arabert_cross_relevance_task4_fold4

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: 0.2574
  • Qwk: 0.3506
  • Mse: 0.2574

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 0.1111 2 0.5156 0.0454 0.5156
No log 0.2222 4 0.4059 0.0584 0.4059
No log 0.3333 6 0.3883 0.0664 0.3883
No log 0.4444 8 0.3323 0.0300 0.3323
No log 0.5556 10 0.3166 0.0769 0.3166
No log 0.6667 12 0.3026 0.0769 0.3026
No log 0.7778 14 0.2709 0.1111 0.2709
No log 0.8889 16 0.2522 0.2158 0.2522
No log 1.0 18 0.2404 0.3048 0.2404
No log 1.1111 20 0.2348 0.2941 0.2348
No log 1.2222 22 0.2318 0.3163 0.2318
No log 1.3333 24 0.2322 0.3054 0.2322
No log 1.4444 26 0.2311 0.2906 0.2311
No log 1.5556 28 0.2619 0.2871 0.2619
No log 1.6667 30 0.2783 0.2455 0.2783
No log 1.7778 32 0.2675 0.3226 0.2675
No log 1.8889 34 0.2424 0.2797 0.2424
No log 2.0 36 0.2390 0.2795 0.2390
No log 2.1111 38 0.2386 0.2863 0.2386
No log 2.2222 40 0.2432 0.3256 0.2432
No log 2.3333 42 0.2747 0.4663 0.2747
No log 2.4444 44 0.2768 0.4585 0.2768
No log 2.5556 46 0.2536 0.3640 0.2536
No log 2.6667 48 0.2422 0.2688 0.2422
No log 2.7778 50 0.2397 0.2728 0.2397
No log 2.8889 52 0.2372 0.2688 0.2372
No log 3.0 54 0.2444 0.3025 0.2444
No log 3.1111 56 0.2512 0.3711 0.2512
No log 3.2222 58 0.2356 0.3252 0.2356
No log 3.3333 60 0.2395 0.2652 0.2395
No log 3.4444 62 0.2364 0.3424 0.2364
No log 3.5556 64 0.2393 0.3891 0.2393
No log 3.6667 66 0.2374 0.3877 0.2374
No log 3.7778 68 0.2344 0.3590 0.2344
No log 3.8889 70 0.2383 0.3509 0.2383
No log 4.0 72 0.2414 0.3561 0.2414
No log 4.1111 74 0.2562 0.3780 0.2562
No log 4.2222 76 0.2480 0.3775 0.2480
No log 4.3333 78 0.2372 0.3346 0.2372
No log 4.4444 80 0.2374 0.3862 0.2374
No log 4.5556 82 0.2400 0.4009 0.2400
No log 4.6667 84 0.2397 0.3761 0.2397
No log 4.7778 86 0.2421 0.3780 0.2421
No log 4.8889 88 0.2422 0.4085 0.2422
No log 5.0 90 0.2475 0.3510 0.2475
No log 5.1111 92 0.2516 0.3213 0.2516
No log 5.2222 94 0.2411 0.3195 0.2411
No log 5.3333 96 0.2411 0.3936 0.2411
No log 5.4444 98 0.2487 0.3948 0.2487
No log 5.5556 100 0.2404 0.3513 0.2404
No log 5.6667 102 0.2386 0.2965 0.2386
No log 5.7778 104 0.2498 0.2913 0.2498
No log 5.8889 106 0.2471 0.3274 0.2471
No log 6.0 108 0.2451 0.4019 0.2451
No log 6.1111 110 0.2598 0.5037 0.2598
No log 6.2222 112 0.2620 0.4876 0.2620
No log 6.3333 114 0.2507 0.4433 0.2507
No log 6.4444 116 0.2412 0.3147 0.2412
No log 6.5556 118 0.2475 0.2991 0.2475
No log 6.6667 120 0.2504 0.3106 0.2504
No log 6.7778 122 0.2428 0.3103 0.2428
No log 6.8889 124 0.2451 0.3410 0.2451
No log 7.0 126 0.2482 0.3509 0.2482
No log 7.1111 128 0.2493 0.3984 0.2493
No log 7.2222 130 0.2509 0.3757 0.2509
No log 7.3333 132 0.2537 0.3713 0.2537
No log 7.4444 134 0.2508 0.3694 0.2508
No log 7.5556 136 0.2502 0.3626 0.2502
No log 7.6667 138 0.2486 0.3673 0.2486
No log 7.7778 140 0.2488 0.3357 0.2488
No log 7.8889 142 0.2538 0.3178 0.2538
No log 8.0 144 0.2653 0.3018 0.2653
No log 8.1111 146 0.2720 0.3029 0.2720
No log 8.2222 148 0.2665 0.3258 0.2665
No log 8.3333 150 0.2598 0.3725 0.2598
No log 8.4444 152 0.2597 0.3889 0.2597
No log 8.5556 154 0.2611 0.4296 0.2611
No log 8.6667 156 0.2627 0.3969 0.2627
No log 8.7778 158 0.2627 0.3787 0.2627
No log 8.8889 160 0.2639 0.3679 0.2639
No log 9.0 162 0.2673 0.3381 0.2673
No log 9.1111 164 0.2677 0.3197 0.2677
No log 9.2222 166 0.2642 0.3383 0.2642
No log 9.3333 168 0.2606 0.3291 0.2606
No log 9.4444 170 0.2580 0.3307 0.2580
No log 9.5556 172 0.2570 0.3373 0.2570
No log 9.6667 174 0.2566 0.3440 0.2566
No log 9.7778 176 0.2569 0.3506 0.2569
No log 9.8889 178 0.2572 0.3506 0.2572
No log 10.0 180 0.2574 0.3506 0.2574

Framework versions

  • Transformers 4.44.0
  • Pytorch 2.4.0
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
3
Safetensors
Model size
135M params
Tensor type
F32
·
Inference API
Unable to determine this model's library. Check the docs .

Model tree for salbatarni/arabert_cross_relevance_task4_fold4

Finetuned
(3647)
this model