8586 | 0.2693 | 0.2551 |
1.2048 | 12.5 | 4500 | 18.2546 | 0.2746 | 0.2508 |
1.2048 | 12.78 | 4600 | 18.1165 | 0.2729 | 0.2526 |
1.2048 | 13.06 | 4700 | 18.8671 | 0.2615 | 0.2417 |
1.2048 | 13.33 | 4800 | 18.8131 | 0.2630 | 0.2466 |
1.2048 | 13.61 | 4900 | 18.3799 | 0.2771 | 0.2568 |
1.0702 | 13.89 | 5000 | 18.5563 | 0.2691 | 0.2416 |
1.0702 | 14.17 | 5100 | 19.0471 | 0.2621 | 0.2449 |
1.0702 | 14.44 | 5200 | 18.6233 | 0.2721 | 0.2451 |
1.0702 | 14.72 | 5300 | 18.8386 | 0.2776 | 0.2590 |
1.0702 | 15.0 | 5400 | 19.5330 | 0.2655 | 0.2479 |
0.9462 | 15.28 | 5500 | 19.8716 | 0.2607 | 0.2499 |
0.9462 | 15.56 | 5600 | 18.5496 | 0.2770 | 0.2568 |
0.9462 | 15.83 | 5700 | 17.8301 | 0.2950 | 0.2697 |
0.9462 | 16.11 | 5800 | 17.7789 | 0.2951 | 0.2724 |
0.9462 | 16.39 | 5900 | 19.3109 | 0.2768 | 0.2673 |
0.8576 | 16.67 | 6000 | 18.1516 | 0.2926 | 0.2660 |
0.8576 | 16.94 | 6100 | 19.5121 | 0.2744 | 0.2574 |
0.8576 | 17.22 | 6200 | 19.5653 | 0.2763 | 0.2625 |
0.8576 | 17.5 | 6300 | 18.3909 | 0.2851 | 0.2640 |
0.8576 | 17.78 | 6400 | 19.0072 | 0.2741 | 0.2467 |
0.7527 | 18.06 | 6500 | 18.6327 | 0.2833 | 0.2648 |
0.7527 | 18.33 | 6600 | 18.9928 | 0.2803 | 0.2563 |
0.7527 | 18.61 | 6700 | 19.7251 | 0.2744 | 0.2603 |
0.7527 | 18.89 | 6800 | 19.1755 | 0.2745 | 0.2576 |
0.7527 | 19.17 | 6900 | 18.4740 | 0.2883 | 0.2686 |
0.7082 | 19.44 | 7000 | 18.9633 | 0.2866 | 0.2668 |
0.7082 | 19.72 | 7100 | 19.5535 | 0.2751 | 0.2648 |
0.7082 | 20.0 | 7200 | 19.1204 | 0.2826 | 0.2612 |
0.7082 | 20.28 | 7300 | 19.4658 | 0.2786 | 0.2618 |
0.7082 | 20.56 | 7400 | 18.3475 | 0.2930 | 0.2716 |
0.6603 | 20.83 | 7500 | 19.6894 | 0.2720 | 0.2568 |
0.6603 | 21.11 | 7600 | 18.3256 | 0.2929 | 0.2727 |
0.6603 | 21.39 | 7700 | 19.0269 | 0.2809 | 0.2712 |
0.6603 | 21.67 | 7800 | 18.9538 | 0.2834 | 0.2658 |
0.6603 | 21.94 | 7900 | 18.8878 | 0.2904 | 0.2742 |
0.6171 | 22.22 | 8000 | 18.9117 | 0.2887 | 0.2728 |
0.6171 | 22.5 | 8100 | 19.0627 | 0.2853 | 0.2737 |
0.6171 | 22.78 | 8200 | 19.1497 | 0.2822 | 0.2722 |
0.6171 | 23.06 | 8300 | 19.3517 | 0.2764 | 0.2673 |
0.6171 | 23.33 | 8400 | 18.9524 | 0.2836 | 0.2706 |
0.5721 | 23.61 | 8500 | 18.4516 | 0.2907 | 0.2749 |
0.5721 | 23.89 | 8600 | 18.7686 | 0.2881 | 0.2749 |
0.5721 | 24.17 | 8700 | 19.0653 | 0.2847 | 0.2697 |
0.5721 | 24.44 | 8800 | 20.1017 | 0.2721 | 0.2664 |
0.5721 | 24.72 | 8900 | 18.7587 | 0.2910 | 0.2743 |
0.5466 | 25.0 | 9000 | 19.4485 | 0.2827 | 0.2741 |
0.5466 | 25.28 | 9100 | 19.2920 | 0.2831 | 0.2677 |
0.5466 | 25.56 | 9200 | 19.1334 | 0.2872 | 0.2735 |
0.5466 | 25.83 | 9300 | 18.9784 | 0.2859 | 0.2694 |
0.5466 | 26.11 | 9400 | 18.7701 | 0.2914 | 0.2763 |
0.5168 | 26.39 | 9500 | 19.3216 | 0.2767 | 0.2665 |
0.5168 | 26.67 | 9600 | 19.3074 | 0.2800 | 0.2745 |
0.5168 | 26.94 | 9700 | 18.6569 | 0.2889 | 0.2722 |
0.5168 | 27.22 | 9800 | 19.3113 | 0.2800 | 0.2703 |
0.5168 | 27.5 | 9900 | 18.8369 | 0.2900 | 0.2774 |
0.5197 | 27.78 | 10000 | 18.7418 | 0.2894 | 0.2771 |
0.5197 | 28.06 | 10100 | 18.8462 | 0.2885 | 0.2754 |
0.5197 | 28.33 | 10200 | 18.6737 | 0.2913 | 0.2785 |
0.5197 | 28.61 | 10300 | 18.8000 | 0.2880 | 0.2755 |
0.5197 | 28.89 | 10400 | 18.5512 | 0.2936 | 0.2793 |
0.5027 | 29.17 | 10500 | 18.5273 | 0.2943 | 0.2809 |
0.5027 | 29.44 | 10600 | 18.5875 | 0.2920 | 0.2797 |
0.5027 | 29.72 | 10700 | 18.6780 | 0.2916 | 0.2807 |
0.5027 | 30.0 | 10800 | 18.6035 | 0.2922 | 0.2796 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.1.1+cu121
- Datasets 2.14.5
- Tokenizers 0.13.3
- Downloads last month
- 1
Model tree for haryoaw/scenario-KD-PO-MSV-EN-EN-D2_data-en-massive_all_1_166
Base model
microsoft/mdeberta-v3-base
Finetuned
haryoaw/scenario-MDBT-TCR-MSV-EN