ArabicNewSplits7_FineTuningAraBERT_run1_AugV5_k18_task3_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: 0.9543
  • Qwk: -0.1128
  • Mse: 0.9543
  • Rmse: 0.9769

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 0.0417 2 3.4611 -0.0152 3.4611 1.8604
No log 0.0833 4 1.8709 0.0560 1.8709 1.3678
No log 0.125 6 1.4564 0.0235 1.4564 1.2068
No log 0.1667 8 1.6745 0.0153 1.6745 1.2940
No log 0.2083 10 1.2798 0.0 1.2798 1.1313
No log 0.25 12 1.0826 0.0279 1.0826 1.0405
No log 0.2917 14 1.1023 0.0048 1.1023 1.0499
No log 0.3333 16 1.0993 -0.0164 1.0993 1.0485
No log 0.375 18 0.8375 -0.2068 0.8375 0.9152
No log 0.4167 20 0.7335 0.0460 0.7335 0.8564
No log 0.4583 22 0.8677 -0.1257 0.8677 0.9315
No log 0.5 24 0.9298 -0.0909 0.9298 0.9642
No log 0.5417 26 0.8831 -0.0870 0.8831 0.9397
No log 0.5833 28 0.8916 -0.0504 0.8916 0.9443
No log 0.625 30 0.9487 0.0909 0.9487 0.9740
No log 0.6667 32 0.9047 0.0111 0.9047 0.9511
No log 0.7083 34 0.9458 0.0878 0.9458 0.9725
No log 0.75 36 0.9012 0.0134 0.9012 0.9493
No log 0.7917 38 0.9472 0.0305 0.9472 0.9732
No log 0.8333 40 0.8250 -0.0801 0.8250 0.9083
No log 0.875 42 0.7754 -0.0753 0.7754 0.8806
No log 0.9167 44 0.7946 -0.0309 0.7946 0.8914
No log 0.9583 46 0.7674 -0.0188 0.7674 0.8760
No log 1.0 48 0.8329 -0.0122 0.8329 0.9126
No log 1.0417 50 0.8878 0.0134 0.8878 0.9422
No log 1.0833 52 0.7686 -0.0766 0.7686 0.8767
No log 1.125 54 0.7778 -0.0695 0.7778 0.8819
No log 1.1667 56 1.2998 -0.0155 1.2998 1.1401
No log 1.2083 58 1.8510 -0.0053 1.8510 1.3605
No log 1.25 60 1.5206 -0.0503 1.5206 1.2331
No log 1.2917 62 1.1744 -0.0234 1.1744 1.0837
No log 1.3333 64 0.9257 -0.0200 0.9257 0.9622
No log 1.375 66 0.9552 -0.0218 0.9552 0.9774
No log 1.4167 68 0.8924 -0.0218 0.8924 0.9447
No log 1.4583 70 0.7715 0.0512 0.7715 0.8784
No log 1.5 72 0.7155 0.0909 0.7155 0.8459
No log 1.5417 74 0.7017 0.1021 0.7017 0.8377
No log 1.5833 76 0.8426 0.0642 0.8426 0.9179
No log 1.625 78 1.0723 0.0518 1.0723 1.0355
No log 1.6667 80 0.9495 -0.0269 0.9495 0.9744
No log 1.7083 82 0.7662 0.1047 0.7662 0.8753
No log 1.75 84 0.7857 -0.0204 0.7857 0.8864
No log 1.7917 86 0.9120 -0.0175 0.9120 0.9550
No log 1.8333 88 1.0453 -0.0013 1.0453 1.0224
No log 1.875 90 0.8986 -0.0442 0.8986 0.9479
No log 1.9167 92 0.8513 -0.0008 0.8513 0.9227
No log 1.9583 94 0.8540 0.0711 0.8540 0.9241
No log 2.0 96 0.7770 0.0296 0.7770 0.8815
No log 2.0417 98 0.7630 -0.0644 0.7630 0.8735
No log 2.0833 100 0.7777 0.0282 0.7777 0.8818
No log 2.125 102 0.8232 0.0316 0.8232 0.9073
No log 2.1667 104 0.7972 0.0867 0.7972 0.8929
No log 2.2083 106 0.7364 -0.0215 0.7364 0.8581
No log 2.25 108 0.7799 -0.1332 0.7799 0.8831
No log 2.2917 110 0.8557 -0.1592 0.8557 0.9250
No log 2.3333 112 0.9124 -0.1466 0.9124 0.9552
No log 2.375 114 0.8699 0.0393 0.8699 0.9327
No log 2.4167 116 0.8838 -0.0209 0.8838 0.9401
No log 2.4583 118 0.8236 0.0303 0.8236 0.9075
No log 2.5 120 0.8499 0.0476 0.8499 0.9219
No log 2.5417 122 0.8043 0.0588 0.8043 0.8968
No log 2.5833 124 0.7307 -0.0033 0.7307 0.8548
No log 2.625 126 0.7649 -0.0679 0.7649 0.8746
No log 2.6667 128 0.7165 -0.0131 0.7165 0.8464
No log 2.7083 130 0.7480 0.0225 0.7480 0.8649
No log 2.75 132 0.7788 0.1097 0.7788 0.8825
No log 2.7917 134 0.8769 0.0786 0.8769 0.9364
No log 2.8333 136 0.8232 0.0867 0.8232 0.9073
No log 2.875 138 0.8012 0.0999 0.8012 0.8951
No log 2.9167 140 0.7686 -0.0179 0.7686 0.8767
No log 2.9583 142 0.7966 -0.0179 0.7966 0.8925
No log 3.0 144 0.9300 0.0871 0.9300 0.9644
No log 3.0417 146 1.0081 0.0333 1.0081 1.0040
No log 3.0833 148 0.9291 -0.1740 0.9291 0.9639
No log 3.125 150 0.9853 -0.0365 0.9853 0.9926
No log 3.1667 152 0.9383 -0.1517 0.9383 0.9687
No log 3.2083 154 1.2111 0.0810 1.2111 1.1005
No log 3.25 156 1.1108 0.0956 1.1108 1.0539
No log 3.2917 158 0.9548 -0.1885 0.9548 0.9772
No log 3.3333 160 0.9898 -0.1156 0.9898 0.9949
No log 3.375 162 0.9645 -0.0457 0.9645 0.9821
No log 3.4167 164 0.9673 -0.1038 0.9673 0.9835
No log 3.4583 166 0.8849 -0.0643 0.8849 0.9407
No log 3.5 168 0.8346 -0.0469 0.8346 0.9135
No log 3.5417 170 0.8484 0.1641 0.8484 0.9211
No log 3.5833 172 0.8578 -0.0774 0.8578 0.9262
No log 3.625 174 0.9291 -0.2388 0.9291 0.9639
No log 3.6667 176 0.9780 -0.0082 0.9780 0.9889
No log 3.7083 178 1.1264 0.0125 1.1264 1.0613
No log 3.75 180 1.3844 -0.0870 1.3844 1.1766
No log 3.7917 182 1.4045 -0.0870 1.4045 1.1851
No log 3.8333 184 1.1842 -0.1384 1.1842 1.0882
No log 3.875 186 1.1489 -0.1293 1.1489 1.0719
No log 3.9167 188 1.1388 -0.0697 1.1388 1.0672
No log 3.9583 190 0.8949 -0.0851 0.8949 0.9460
No log 4.0 192 0.9606 0.0092 0.9606 0.9801
No log 4.0417 194 0.9537 0.0092 0.9537 0.9766
No log 4.0833 196 0.8376 -0.0334 0.8376 0.9152
No log 4.125 198 0.9769 0.0476 0.9769 0.9884
No log 4.1667 200 0.9983 0.0442 0.9983 0.9991
No log 4.2083 202 0.9399 -0.0457 0.9399 0.9695
No log 4.25 204 1.0482 0.0125 1.0482 1.0238
No log 4.2917 206 1.1856 0.0516 1.1856 1.0888
No log 4.3333 208 1.0907 0.0794 1.0907 1.0444
No log 4.375 210 1.0206 -0.0822 1.0206 1.0103
No log 4.4167 212 1.3393 -0.0237 1.3393 1.1573
No log 4.4583 214 1.4188 -0.0311 1.4188 1.1911
No log 4.5 216 1.1209 -0.0500 1.1209 1.0587
No log 4.5417 218 0.8949 -0.2683 0.8949 0.9460
No log 4.5833 220 0.9354 -0.0492 0.9354 0.9672
No log 4.625 222 0.9610 -0.0685 0.9610 0.9803
No log 4.6667 224 0.8999 -0.0528 0.8999 0.9487
No log 4.7083 226 0.8561 -0.2374 0.8561 0.9252
No log 4.75 228 0.9899 -0.0056 0.9899 0.9949
No log 4.7917 230 1.0328 -0.0101 1.0328 1.0163
No log 4.8333 232 0.9508 -0.0408 0.9508 0.9751
No log 4.875 234 0.8725 -0.0658 0.8725 0.9341
No log 4.9167 236 0.8413 -0.0427 0.8413 0.9172
No log 4.9583 238 0.8309 0.0181 0.8309 0.9116
No log 5.0 240 0.7862 0.0571 0.7862 0.8867
No log 5.0417 242 0.7540 0.0524 0.7540 0.8684
No log 5.0833 244 0.8105 -0.1239 0.8105 0.9003
No log 5.125 246 0.8880 -0.0351 0.8880 0.9423
No log 5.1667 248 0.9101 -0.1244 0.9101 0.9540
No log 5.2083 250 0.9108 -0.1841 0.9108 0.9544
No log 5.25 252 0.9414 -0.1219 0.9414 0.9702
No log 5.2917 254 0.9494 -0.1334 0.9494 0.9744
No log 5.3333 256 0.9632 -0.1394 0.9632 0.9814
No log 5.375 258 0.9871 -0.1690 0.9871 0.9935
No log 5.4167 260 1.0754 -0.1152 1.0754 1.0370
No log 5.4583 262 1.0753 -0.1152 1.0753 1.0370
No log 5.5 264 1.0898 -0.1152 1.0898 1.0439
No log 5.5417 266 0.9872 -0.2175 0.9872 0.9936
No log 5.5833 268 1.0207 -0.0927 1.0207 1.0103
No log 5.625 270 0.9983 -0.0920 0.9983 0.9991
No log 5.6667 272 0.9809 -0.2087 0.9809 0.9904
No log 5.7083 274 1.0760 -0.1722 1.0760 1.0373
No log 5.75 276 1.1231 -0.1012 1.1231 1.0598
No log 5.7917 278 1.1605 -0.1261 1.1605 1.0773
No log 5.8333 280 1.1073 -0.1463 1.1073 1.0523
No log 5.875 282 1.0365 -0.1043 1.0365 1.0181
No log 5.9167 284 0.9592 -0.1407 0.9592 0.9794
No log 5.9583 286 0.9084 -0.0672 0.9084 0.9531
No log 6.0 288 0.8789 -0.0099 0.8789 0.9375
No log 6.0417 290 0.8568 -0.0099 0.8568 0.9257
No log 6.0833 292 0.8792 -0.0217 0.8792 0.9377
No log 6.125 294 0.9978 0.0446 0.9978 0.9989
No log 6.1667 296 1.0465 0.0446 1.0465 1.0230
No log 6.2083 298 0.9807 -0.0409 0.9807 0.9903
No log 6.25 300 1.0806 0.0200 1.0806 1.0395
No log 6.2917 302 1.3470 -0.1268 1.3470 1.1606
No log 6.3333 304 1.2618 -0.1732 1.2618 1.1233
No log 6.375 306 1.0864 -0.0646 1.0864 1.0423
No log 6.4167 308 1.1261 -0.0905 1.1261 1.0612
No log 6.4583 310 1.0945 0.0067 1.0945 1.0462
No log 6.5 312 0.9445 -0.0920 0.9445 0.9718
No log 6.5417 314 0.8828 -0.0939 0.8828 0.9396
No log 6.5833 316 0.8618 -0.0557 0.8618 0.9283
No log 6.625 318 0.8483 -0.0984 0.8483 0.9210
No log 6.6667 320 0.8749 -0.0647 0.8749 0.9354
No log 6.7083 322 0.9119 -0.0660 0.9119 0.9549
No log 6.75 324 0.9149 -0.1166 0.9149 0.9565
No log 6.7917 326 0.9231 -0.0764 0.9231 0.9608
No log 6.8333 328 0.9653 -0.1093 0.9653 0.9825
No log 6.875 330 0.9784 -0.1093 0.9784 0.9891
No log 6.9167 332 0.9236 -0.0588 0.9236 0.9610
No log 6.9583 334 0.9477 -0.1459 0.9477 0.9735
No log 7.0 336 0.9822 -0.1355 0.9822 0.9910
No log 7.0417 338 1.0405 -0.1232 1.0405 1.0200
No log 7.0833 340 1.1159 0.0497 1.1159 1.0564
No log 7.125 342 1.0729 0.0497 1.0729 1.0358
No log 7.1667 344 1.0171 -0.0717 1.0171 1.0085
No log 7.2083 346 1.1561 -0.0476 1.1561 1.0752
No log 7.25 348 1.1815 -0.0120 1.1815 1.0870
No log 7.2917 350 1.0872 -0.1325 1.0872 1.0427
No log 7.3333 352 1.0041 -0.0695 1.0041 1.0021
No log 7.375 354 0.9728 -0.0970 0.9728 0.9863
No log 7.4167 356 0.9408 -0.1770 0.9408 0.9700
No log 7.4583 358 0.9410 -0.1826 0.9410 0.9700
No log 7.5 360 0.8908 -0.2580 0.8908 0.9438
No log 7.5417 362 0.8726 -0.2614 0.8726 0.9341
No log 7.5833 364 0.8984 0.0225 0.8984 0.9479
No log 7.625 366 0.9137 -0.1121 0.9137 0.9559
No log 7.6667 368 0.9373 -0.0533 0.9373 0.9681
No log 7.7083 370 0.9607 -0.2017 0.9607 0.9801
No log 7.75 372 0.9631 -0.2017 0.9631 0.9814
No log 7.7917 374 0.9557 -0.0479 0.9557 0.9776
No log 7.8333 376 1.0390 -0.0425 1.0390 1.0193
No log 7.875 378 1.0777 -0.0861 1.0777 1.0381
No log 7.9167 380 1.0874 -0.0474 1.0874 1.0428
No log 7.9583 382 1.0257 -0.0946 1.0257 1.0128
No log 8.0 384 1.0096 -0.1170 1.0096 1.0048
No log 8.0417 386 1.0034 -0.0784 1.0034 1.0017
No log 8.0833 388 1.0344 -0.0679 1.0344 1.0170
No log 8.125 390 1.1565 -0.0583 1.1565 1.0754
No log 8.1667 392 1.1358 -0.0163 1.1358 1.0658
No log 8.2083 394 1.0059 0.0129 1.0059 1.0029
No log 8.25 396 0.9698 -0.0138 0.9698 0.9848
No log 8.2917 398 0.9436 -0.0491 0.9436 0.9714
No log 8.3333 400 0.9680 -0.0669 0.9680 0.9839
No log 8.375 402 0.9674 -0.0991 0.9674 0.9836
No log 8.4167 404 0.9480 -0.0339 0.9480 0.9737
No log 8.4583 406 0.9779 -0.1205 0.9779 0.9889
No log 8.5 408 0.9945 -0.0734 0.9945 0.9972
No log 8.5417 410 1.0008 -0.0844 1.0008 1.0004
No log 8.5833 412 0.9935 -0.0844 0.9935 0.9967
No log 8.625 414 0.9931 -0.0744 0.9931 0.9966
No log 8.6667 416 0.9825 -0.0343 0.9825 0.9912
No log 8.7083 418 0.9573 -0.0734 0.9573 0.9784
No log 8.75 420 0.9229 -0.1280 0.9229 0.9607
No log 8.7917 422 0.8934 -0.0408 0.8934 0.9452
No log 8.8333 424 0.9207 -0.0896 0.9207 0.9595
No log 8.875 426 0.9305 -0.0508 0.9305 0.9646
No log 8.9167 428 0.9134 0.0093 0.9134 0.9557
No log 8.9583 430 0.9454 -0.0706 0.9454 0.9723
No log 9.0 432 1.0018 -0.0171 1.0018 1.0009
No log 9.0417 434 1.0154 -0.0804 1.0154 1.0077
No log 9.0833 436 1.0166 0.0045 1.0166 1.0082
No log 9.125 438 1.0215 -0.0393 1.0215 1.0107
No log 9.1667 440 1.0460 -0.0534 1.0460 1.0228
No log 9.2083 442 1.1211 0.0038 1.1211 1.0588
No log 9.25 444 1.0750 -0.1044 1.0750 1.0368
No log 9.2917 446 0.9821 -0.1447 0.9821 0.9910
No log 9.3333 448 0.9275 -0.1521 0.9275 0.9630
No log 9.375 450 0.9019 -0.1524 0.9019 0.9497
No log 9.4167 452 0.8946 -0.1524 0.8946 0.9458
No log 9.4583 454 0.9394 -0.1568 0.9394 0.9692
No log 9.5 456 1.0438 -0.0056 1.0438 1.0216
No log 9.5417 458 1.0303 -0.0442 1.0303 1.0151
No log 9.5833 460 1.0422 -0.0728 1.0422 1.0209
No log 9.625 462 1.0099 -0.1234 1.0099 1.0049
No log 9.6667 464 1.0372 -0.0658 1.0372 1.0184
No log 9.7083 466 1.0399 -0.0351 1.0399 1.0198
No log 9.75 468 1.0128 -0.0822 1.0128 1.0064
No log 9.7917 470 0.9454 -0.0171 0.9454 0.9723
No log 9.8333 472 0.8871 -0.0533 0.8871 0.9419
No log 9.875 474 0.8848 -0.0734 0.8848 0.9406
No log 9.9167 476 0.9057 -0.0127 0.9057 0.9517
No log 9.9583 478 0.9267 -0.0099 0.9267 0.9627
No log 10.0 480 1.0283 -0.0309 1.0283 1.0141
No log 10.0417 482 1.0448 0.0084 1.0448 1.0222
No log 10.0833 484 0.9861 -0.0347 0.9861 0.9930
No log 10.125 486 1.0066 0.0087 1.0066 1.0033
No log 10.1667 488 1.0883 -0.0175 1.0883 1.0432
No log 10.2083 490 1.0678 -0.0214 1.0678 1.0334
No log 10.25 492 0.9649 -0.0611 0.9649 0.9823
No log 10.2917 494 0.9612 0.0181 0.9612 0.9804
No log 10.3333 496 1.0352 -0.0031 1.0352 1.0174
No log 10.375 498 0.9926 0.0470 0.9926 0.9963
0.3887 10.4167 500 0.9584 -0.0755 0.9584 0.9790
0.3887 10.4583 502 1.0489 -0.0055 1.0489 1.0242
0.3887 10.5 504 1.0915 0.0551 1.0915 1.0448
0.3887 10.5417 506 1.0713 -0.1656 1.0713 1.0350
0.3887 10.5833 508 1.1419 0.0461 1.1419 1.0686
0.3887 10.625 510 1.1540 0.0419 1.1540 1.0743
0.3887 10.6667 512 1.0454 -0.0395 1.0454 1.0225
0.3887 10.7083 514 1.0314 0.0262 1.0314 1.0156
0.3887 10.75 516 1.1115 0.0190 1.1115 1.0543
0.3887 10.7917 518 1.0799 0.0138 1.0799 1.0392
0.3887 10.8333 520 0.9900 -0.0368 0.9900 0.9950
0.3887 10.875 522 0.9908 0.0265 0.9908 0.9954
0.3887 10.9167 524 1.1414 0.0341 1.1414 1.0683
0.3887 10.9583 526 1.1815 0.0316 1.1815 1.0870
0.3887 11.0 528 1.0850 0.0448 1.0850 1.0416
0.3887 11.0417 530 1.0060 -0.0784 1.0060 1.0030
0.3887 11.0833 532 1.0100 -0.1406 1.0100 1.0050
0.3887 11.125 534 1.0096 -0.1406 1.0096 1.0048
0.3887 11.1667 536 1.0363 0.0573 1.0363 1.0180
0.3887 11.2083 538 1.1131 0.0781 1.1131 1.0550
0.3887 11.25 540 1.1028 0.0473 1.1028 1.0501
0.3887 11.2917 542 1.0503 -0.1456 1.0503 1.0248
0.3887 11.3333 544 1.0465 -0.1258 1.0465 1.0230
0.3887 11.375 546 0.9836 -0.1355 0.9836 0.9918
0.3887 11.4167 548 0.9183 -0.0923 0.9183 0.9583
0.3887 11.4583 550 0.9176 -0.0616 0.9176 0.9579
0.3887 11.5 552 0.9375 -0.0241 0.9375 0.9682
0.3887 11.5417 554 0.9556 -0.0616 0.9556 0.9775
0.3887 11.5833 556 0.9555 -0.1569 0.9555 0.9775
0.3887 11.625 558 0.9730 -0.1459 0.9730 0.9864
0.3887 11.6667 560 0.9716 -0.1458 0.9716 0.9857
0.3887 11.7083 562 0.9537 -0.1176 0.9537 0.9766
0.3887 11.75 564 0.9543 -0.1128 0.9543 0.9769

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0+cu118
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
0
Safetensors
Model size
135M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for MayBashendy/ArabicNewSplits7_FineTuningAraBERT_run1_AugV5_k18_task3_organization

Finetuned
(2729)
this model