ArabicNewSplits7_OSS_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k8_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: 0.7562
  • Qwk: 0.3699
  • Mse: 0.7562
  • Rmse: 0.8696

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 0.05 2 2.5557 -0.0109 2.5557 1.5987
No log 0.1 4 1.3520 0.1234 1.3520 1.1628
No log 0.15 6 0.7202 0.1372 0.7202 0.8487
No log 0.2 8 0.9780 0.1918 0.9780 0.9889
No log 0.25 10 1.1714 0.0324 1.1714 1.0823
No log 0.3 12 1.2621 0.1016 1.2621 1.1234
No log 0.35 14 1.0702 0.1944 1.0702 1.0345
No log 0.4 16 0.8701 0.2027 0.8701 0.9328
No log 0.45 18 0.8076 0.0 0.8076 0.8987
No log 0.5 20 0.7615 0.0 0.7615 0.8726
No log 0.55 22 0.7492 0.0 0.7492 0.8656
No log 0.6 24 0.7533 0.2476 0.7533 0.8679
No log 0.65 26 0.6945 0.3006 0.6945 0.8334
No log 0.7 28 0.6619 0.3238 0.6619 0.8136
No log 0.75 30 0.7215 0.3169 0.7215 0.8494
No log 0.8 32 0.9027 0.2562 0.9027 0.9501
No log 0.85 34 1.1208 0.0264 1.1208 1.0587
No log 0.9 36 1.1017 -0.1391 1.1017 1.0496
No log 0.95 38 0.9895 -0.0970 0.9895 0.9948
No log 1.0 40 1.0519 -0.0066 1.0519 1.0256
No log 1.05 42 1.1037 0.0076 1.1037 1.0506
No log 1.1 44 1.1540 0.0838 1.1540 1.0742
No log 1.15 46 1.0759 0.0095 1.0759 1.0372
No log 1.2 48 1.0552 0.0741 1.0552 1.0272
No log 1.25 50 1.0835 0.1463 1.0835 1.0409
No log 1.3 52 1.0228 0.2227 1.0228 1.0113
No log 1.35 54 0.9181 0.3287 0.9181 0.9582
No log 1.4 56 0.8919 0.2923 0.8919 0.9444
No log 1.45 58 0.8157 0.1539 0.8157 0.9032
No log 1.5 60 0.8552 0.0944 0.8552 0.9248
No log 1.55 62 0.8085 0.0393 0.8085 0.8992
No log 1.6 64 0.7772 0.1459 0.7772 0.8816
No log 1.65 66 0.7415 0.3384 0.7415 0.8611
No log 1.7 68 0.6613 0.4606 0.6613 0.8132
No log 1.75 70 0.6649 0.4476 0.6649 0.8154
No log 1.8 72 0.7581 0.4648 0.7581 0.8707
No log 1.85 74 0.8269 0.3473 0.8269 0.9094
No log 1.9 76 0.7962 0.2116 0.7962 0.8923
No log 1.95 78 0.7817 0.1699 0.7817 0.8841
No log 2.0 80 0.7694 0.1268 0.7694 0.8772
No log 2.05 82 0.7946 0.1373 0.7946 0.8914
No log 2.1 84 0.8436 0.3194 0.8436 0.9185
No log 2.15 86 0.8501 0.3194 0.8501 0.9220
No log 2.2 88 0.8804 0.2643 0.8804 0.9383
No log 2.25 90 0.9316 0.2029 0.9316 0.9652
No log 2.3 92 0.9983 0.1682 0.9983 0.9991
No log 2.35 94 1.0745 0.2910 1.0745 1.0366
No log 2.4 96 1.1425 0.2730 1.1425 1.0689
No log 2.45 98 1.2190 0.2840 1.2190 1.1041
No log 2.5 100 1.3449 0.1916 1.3449 1.1597
No log 2.55 102 1.3656 0.2215 1.3656 1.1686
No log 2.6 104 1.2229 0.2987 1.2229 1.1058
No log 2.65 106 1.0185 0.2389 1.0185 1.0092
No log 2.7 108 0.9542 0.3019 0.9542 0.9769
No log 2.75 110 0.9379 0.3538 0.9379 0.9684
No log 2.8 112 0.8742 0.3409 0.8742 0.9350
No log 2.85 114 0.8114 0.3846 0.8114 0.9008
No log 2.9 116 0.7682 0.3746 0.7682 0.8765
No log 2.95 118 0.7837 0.3425 0.7837 0.8853
No log 3.0 120 0.8186 0.3433 0.8186 0.9047
No log 3.05 122 0.9245 0.3217 0.9245 0.9615
No log 3.1 124 0.9815 0.3727 0.9815 0.9907
No log 3.15 126 1.0473 0.3662 1.0473 1.0234
No log 3.2 128 1.0844 0.3608 1.0844 1.0413
No log 3.25 130 1.1404 0.3505 1.1404 1.0679
No log 3.3 132 1.1143 0.3203 1.1143 1.0556
No log 3.35 134 0.9726 0.2892 0.9726 0.9862
No log 3.4 136 0.8632 0.2227 0.8632 0.9291
No log 3.45 138 0.8596 0.2171 0.8596 0.9272
No log 3.5 140 0.9318 0.2892 0.9318 0.9653
No log 3.55 142 1.0148 0.2872 1.0148 1.0074
No log 3.6 144 1.0959 0.3555 1.0959 1.0469
No log 3.65 146 1.0236 0.2568 1.0236 1.0117
No log 3.7 148 0.9054 0.2836 0.9054 0.9515
No log 3.75 150 0.8654 0.2866 0.8654 0.9303
No log 3.8 152 0.8568 0.2899 0.8568 0.9257
No log 3.85 154 0.8733 0.3433 0.8733 0.9345
No log 3.9 156 0.9238 0.3137 0.9238 0.9611
No log 3.95 158 0.9777 0.3538 0.9777 0.9888
No log 4.0 160 0.8803 0.4032 0.8803 0.9383
No log 4.05 162 0.7471 0.3913 0.7471 0.8644
No log 4.1 164 0.7130 0.2871 0.7130 0.8444
No log 4.15 166 0.7913 0.2883 0.7913 0.8895
No log 4.2 168 0.8273 0.2589 0.8273 0.9095
No log 4.25 170 0.8799 0.2387 0.8799 0.9380
No log 4.3 172 1.0586 0.2707 1.0586 1.0289
No log 4.35 174 1.3079 0.2156 1.3078 1.1436
No log 4.4 176 1.3141 0.2008 1.3141 1.1463
No log 4.45 178 1.1349 0.2392 1.1349 1.0653
No log 4.5 180 1.0644 0.3161 1.0644 1.0317
No log 4.55 182 0.9420 0.3559 0.9420 0.9706
No log 4.6 184 0.8876 0.3618 0.8876 0.9421
No log 4.65 186 0.8079 0.3195 0.8079 0.8988
No log 4.7 188 0.7908 0.3195 0.7908 0.8893
No log 4.75 190 0.8670 0.4032 0.8670 0.9311
No log 4.8 192 1.1604 0.3412 1.1604 1.0772
No log 4.85 194 1.3661 0.3510 1.3661 1.1688
No log 4.9 196 1.3620 0.3509 1.3620 1.1670
No log 4.95 198 1.0754 0.3003 1.0754 1.0370
No log 5.0 200 0.8743 0.3645 0.8743 0.9350
No log 5.05 202 0.8825 0.3159 0.8825 0.9394
No log 5.1 204 1.0261 0.3849 1.0261 1.0130
No log 5.15 206 1.0794 0.4116 1.0794 1.0389
No log 5.2 208 0.9734 0.3849 0.9734 0.9866
No log 5.25 210 0.8809 0.4332 0.8809 0.9386
No log 5.3 212 0.7995 0.3543 0.7995 0.8941
No log 5.35 214 0.7300 0.3824 0.7300 0.8544
No log 5.4 216 0.6713 0.2839 0.6713 0.8193
No log 5.45 218 0.6708 0.3500 0.6708 0.8190
No log 5.5 220 0.7279 0.3645 0.7279 0.8532
No log 5.55 222 0.7737 0.3847 0.7737 0.8796
No log 5.6 224 0.8127 0.3892 0.8127 0.9015
No log 5.65 226 0.9289 0.3597 0.9289 0.9638
No log 5.7 228 1.0246 0.3503 1.0246 1.0122
No log 5.75 230 1.0520 0.3503 1.0520 1.0257
No log 5.8 232 1.0846 0.3455 1.0846 1.0414
No log 5.85 234 1.0424 0.3154 1.0424 1.0210
No log 5.9 236 0.9727 0.3052 0.9727 0.9863
No log 5.95 238 1.0000 0.3697 1.0000 1.0000
No log 6.0 240 1.0950 0.3358 1.0950 1.0464
No log 6.05 242 1.0870 0.3129 1.0870 1.0426
No log 6.1 244 0.9516 0.3650 0.9516 0.9755
No log 6.15 246 0.8101 0.2692 0.8101 0.9001
No log 6.2 248 0.7525 0.3494 0.7525 0.8675
No log 6.25 250 0.7399 0.2819 0.7399 0.8602
No log 6.3 252 0.7912 0.3475 0.7912 0.8895
No log 6.35 254 0.9267 0.3613 0.9267 0.9627
No log 6.4 256 0.9245 0.3613 0.9245 0.9615
No log 6.45 258 0.7983 0.3499 0.7983 0.8935
No log 6.5 260 0.7217 0.4020 0.7217 0.8495
No log 6.55 262 0.7027 0.4020 0.7027 0.8383
No log 6.6 264 0.7420 0.3384 0.7420 0.8614
No log 6.65 266 0.8657 0.3657 0.8657 0.9304
No log 6.7 268 1.0013 0.3317 1.0013 1.0006
No log 6.75 270 1.0088 0.3662 1.0088 1.0044
No log 6.8 272 0.8769 0.3618 0.8769 0.9364
No log 6.85 274 0.7035 0.4251 0.7035 0.8388
No log 6.9 276 0.6359 0.3341 0.6359 0.7974
No log 6.95 278 0.6218 0.3893 0.6218 0.7885
No log 7.0 280 0.6421 0.3267 0.6421 0.8013
No log 7.05 282 0.7697 0.4123 0.7697 0.8773
No log 7.1 284 0.9153 0.3466 0.9153 0.9567
No log 7.15 286 0.9488 0.3669 0.9488 0.9740
No log 7.2 288 0.8454 0.3913 0.8454 0.9194
No log 7.25 290 0.7418 0.3662 0.7418 0.8613
No log 7.3 292 0.6934 0.3518 0.6934 0.8327
No log 7.35 294 0.6690 0.3099 0.6690 0.8179
No log 7.4 296 0.6682 0.4329 0.6682 0.8175
No log 7.45 298 0.7267 0.4404 0.7267 0.8524
No log 7.5 300 0.7144 0.4018 0.7144 0.8452
No log 7.55 302 0.6181 0.4808 0.6181 0.7862
No log 7.6 304 0.6078 0.4582 0.6078 0.7796
No log 7.65 306 0.6819 0.3963 0.6819 0.8258
No log 7.7 308 0.6959 0.3329 0.6959 0.8342
No log 7.75 310 0.7204 0.3737 0.7204 0.8487
No log 7.8 312 0.7510 0.3196 0.7510 0.8666
No log 7.85 314 0.7986 0.3060 0.7986 0.8937
No log 7.9 316 0.8935 0.2492 0.8935 0.9452
No log 7.95 318 1.0482 0.2659 1.0482 1.0238
No log 8.0 320 1.1497 0.2398 1.1497 1.0722
No log 8.05 322 1.0853 0.2461 1.0853 1.0418
No log 8.1 324 0.8982 0.2812 0.8982 0.9477
No log 8.15 326 0.7810 0.2883 0.7810 0.8837
No log 8.2 328 0.7452 0.2718 0.7452 0.8633
No log 8.25 330 0.7539 0.2883 0.7539 0.8683
No log 8.3 332 0.8486 0.3008 0.8486 0.9212
No log 8.35 334 0.9382 0.3076 0.9382 0.9686
No log 8.4 336 0.8991 0.3799 0.8991 0.9482
No log 8.45 338 0.8649 0.3799 0.8649 0.9300
No log 8.5 340 0.7619 0.3843 0.7619 0.8729
No log 8.55 342 0.6785 0.3763 0.6785 0.8237
No log 8.6 344 0.6574 0.3763 0.6574 0.8108
No log 8.65 346 0.6757 0.3737 0.6757 0.8220
No log 8.7 348 0.7747 0.4102 0.7747 0.8802
No log 8.75 350 0.8890 0.4348 0.8890 0.9429
No log 8.8 352 0.9016 0.4270 0.9016 0.9495
No log 8.85 354 0.7952 0.4142 0.7952 0.8917
No log 8.9 356 0.6618 0.3794 0.6618 0.8135
No log 8.95 358 0.6246 0.3594 0.6246 0.7903
No log 9.0 360 0.6402 0.3794 0.6402 0.8001
No log 9.05 362 0.7336 0.4721 0.7336 0.8565
No log 9.1 364 0.8488 0.4142 0.8488 0.9213
No log 9.15 366 0.9212 0.4246 0.9212 0.9598
No log 9.2 368 0.9234 0.4108 0.9234 0.9609
No log 9.25 370 0.8164 0.4142 0.8164 0.9035
No log 9.3 372 0.6873 0.3918 0.6873 0.8290
No log 9.35 374 0.6386 0.3312 0.6386 0.7991
No log 9.4 376 0.6244 0.3312 0.6244 0.7902
No log 9.45 378 0.6249 0.3312 0.6249 0.7905
No log 9.5 380 0.6853 0.4089 0.6853 0.8278
No log 9.55 382 0.8300 0.4030 0.8300 0.9111
No log 9.6 384 0.9393 0.4315 0.9393 0.9692
No log 9.65 386 0.9600 0.4149 0.9600 0.9798
No log 9.7 388 0.8846 0.3601 0.8846 0.9405
No log 9.75 390 0.7857 0.4404 0.7857 0.8864
No log 9.8 392 0.7322 0.4251 0.7322 0.8557
No log 9.85 394 0.6941 0.3261 0.6941 0.8331
No log 9.9 396 0.6943 0.3261 0.6943 0.8332
No log 9.95 398 0.7025 0.4014 0.7025 0.8382
No log 10.0 400 0.7361 0.4251 0.7361 0.8580
No log 10.05 402 0.8080 0.4030 0.8080 0.8989
No log 10.1 404 0.8777 0.3827 0.8777 0.9368
No log 10.15 406 0.8670 0.3310 0.8670 0.9311
No log 10.2 408 0.8058 0.3008 0.8058 0.8977
No log 10.25 410 0.7995 0.3008 0.7995 0.8941
No log 10.3 412 0.7901 0.3008 0.7901 0.8889
No log 10.35 414 0.8015 0.2643 0.8015 0.8953
No log 10.4 416 0.8377 0.2892 0.8377 0.9153
No log 10.45 418 0.8139 0.2697 0.8139 0.9021
No log 10.5 420 0.7883 0.1946 0.7883 0.8879
No log 10.55 422 0.8055 0.1946 0.8055 0.8975
No log 10.6 424 0.8577 0.3319 0.8577 0.9261
No log 10.65 426 0.8514 0.3731 0.8514 0.9227
No log 10.7 428 0.8420 0.4102 0.8420 0.9176
No log 10.75 430 0.7844 0.4175 0.7844 0.8857
No log 10.8 432 0.7945 0.4175 0.7945 0.8914
No log 10.85 434 0.7965 0.4275 0.7965 0.8925
No log 10.9 436 0.7741 0.3008 0.7741 0.8798
No log 10.95 438 0.7716 0.3450 0.7716 0.8784
No log 11.0 440 0.7532 0.3590 0.7532 0.8679
No log 11.05 442 0.7613 0.3450 0.7613 0.8725
No log 11.1 444 0.7814 0.3562 0.7814 0.8840
No log 11.15 446 0.8252 0.3076 0.8252 0.9084
No log 11.2 448 0.7928 0.3433 0.7928 0.8904
No log 11.25 450 0.7632 0.4562 0.7632 0.8736
No log 11.3 452 0.7467 0.4562 0.7467 0.8641
No log 11.35 454 0.7273 0.4562 0.7273 0.8528
No log 11.4 456 0.7359 0.4175 0.7359 0.8579
No log 11.45 458 0.7671 0.4030 0.7671 0.8759
No log 11.5 460 0.7560 0.4030 0.7560 0.8695
No log 11.55 462 0.7125 0.3799 0.7125 0.8441
No log 11.6 464 0.7272 0.3799 0.7272 0.8527
No log 11.65 466 0.7677 0.4030 0.7677 0.8762
No log 11.7 468 0.7934 0.3665 0.7934 0.8908
No log 11.75 470 0.7636 0.3497 0.7636 0.8738
No log 11.8 472 0.7188 0.3261 0.7188 0.8478
No log 11.85 474 0.6902 0.3737 0.6902 0.8308
No log 11.9 476 0.7141 0.3261 0.7141 0.8450
No log 11.95 478 0.8117 0.3719 0.8117 0.9010
No log 12.0 480 0.8741 0.3559 0.8741 0.9349
No log 12.05 482 0.8300 0.3346 0.8300 0.9110
No log 12.1 484 0.7262 0.3450 0.7262 0.8522
No log 12.15 486 0.6782 0.3737 0.6782 0.8235
No log 12.2 488 0.6643 0.4489 0.6643 0.8151
No log 12.25 490 0.6844 0.3167 0.6844 0.8273
No log 12.3 492 0.7589 0.3630 0.7589 0.8711
No log 12.35 494 0.8125 0.3601 0.8125 0.9014
No log 12.4 496 0.8106 0.3665 0.8106 0.9003
No log 12.45 498 0.8167 0.3782 0.8167 0.9037
0.3588 12.5 500 0.8744 0.3827 0.8744 0.9351
0.3588 12.55 502 0.8466 0.3988 0.8466 0.9201
0.3588 12.6 504 0.7667 0.3333 0.7667 0.8756
0.3588 12.65 506 0.7560 0.3280 0.7560 0.8695
0.3588 12.7 508 0.7462 0.3280 0.7462 0.8639
0.3588 12.75 510 0.7453 0.3699 0.7453 0.8633
0.3588 12.8 512 0.7562 0.3699 0.7562 0.8696

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_OSS_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k8_task7_organization

Finetuned
(4019)
this model