arabert_cross_relevance_task4_fold0
This model is a fine-tuned version of aubmindlab/bert-base-arabertv02 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2736
- Qwk: 0.1560
- Mse: 0.2738
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Qwk | Mse |
---|---|---|---|---|---|
No log | 0.1176 | 2 | 0.5899 | 0.0303 | 0.5899 |
No log | 0.2353 | 4 | 0.2991 | 0.0083 | 0.2993 |
No log | 0.3529 | 6 | 0.2856 | 0.0046 | 0.2859 |
No log | 0.4706 | 8 | 0.3724 | 0.0918 | 0.3727 |
No log | 0.5882 | 10 | 0.3677 | 0.1010 | 0.3680 |
No log | 0.7059 | 12 | 0.2967 | 0.0745 | 0.2971 |
No log | 0.8235 | 14 | 0.2878 | 0.0745 | 0.2883 |
No log | 0.9412 | 16 | 0.3092 | 0.0849 | 0.3099 |
No log | 1.0588 | 18 | 0.3111 | 0.0769 | 0.3118 |
No log | 1.1765 | 20 | 0.3278 | 0.0860 | 0.3286 |
No log | 1.2941 | 22 | 0.3356 | 0.0940 | 0.3363 |
No log | 1.4118 | 24 | 0.3182 | 0.0930 | 0.3188 |
No log | 1.5294 | 26 | 0.2840 | 0.0711 | 0.2844 |
No log | 1.6471 | 28 | 0.2640 | 0.0745 | 0.2644 |
No log | 1.7647 | 30 | 0.2617 | 0.0787 | 0.2620 |
No log | 1.8824 | 32 | 0.2721 | 0.0686 | 0.2724 |
No log | 2.0 | 34 | 0.2784 | 0.0709 | 0.2787 |
No log | 2.1176 | 36 | 0.2850 | 0.1325 | 0.2852 |
No log | 2.2353 | 38 | 0.2912 | 0.2051 | 0.2915 |
No log | 2.3529 | 40 | 0.2922 | 0.1733 | 0.2925 |
No log | 2.4706 | 42 | 0.2846 | 0.1093 | 0.2850 |
No log | 2.5882 | 44 | 0.2920 | 0.1049 | 0.2924 |
No log | 2.7059 | 46 | 0.3062 | 0.1101 | 0.3066 |
No log | 2.8235 | 48 | 0.2956 | 0.1049 | 0.2961 |
No log | 2.9412 | 50 | 0.2664 | 0.1066 | 0.2668 |
No log | 3.0588 | 52 | 0.2583 | 0.1197 | 0.2587 |
No log | 3.1765 | 54 | 0.2638 | 0.1594 | 0.2642 |
No log | 3.2941 | 56 | 0.2611 | 0.1408 | 0.2615 |
No log | 3.4118 | 58 | 0.2742 | 0.1087 | 0.2747 |
No log | 3.5294 | 60 | 0.2867 | 0.0956 | 0.2872 |
No log | 3.6471 | 62 | 0.3015 | 0.0985 | 0.3019 |
No log | 3.7647 | 64 | 0.2991 | 0.1076 | 0.2995 |
No log | 3.8824 | 66 | 0.2717 | 0.1219 | 0.2721 |
No log | 4.0 | 68 | 0.2625 | 0.1787 | 0.2628 |
No log | 4.1176 | 70 | 0.2648 | 0.2139 | 0.2651 |
No log | 4.2353 | 72 | 0.2608 | 0.1889 | 0.2611 |
No log | 4.3529 | 74 | 0.2596 | 0.1230 | 0.2599 |
No log | 4.4706 | 76 | 0.2657 | 0.1072 | 0.2660 |
No log | 4.5882 | 78 | 0.2691 | 0.1064 | 0.2695 |
No log | 4.7059 | 80 | 0.2573 | 0.0987 | 0.2577 |
No log | 4.8235 | 82 | 0.2467 | 0.1026 | 0.2470 |
No log | 4.9412 | 84 | 0.2538 | 0.1374 | 0.2541 |
No log | 5.0588 | 86 | 0.2632 | 0.1541 | 0.2635 |
No log | 5.1765 | 88 | 0.2699 | 0.1482 | 0.2702 |
No log | 5.2941 | 90 | 0.2684 | 0.1464 | 0.2687 |
No log | 5.4118 | 92 | 0.2643 | 0.1446 | 0.2646 |
No log | 5.5294 | 94 | 0.2663 | 0.1381 | 0.2665 |
No log | 5.6471 | 96 | 0.2594 | 0.1117 | 0.2597 |
No log | 5.7647 | 98 | 0.2596 | 0.1043 | 0.2599 |
No log | 5.8824 | 100 | 0.2557 | 0.1095 | 0.2559 |
No log | 6.0 | 102 | 0.2552 | 0.1192 | 0.2555 |
No log | 6.1176 | 104 | 0.2623 | 0.1286 | 0.2626 |
No log | 6.2353 | 106 | 0.2589 | 0.1349 | 0.2592 |
No log | 6.3529 | 108 | 0.2591 | 0.1493 | 0.2593 |
No log | 6.4706 | 110 | 0.2623 | 0.1697 | 0.2626 |
No log | 6.5882 | 112 | 0.2713 | 0.1744 | 0.2716 |
No log | 6.7059 | 114 | 0.2740 | 0.1744 | 0.2743 |
No log | 6.8235 | 116 | 0.2753 | 0.1499 | 0.2756 |
No log | 6.9412 | 118 | 0.2729 | 0.1471 | 0.2732 |
No log | 7.0588 | 120 | 0.2685 | 0.1437 | 0.2688 |
No log | 7.1765 | 122 | 0.2665 | 0.1388 | 0.2667 |
No log | 7.2941 | 124 | 0.2587 | 0.1432 | 0.2589 |
No log | 7.4118 | 126 | 0.2538 | 0.1376 | 0.2541 |
No log | 7.5294 | 128 | 0.2554 | 0.1388 | 0.2557 |
No log | 7.6471 | 130 | 0.2570 | 0.1592 | 0.2572 |
No log | 7.7647 | 132 | 0.2582 | 0.1593 | 0.2584 |
No log | 7.8824 | 134 | 0.2619 | 0.1575 | 0.2621 |
No log | 8.0 | 136 | 0.2676 | 0.1543 | 0.2678 |
No log | 8.1176 | 138 | 0.2711 | 0.1419 | 0.2714 |
No log | 8.2353 | 140 | 0.2745 | 0.1388 | 0.2747 |
No log | 8.3529 | 142 | 0.2729 | 0.1388 | 0.2731 |
No log | 8.4706 | 144 | 0.2716 | 0.1325 | 0.2718 |
No log | 8.5882 | 146 | 0.2731 | 0.1295 | 0.2734 |
No log | 8.7059 | 148 | 0.2726 | 0.1287 | 0.2729 |
No log | 8.8235 | 150 | 0.2748 | 0.1306 | 0.2751 |
No log | 8.9412 | 152 | 0.2760 | 0.1400 | 0.2762 |
No log | 9.0588 | 154 | 0.2751 | 0.1400 | 0.2754 |
No log | 9.1765 | 156 | 0.2728 | 0.1478 | 0.2731 |
No log | 9.2941 | 158 | 0.2721 | 0.1478 | 0.2723 |
No log | 9.4118 | 160 | 0.2716 | 0.1478 | 0.2718 |
No log | 9.5294 | 162 | 0.2714 | 0.1478 | 0.2717 |
No log | 9.6471 | 164 | 0.2722 | 0.1478 | 0.2725 |
No log | 9.7647 | 166 | 0.2731 | 0.1560 | 0.2733 |
No log | 9.8824 | 168 | 0.2734 | 0.1560 | 0.2736 |
No log | 10.0 | 170 | 0.2736 | 0.1560 | 0.2738 |
Framework versions
- Transformers 4.44.0
- Pytorch 2.4.0
- Datasets 2.21.0
- Tokenizers 0.19.1
- Downloads last month
- 0
Model tree for salbatarni/arabert_cross_relevance_task4_fold0
Base model
aubmindlab/bert-base-arabertv02