ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run1_AugV5_k4_task7_organization

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: 0.7441
  • Qwk: 0.3088
  • Mse: 0.7441
  • Rmse: 0.8626

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 Qwk Mse Rmse
No log 0.2 2 2.4280 -0.0646 2.4280 1.5582
No log 0.4 4 1.0881 0.2875 1.0881 1.0431
No log 0.6 6 1.0474 -0.1517 1.0474 1.0234
No log 0.8 8 1.3691 -0.1706 1.3691 1.1701
No log 1.0 10 1.2737 -0.1706 1.2737 1.1286
No log 1.2 12 1.0007 0.0283 1.0007 1.0003
No log 1.4 14 0.9275 0.1183 0.9275 0.9630
No log 1.6 16 0.8221 0.0428 0.8221 0.9067
No log 1.8 18 0.8087 0.0 0.8087 0.8993
No log 2.0 20 0.7870 0.0 0.7870 0.8871
No log 2.2 22 0.7671 0.0 0.7671 0.8758
No log 2.4 24 0.7781 0.0 0.7781 0.8821
No log 2.6 26 0.8770 -0.0320 0.8770 0.9365
No log 2.8 28 1.0144 -0.0076 1.0144 1.0072
No log 3.0 30 0.8925 -0.0700 0.8925 0.9447
No log 3.2 32 0.7964 0.0481 0.7964 0.8924
No log 3.4 34 0.7693 0.1674 0.7693 0.8771
No log 3.6 36 0.8323 0.2285 0.8323 0.9123
No log 3.8 38 0.8663 0.2319 0.8663 0.9308
No log 4.0 40 0.9179 -0.0045 0.9179 0.9581
No log 4.2 42 1.2157 0.0367 1.2157 1.1026
No log 4.4 44 1.0916 -0.0033 1.0916 1.0448
No log 4.6 46 0.8641 0.2063 0.8641 0.9296
No log 4.8 48 0.8579 0.1550 0.8579 0.9262
No log 5.0 50 0.8885 0.1815 0.8885 0.9426
No log 5.2 52 0.8945 0.1766 0.8945 0.9458
No log 5.4 54 0.8386 0.1699 0.8386 0.9157
No log 5.6 56 0.8781 0.0410 0.8781 0.9371
No log 5.8 58 1.0238 0.0975 1.0238 1.0118
No log 6.0 60 1.0113 0.1259 1.0113 1.0056
No log 6.2 62 0.8929 0.1498 0.8929 0.9449
No log 6.4 64 0.8862 0.1541 0.8862 0.9414
No log 6.6 66 0.8867 0.1541 0.8867 0.9416
No log 6.8 68 0.8958 0.0930 0.8958 0.9464
No log 7.0 70 0.8884 0.1760 0.8884 0.9425
No log 7.2 72 0.9050 0.1866 0.9050 0.9513
No log 7.4 74 1.0495 0.1271 1.0495 1.0245
No log 7.6 76 0.9746 0.1712 0.9746 0.9872
No log 7.8 78 0.8469 0.1303 0.8469 0.9203
No log 8.0 80 0.8490 0.1379 0.8490 0.9214
No log 8.2 82 0.8602 0.1969 0.8602 0.9275
No log 8.4 84 0.8646 0.2747 0.8646 0.9298
No log 8.6 86 0.8753 0.2987 0.8753 0.9356
No log 8.8 88 0.9014 0.2593 0.9014 0.9494
No log 9.0 90 0.8741 0.2888 0.8741 0.9349
No log 9.2 92 0.8777 0.2256 0.8777 0.9369
No log 9.4 94 0.8431 0.2936 0.8431 0.9182
No log 9.6 96 0.8234 0.3296 0.8234 0.9074
No log 9.8 98 0.8382 0.3060 0.8382 0.9155
No log 10.0 100 0.8438 0.3060 0.8438 0.9186
No log 10.2 102 0.8347 0.3478 0.8347 0.9136
No log 10.4 104 0.9014 0.0678 0.9014 0.9494
No log 10.6 106 0.8910 0.0702 0.8910 0.9439
No log 10.8 108 0.8374 0.1379 0.8374 0.9151
No log 11.0 110 0.8317 0.2475 0.8317 0.9120
No log 11.2 112 0.8199 0.2360 0.8199 0.9055
No log 11.4 114 0.8325 0.1797 0.8325 0.9124
No log 11.6 116 0.8529 0.1471 0.8529 0.9235
No log 11.8 118 0.8190 0.1471 0.8190 0.9050
No log 12.0 120 0.7978 0.3002 0.7978 0.8932
No log 12.2 122 0.8029 0.2973 0.8029 0.8961
No log 12.4 124 0.8591 0.2498 0.8591 0.9269
No log 12.6 126 0.8539 0.2633 0.8539 0.9241
No log 12.8 128 0.8333 0.2561 0.8333 0.9129
No log 13.0 130 0.8075 0.1471 0.8075 0.8986
No log 13.2 132 0.8151 0.1341 0.8151 0.9028
No log 13.4 134 0.9087 0.2899 0.9087 0.9533
No log 13.6 136 1.0867 0.1142 1.0867 1.0425
No log 13.8 138 1.0656 0.2055 1.0656 1.0323
No log 14.0 140 0.9129 0.2495 0.9129 0.9555
No log 14.2 142 0.8662 0.2053 0.8662 0.9307
No log 14.4 144 0.9314 0.1433 0.9314 0.9651
No log 14.6 146 0.8753 0.1795 0.8753 0.9356
No log 14.8 148 0.8144 0.2447 0.8144 0.9024
No log 15.0 150 0.8427 0.1740 0.8427 0.9180
No log 15.2 152 0.8586 0.2027 0.8586 0.9266
No log 15.4 154 0.8176 0.2590 0.8176 0.9042
No log 15.6 156 0.8087 0.2058 0.8087 0.8993
No log 15.8 158 0.8122 0.1509 0.8122 0.9012
No log 16.0 160 0.7792 0.0741 0.7792 0.8827
No log 16.2 162 0.7818 0.2007 0.7818 0.8842
No log 16.4 164 0.8351 0.2995 0.8351 0.9138
No log 16.6 166 0.8607 0.2781 0.8607 0.9278
No log 16.8 168 0.8056 0.2784 0.8056 0.8976
No log 17.0 170 0.7397 0.1386 0.7397 0.8601
No log 17.2 172 0.7633 0.1528 0.7633 0.8737
No log 17.4 174 0.7725 0.2611 0.7725 0.8789
No log 17.6 176 0.7114 0.1569 0.7114 0.8435
No log 17.8 178 0.6806 0.1094 0.6806 0.8250
No log 18.0 180 0.6980 0.2981 0.6980 0.8355
No log 18.2 182 0.7073 0.2784 0.7073 0.8410
No log 18.4 184 0.7062 0.2563 0.7062 0.8404
No log 18.6 186 0.7244 0.1969 0.7244 0.8511
No log 18.8 188 0.7478 0.2371 0.7478 0.8648
No log 19.0 190 0.7659 0.2688 0.7659 0.8752
No log 19.2 192 0.7740 0.2688 0.7740 0.8798
No log 19.4 194 0.7655 0.2688 0.7655 0.8749
No log 19.6 196 0.7628 0.2400 0.7628 0.8734
No log 19.8 198 0.7577 0.2661 0.7577 0.8705
No log 20.0 200 0.7963 0.3060 0.7963 0.8924
No log 20.2 202 0.7775 0.2784 0.7775 0.8817
No log 20.4 204 0.7537 0.2784 0.7537 0.8682
No log 20.6 206 0.7288 0.2589 0.7288 0.8537
No log 20.8 208 0.7145 0.2319 0.7145 0.8453
No log 21.0 210 0.7301 0.0755 0.7301 0.8544
No log 21.2 212 0.7600 0.1729 0.7600 0.8718
No log 21.4 214 0.7573 0.1582 0.7573 0.8702
No log 21.6 216 0.8006 0.3500 0.8006 0.8947
No log 21.8 218 0.8706 0.3480 0.8706 0.9331
No log 22.0 220 0.8659 0.3653 0.8659 0.9305
No log 22.2 222 0.7896 0.3060 0.7896 0.8886
No log 22.4 224 0.7553 0.1901 0.7553 0.8691
No log 22.6 226 0.7604 0.0755 0.7604 0.8720
No log 22.8 228 0.7820 0.1130 0.7820 0.8843
No log 23.0 230 0.7619 0.2287 0.7619 0.8729
No log 23.2 232 0.7713 0.1901 0.7713 0.8783
No log 23.4 234 0.8220 0.3157 0.8220 0.9066
No log 23.6 236 0.8328 0.3157 0.8328 0.9126
No log 23.8 238 0.8035 0.3127 0.8035 0.8964
No log 24.0 240 0.7677 0.1303 0.7677 0.8762
No log 24.2 242 0.7570 0.1850 0.7570 0.8700
No log 24.4 244 0.7571 0.1131 0.7571 0.8701
No log 24.6 246 0.7416 0.2148 0.7416 0.8611
No log 24.8 248 0.7365 0.1853 0.7365 0.8582
No log 25.0 250 0.7708 0.3287 0.7708 0.8779
No log 25.2 252 0.7918 0.3368 0.7918 0.8898
No log 25.4 254 0.7537 0.2129 0.7537 0.8681
No log 25.6 256 0.7265 0.2327 0.7265 0.8523
No log 25.8 258 0.7299 0.2327 0.7299 0.8543
No log 26.0 260 0.7616 0.2471 0.7616 0.8727
No log 26.2 262 0.8261 0.3475 0.8261 0.9089
No log 26.4 264 0.8106 0.3127 0.8106 0.9004
No log 26.6 266 0.7694 0.2558 0.7694 0.8772
No log 26.8 268 0.7531 0.3050 0.7531 0.8678
No log 27.0 270 0.7440 0.3369 0.7440 0.8626
No log 27.2 272 0.7383 0.3369 0.7383 0.8593
No log 27.4 274 0.7359 0.3050 0.7359 0.8578
No log 27.6 276 0.7352 0.1953 0.7352 0.8574
No log 27.8 278 0.7769 0.3471 0.7769 0.8814
No log 28.0 280 0.7917 0.3737 0.7917 0.8897
No log 28.2 282 0.7800 0.3544 0.7800 0.8832
No log 28.4 284 0.7564 0.3471 0.7564 0.8697
No log 28.6 286 0.7106 0.2943 0.7106 0.8430
No log 28.8 288 0.6972 0.2965 0.6972 0.8350
No log 29.0 290 0.6907 0.3407 0.6907 0.8311
No log 29.2 292 0.7038 0.1775 0.7038 0.8389
No log 29.4 294 0.6931 0.2484 0.6931 0.8325
No log 29.6 296 0.7004 0.2407 0.7004 0.8369
No log 29.8 298 0.7603 0.3918 0.7603 0.8720
No log 30.0 300 0.8116 0.4167 0.8116 0.9009
No log 30.2 302 0.7948 0.3918 0.7948 0.8915
No log 30.4 304 0.7539 0.2847 0.7539 0.8682
No log 30.6 306 0.7512 0.3171 0.7512 0.8667
No log 30.8 308 0.7763 0.2834 0.7763 0.8811
No log 31.0 310 0.7974 0.2566 0.7974 0.8930
No log 31.2 312 0.8091 0.2514 0.8091 0.8995
No log 31.4 314 0.7969 0.2564 0.7969 0.8927
No log 31.6 316 0.7808 0.1886 0.7808 0.8836
No log 31.8 318 0.7703 0.2475 0.7703 0.8777
No log 32.0 320 0.7665 0.2749 0.7665 0.8755
No log 32.2 322 0.7553 0.2532 0.7553 0.8691
No log 32.4 324 0.7565 0.2247 0.7565 0.8698
No log 32.6 326 0.7605 0.2683 0.7605 0.8721
No log 32.8 328 0.7763 0.2563 0.7763 0.8811
No log 33.0 330 0.7859 0.2479 0.7859 0.8865
No log 33.2 332 0.7889 0.2563 0.7889 0.8882
No log 33.4 334 0.7826 0.2563 0.7826 0.8847
No log 33.6 336 0.7689 0.2563 0.7689 0.8769
No log 33.8 338 0.7572 0.2838 0.7572 0.8702
No log 34.0 340 0.7469 0.2532 0.7469 0.8642
No log 34.2 342 0.7397 0.3224 0.7397 0.8601
No log 34.4 344 0.7330 0.2913 0.7330 0.8562
No log 34.6 346 0.7437 0.2751 0.7437 0.8624
No log 34.8 348 0.7841 0.3475 0.7841 0.8855
No log 35.0 350 0.8427 0.3544 0.8427 0.9180
No log 35.2 352 0.8623 0.3544 0.8623 0.9286
No log 35.4 354 0.8327 0.3544 0.8327 0.9125
No log 35.6 356 0.7730 0.3723 0.7730 0.8792
No log 35.8 358 0.7276 0.3050 0.7276 0.8530
No log 36.0 360 0.7192 0.3675 0.7192 0.8481
No log 36.2 362 0.7212 0.2965 0.7212 0.8492
No log 36.4 364 0.7321 0.3599 0.7321 0.8556
No log 36.6 366 0.7521 0.3763 0.7521 0.8673
No log 36.8 368 0.7615 0.4020 0.7615 0.8726
No log 37.0 370 0.7715 0.3942 0.7715 0.8783
No log 37.2 372 0.7674 0.4190 0.7674 0.8760
No log 37.4 374 0.7396 0.3762 0.7396 0.8600
No log 37.6 376 0.7318 0.2682 0.7318 0.8555
No log 37.8 378 0.7326 0.2986 0.7326 0.8559
No log 38.0 380 0.7203 0.2451 0.7203 0.8487
No log 38.2 382 0.7118 0.1760 0.7118 0.8437
No log 38.4 384 0.7190 0.2063 0.7190 0.8479
No log 38.6 386 0.7312 0.1953 0.7312 0.8551
No log 38.8 388 0.7347 0.1953 0.7347 0.8571
No log 39.0 390 0.7449 0.2558 0.7449 0.8631
No log 39.2 392 0.7482 0.2684 0.7482 0.8650
No log 39.4 394 0.7577 0.3266 0.7577 0.8705
No log 39.6 396 0.7886 0.2568 0.7886 0.8880
No log 39.8 398 0.8069 0.2819 0.8069 0.8983
No log 40.0 400 0.7804 0.3213 0.7804 0.8834
No log 40.2 402 0.7670 0.3530 0.7670 0.8758
No log 40.4 404 0.7696 0.3762 0.7696 0.8772
No log 40.6 406 0.7765 0.3051 0.7765 0.8812
No log 40.8 408 0.7740 0.3051 0.7740 0.8798
No log 41.0 410 0.7634 0.2319 0.7634 0.8738
No log 41.2 412 0.7536 0.2099 0.7536 0.8681
No log 41.4 414 0.7528 0.2063 0.7528 0.8676
No log 41.6 416 0.7517 0.1636 0.7517 0.8670
No log 41.8 418 0.7511 0.1723 0.7511 0.8667
No log 42.0 420 0.7502 0.1051 0.7502 0.8661
No log 42.2 422 0.7505 0.1011 0.7505 0.8663
No log 42.4 424 0.7553 0.1952 0.7553 0.8691
No log 42.6 426 0.7615 0.2973 0.7615 0.8726
No log 42.8 428 0.7690 0.3355 0.7690 0.8769
No log 43.0 430 0.7618 0.3355 0.7618 0.8728
No log 43.2 432 0.7440 0.2621 0.7440 0.8626
No log 43.4 434 0.7387 0.1952 0.7387 0.8595
No log 43.6 436 0.7483 0.0687 0.7483 0.8650
No log 43.8 438 0.7438 0.0679 0.7438 0.8624
No log 44.0 440 0.7253 0.1051 0.7253 0.8516
No log 44.2 442 0.7216 0.1353 0.7216 0.8494
No log 44.4 444 0.7410 0.2847 0.7410 0.8608
No log 44.6 446 0.7733 0.3996 0.7733 0.8794
No log 44.8 448 0.8067 0.3996 0.8067 0.8981
No log 45.0 450 0.7920 0.3996 0.7920 0.8899
No log 45.2 452 0.7533 0.3737 0.7533 0.8679
No log 45.4 454 0.7297 0.2847 0.7297 0.8542
No log 45.6 456 0.7163 0.3649 0.7163 0.8463
No log 45.8 458 0.7164 0.2683 0.7164 0.8464
No log 46.0 460 0.7261 0.2652 0.7261 0.8521
No log 46.2 462 0.7270 0.2652 0.7270 0.8527
No log 46.4 464 0.7175 0.3239 0.7175 0.8470
No log 46.6 466 0.7056 0.2862 0.7056 0.8400
No log 46.8 468 0.7061 0.3280 0.7061 0.8403
No log 47.0 470 0.7150 0.3183 0.7150 0.8456
No log 47.2 472 0.7168 0.3183 0.7168 0.8466
No log 47.4 474 0.7073 0.3280 0.7073 0.8410
No log 47.6 476 0.7004 0.2965 0.7004 0.8369
No log 47.8 478 0.7034 0.3945 0.7034 0.8387
No log 48.0 480 0.7255 0.3737 0.7255 0.8518
No log 48.2 482 0.7459 0.3996 0.7459 0.8636
No log 48.4 484 0.7573 0.3996 0.7573 0.8702
No log 48.6 486 0.7483 0.3996 0.7483 0.8651
No log 48.8 488 0.7261 0.3737 0.7261 0.8521
No log 49.0 490 0.7037 0.2913 0.7037 0.8388
No log 49.2 492 0.7016 0.2590 0.7016 0.8376
No log 49.4 494 0.7008 0.2936 0.7008 0.8371
No log 49.6 496 0.7037 0.2913 0.7037 0.8389
No log 49.8 498 0.7121 0.2847 0.7121 0.8439
0.2538 50.0 500 0.7385 0.3996 0.7385 0.8594
0.2538 50.2 502 0.7637 0.3996 0.7637 0.8739
0.2538 50.4 504 0.7736 0.3475 0.7736 0.8795
0.2538 50.6 506 0.7671 0.3475 0.7671 0.8758
0.2538 50.8 508 0.7650 0.3996 0.7650 0.8746
0.2538 51.0 510 0.7441 0.3088 0.7441 0.8626

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0+cu118
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
-
Safetensors
Model size
0.1B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for MayBashendy/ArabicNewSplits7_usingWellWrittenEssays_FineTuningAraBERT_run1_AugV5_k4_task7_organization

Finetuned
(4033)
this model