ArabicNewSplits7_FineTuningAraBERT_run1_AugV5_k20_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: 0.7868
  • Qwk: 0.0116
  • Mse: 0.7868
  • Rmse: 0.8870

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 0.0377 2 3.7032 -0.0073 3.7032 1.9244
No log 0.0755 4 2.0761 0.0348 2.0761 1.4409
No log 0.1132 6 1.2712 -0.0746 1.2712 1.1275
No log 0.1509 8 1.3514 0.0283 1.3514 1.1625
No log 0.1887 10 0.9679 -0.0545 0.9679 0.9838
No log 0.2264 12 0.7573 -0.0309 0.7573 0.8702
No log 0.2642 14 0.9645 0.0543 0.9645 0.9821
No log 0.3019 16 1.0338 -0.0744 1.0338 1.0168
No log 0.3396 18 1.0595 -0.0247 1.0595 1.0293
No log 0.3774 20 1.0185 -0.0247 1.0185 1.0092
No log 0.4151 22 0.8154 0.1047 0.8154 0.9030
No log 0.4528 24 0.7299 0.1082 0.7299 0.8544
No log 0.4906 26 0.7400 0.0460 0.7400 0.8602
No log 0.5283 28 0.8081 -0.0309 0.8081 0.8989
No log 0.5660 30 0.8773 0.1107 0.8773 0.9366
No log 0.6038 32 0.9366 -0.0949 0.9366 0.9678
No log 0.6415 34 0.9683 -0.0345 0.9683 0.9840
No log 0.6792 36 1.0020 -0.0435 1.0020 1.0010
No log 0.7170 38 0.9911 -0.0992 0.9911 0.9955
No log 0.7547 40 0.9251 -0.0930 0.9251 0.9618
No log 0.7925 42 0.9406 -0.0617 0.9406 0.9699
No log 0.8302 44 0.9932 -0.0658 0.9932 0.9966
No log 0.8679 46 1.0306 -0.0704 1.0306 1.0152
No log 0.9057 48 1.0964 -0.0149 1.0964 1.0471
No log 0.9434 50 1.0048 -0.0668 1.0048 1.0024
No log 0.9811 52 1.0239 -0.0982 1.0239 1.0119
No log 1.0189 54 1.1424 -0.0247 1.1424 1.0688
No log 1.0566 56 1.0895 0.0279 1.0895 1.0438
No log 1.0943 58 0.9856 -0.0966 0.9856 0.9928
No log 1.1321 60 0.8875 -0.0490 0.8875 0.9421
No log 1.1698 62 0.8901 -0.0861 0.8901 0.9435
No log 1.2075 64 0.8203 -0.0766 0.8203 0.9057
No log 1.2453 66 0.8302 -0.1738 0.8302 0.9112
No log 1.2830 68 0.8504 -0.1249 0.8504 0.9222
No log 1.3208 70 0.8280 -0.1715 0.8280 0.9100
No log 1.3585 72 0.8241 -0.1244 0.8241 0.9078
No log 1.3962 74 0.8068 0.0296 0.8068 0.8982
No log 1.4340 76 0.7848 -0.1223 0.7848 0.8859
No log 1.4717 78 0.7812 -0.0069 0.7812 0.8839
No log 1.5094 80 0.7991 -0.1067 0.7991 0.8939
No log 1.5472 82 0.8125 -0.1067 0.8125 0.9014
No log 1.5849 84 0.8489 -0.0679 0.8489 0.9214
No log 1.6226 86 0.8764 -0.1244 0.8764 0.9362
No log 1.6604 88 0.8518 -0.0240 0.8518 0.9229
No log 1.6981 90 0.8021 -0.0520 0.8021 0.8956
No log 1.7358 92 0.8508 -0.0473 0.8508 0.9224
No log 1.7736 94 0.8413 -0.0644 0.8413 0.9172
No log 1.8113 96 0.8432 -0.0215 0.8432 0.9183
No log 1.8491 98 0.8409 -0.0739 0.8409 0.9170
No log 1.8868 100 0.8051 -0.0160 0.8051 0.8972
No log 1.9245 102 0.7942 -0.0662 0.7942 0.8912
No log 1.9623 104 0.7914 -0.0644 0.7914 0.8896
No log 2.0 106 0.7828 -0.0662 0.7828 0.8848
No log 2.0377 108 0.7706 -0.0571 0.7706 0.8778
No log 2.0755 110 0.7715 -0.1074 0.7715 0.8783
No log 2.1132 112 0.7850 -0.0939 0.7850 0.8860
No log 2.1509 114 0.7932 -0.0322 0.7932 0.8906
No log 2.1887 116 0.7789 0.0116 0.7789 0.8825
No log 2.2264 118 0.7764 0.0089 0.7764 0.8811
No log 2.2642 120 0.7527 0.0355 0.7527 0.8676
No log 2.3019 122 0.7315 0.0922 0.7315 0.8553
No log 2.3396 124 0.7686 0.0249 0.7686 0.8767
No log 2.3774 126 0.8749 0.0753 0.8749 0.9354
No log 2.4151 128 0.8891 0.0758 0.8891 0.9429
No log 2.4528 130 0.8407 -0.0334 0.8407 0.9169
No log 2.4906 132 0.9257 -0.0317 0.9257 0.9621
No log 2.5283 134 0.8595 0.0172 0.8595 0.9271
No log 2.5660 136 0.8423 -0.0465 0.8423 0.9178
No log 2.6038 138 0.8206 -0.1026 0.8206 0.9059
No log 2.6415 140 0.8319 0.0155 0.8319 0.9121
No log 2.6792 142 0.8455 -0.0322 0.8455 0.9195
No log 2.7170 144 0.8281 -0.0939 0.8281 0.9100
No log 2.7547 146 0.8104 0.0395 0.8104 0.9002
No log 2.7925 148 0.8345 0.0395 0.8345 0.9135
No log 2.8302 150 0.8689 -0.1474 0.8689 0.9321
No log 2.8679 152 0.9119 0.0377 0.9119 0.9549
No log 2.9057 154 0.8927 0.0770 0.8927 0.9448
No log 2.9434 156 0.8270 -0.0284 0.8270 0.9094
No log 2.9811 158 0.7470 0.0 0.7470 0.8643
No log 3.0189 160 0.8345 0.0043 0.8345 0.9135
No log 3.0566 162 0.8382 -0.0008 0.8382 0.9155
No log 3.0943 164 0.7493 -0.0160 0.7493 0.8656
No log 3.1321 166 0.7781 -0.1001 0.7781 0.8821
No log 3.1698 168 0.8101 0.0673 0.8101 0.9000
No log 3.2075 170 0.8416 0.1174 0.8416 0.9174
No log 3.2453 172 0.8152 -0.1263 0.8152 0.9029
No log 3.2830 174 0.7782 0.0334 0.7782 0.8821
No log 3.3208 176 0.7795 0.0334 0.7795 0.8829
No log 3.3585 178 0.7590 0.0555 0.7590 0.8712
No log 3.3962 180 0.8060 -0.1329 0.8060 0.8978
No log 3.4340 182 0.8317 -0.2071 0.8317 0.9120
No log 3.4717 184 0.7946 -0.0322 0.7946 0.8914
No log 3.5094 186 0.7701 -0.0550 0.7701 0.8775
No log 3.5472 188 0.7713 -0.0550 0.7713 0.8782
No log 3.5849 190 0.7743 -0.0027 0.7743 0.8799
No log 3.6226 192 0.7903 -0.0366 0.7903 0.8890
No log 3.6604 194 0.8782 -0.0843 0.8782 0.9371
No log 3.6981 196 0.9049 -0.0685 0.9049 0.9512
No log 3.7358 198 0.8133 -0.0591 0.8133 0.9018
No log 3.7736 200 0.8109 -0.0599 0.8109 0.9005
No log 3.8113 202 0.8331 -0.1180 0.8331 0.9128
No log 3.8491 204 0.7797 0.0116 0.7797 0.8830
No log 3.8868 206 0.8604 -0.1709 0.8604 0.9276
No log 3.9245 208 0.8884 -0.0761 0.8884 0.9426
No log 3.9623 210 0.8629 -0.0801 0.8629 0.9289
No log 4.0 212 0.8057 -0.0620 0.8057 0.8976
No log 4.0377 214 0.7798 0.0922 0.7798 0.8831
No log 4.0755 216 0.7716 0.1552 0.7716 0.8784
No log 4.1132 218 0.7870 -0.0345 0.7870 0.8871
No log 4.1509 220 0.7764 -0.0406 0.7764 0.8811
No log 4.1887 222 0.7059 -0.0035 0.7059 0.8402
No log 4.2264 224 0.6919 0.0964 0.6919 0.8318
No log 4.2642 226 0.6949 -0.0035 0.6949 0.8336
No log 4.3019 228 0.6987 0.1021 0.6987 0.8359
No log 4.3396 230 0.7406 0.0033 0.7406 0.8606
No log 4.3774 232 0.7771 -0.0363 0.7771 0.8815
No log 4.4151 234 0.7730 -0.1329 0.7730 0.8792
No log 4.4528 236 0.8755 -0.0346 0.8755 0.9357
No log 4.4906 238 0.8875 -0.0346 0.8875 0.9421
No log 4.5283 240 0.8136 -0.0144 0.8136 0.9020
No log 4.5660 242 0.8033 -0.0648 0.8033 0.8963
No log 4.6038 244 0.7838 -0.0837 0.7838 0.8853
No log 4.6415 246 0.7468 0.0587 0.7468 0.8642
No log 4.6792 248 0.7397 0.0587 0.7397 0.8600
No log 4.7170 250 0.7695 0.0869 0.7695 0.8772
No log 4.7547 252 0.8043 -0.0027 0.8043 0.8968
No log 4.7925 254 0.8648 0.0718 0.8648 0.9299
No log 4.8302 256 0.9163 0.0008 0.9163 0.9572
No log 4.8679 258 0.8698 -0.0806 0.8698 0.9326
No log 4.9057 260 0.9037 -0.0459 0.9037 0.9506
No log 4.9434 262 0.9185 -0.1009 0.9185 0.9584
No log 4.9811 264 1.0093 0.0147 1.0093 1.0046
No log 5.0189 266 0.9006 -0.0492 0.9006 0.9490
No log 5.0566 268 0.9108 -0.0588 0.9108 0.9544
No log 5.0943 270 0.8815 -0.1013 0.8815 0.9389
No log 5.1321 272 0.8886 0.0249 0.8886 0.9427
No log 5.1698 274 1.0624 -0.0060 1.0624 1.0307
No log 5.2075 276 1.0092 -0.0099 1.0092 1.0046
No log 5.2453 278 0.8522 -0.0762 0.8522 0.9231
No log 5.2830 280 0.8620 0.0236 0.8620 0.9284
No log 5.3208 282 0.8757 0.0123 0.8757 0.9358
No log 5.3585 284 0.7786 -0.0145 0.7786 0.8824
No log 5.3962 286 0.7955 -0.0363 0.7955 0.8919
No log 5.4340 288 0.8362 0.0206 0.8362 0.9144
No log 5.4717 290 0.8382 0.0206 0.8382 0.9155
No log 5.5094 292 0.8008 -0.0614 0.8008 0.8948
No log 5.5472 294 0.8130 -0.0228 0.8130 0.9017
No log 5.5849 296 0.7830 -0.0264 0.7830 0.8849
No log 5.6226 298 0.7512 0.0555 0.7512 0.8667
No log 5.6604 300 0.8683 -0.0262 0.8683 0.9318
No log 5.6981 302 0.8806 -0.0317 0.8806 0.9384
No log 5.7358 304 0.7855 0.0148 0.7855 0.8863
No log 5.7736 306 0.7510 0.1486 0.7510 0.8666
No log 5.8113 308 0.7451 0.1622 0.7451 0.8632
No log 5.8491 310 0.7535 0.1141 0.7535 0.8680
No log 5.8868 312 0.7665 0.0122 0.7665 0.8755
No log 5.9245 314 0.7887 -0.0738 0.7887 0.8881
No log 5.9623 316 0.8393 -0.0934 0.8393 0.9161
No log 6.0 318 0.8865 -0.0616 0.8865 0.9415
No log 6.0377 320 0.8620 -0.0040 0.8620 0.9285
No log 6.0755 322 0.7625 -0.0385 0.7625 0.8732
No log 6.1132 324 0.7758 0.1259 0.7758 0.8808
No log 6.1509 326 0.7725 0.1259 0.7725 0.8789
No log 6.1887 328 0.7111 0.1379 0.7111 0.8433
No log 6.2264 330 0.7378 -0.1010 0.7378 0.8590
No log 6.2642 332 0.7699 -0.0774 0.7699 0.8775
No log 6.3019 334 0.7665 -0.0363 0.7665 0.8755
No log 6.3396 336 0.7329 -0.0451 0.7329 0.8561
No log 6.3774 338 0.6967 0.0506 0.6967 0.8347
No log 6.4151 340 0.7152 0.1259 0.7152 0.8457
No log 6.4528 342 0.7439 0.1254 0.7439 0.8625
No log 6.4906 344 0.7454 0.0926 0.7454 0.8634
No log 6.5283 346 0.8331 0.0763 0.8331 0.9128
No log 6.5660 348 0.9102 0.0824 0.9102 0.9541
No log 6.6038 350 0.8584 0.1157 0.8584 0.9265
No log 6.6415 352 0.7883 -0.1146 0.7883 0.8879
No log 6.6792 354 0.7629 -0.0465 0.7629 0.8734
No log 6.7170 356 0.7761 -0.0798 0.7761 0.8809
No log 6.7547 358 0.7958 -0.1091 0.7958 0.8921
No log 6.7925 360 0.7468 0.0141 0.7468 0.8642
No log 6.8302 362 0.7347 0.0821 0.7347 0.8572
No log 6.8679 364 0.7234 0.0922 0.7234 0.8505
No log 6.9057 366 0.8253 0.0347 0.8253 0.9084
No log 6.9434 368 0.9840 0.1139 0.9840 0.9920
No log 6.9811 370 0.9675 0.0107 0.9675 0.9836
No log 7.0189 372 0.8808 0.0053 0.8808 0.9385
No log 7.0566 374 0.8542 0.0359 0.8542 0.9242
No log 7.0943 376 0.8271 0.0053 0.8271 0.9094
No log 7.1321 378 0.9461 -0.0722 0.9461 0.9727
No log 7.1698 380 1.0388 0.0837 1.0388 1.0192
No log 7.2075 382 0.9689 0.0481 0.9689 0.9843
No log 7.2453 384 0.8347 -0.1086 0.8347 0.9136
No log 7.2830 386 0.8055 -0.0056 0.8055 0.8975
No log 7.3208 388 0.8426 0.0709 0.8426 0.9180
No log 7.3585 390 0.8230 0.0 0.8230 0.9072
No log 7.3962 392 0.8381 -0.0786 0.8381 0.9155
No log 7.4340 394 0.8780 -0.0477 0.8780 0.9370
No log 7.4717 396 0.8286 -0.0633 0.8286 0.9103
No log 7.5094 398 0.7614 -0.0363 0.7614 0.8726
No log 7.5472 400 0.7377 -0.0096 0.7377 0.8589
No log 7.5849 402 0.7506 0.0863 0.7506 0.8663
No log 7.6226 404 0.7469 0.0030 0.7469 0.8642
No log 7.6604 406 0.8228 -0.0551 0.8228 0.9071
No log 7.6981 408 1.0422 0.1138 1.0422 1.0209
No log 7.7358 410 1.1366 0.0610 1.1366 1.0661
No log 7.7736 412 1.0510 0.0566 1.0510 1.0252
No log 7.8113 414 0.8844 0.0028 0.8844 0.9404
No log 7.8491 416 0.8392 -0.0408 0.8392 0.9161
No log 7.8868 418 0.8515 0.0757 0.8515 0.9227
No log 7.9245 420 0.8085 0.0053 0.8085 0.8992
No log 7.9623 422 0.8489 -0.1301 0.8489 0.9214
No log 8.0 424 0.9176 0.0180 0.9176 0.9579
No log 8.0377 426 0.9136 -0.0164 0.9136 0.9558
No log 8.0755 428 0.8489 -0.1709 0.8489 0.9214
No log 8.1132 430 0.7541 -0.1399 0.7541 0.8684
No log 8.1509 432 0.7156 -0.0473 0.7156 0.8459
No log 8.1887 434 0.7098 0.1902 0.7098 0.8425
No log 8.2264 436 0.7289 0.1259 0.7289 0.8538
No log 8.2642 438 0.7125 0.1318 0.7125 0.8441
No log 8.3019 440 0.7148 0.0 0.7148 0.8455
No log 8.3396 442 0.7574 -0.0939 0.7574 0.8703
No log 8.3774 444 0.7704 -0.0881 0.7704 0.8777
No log 8.4151 446 0.7661 0.0094 0.7661 0.8753
No log 8.4528 448 0.7781 0.0471 0.7781 0.8821
No log 8.4906 450 0.8025 0.0488 0.8025 0.8958
No log 8.5283 452 0.8373 0.0187 0.8373 0.9151
No log 8.5660 454 0.8637 -0.0458 0.8637 0.9293
No log 8.6038 456 0.8876 -0.1054 0.8876 0.9421
No log 8.6415 458 0.8321 -0.1413 0.8321 0.9122
No log 8.6792 460 0.7490 -0.1137 0.7490 0.8654
No log 8.7170 462 0.7126 0.0 0.7126 0.8441
No log 8.7547 464 0.7222 -0.1067 0.7222 0.8498
No log 8.7925 466 0.7777 -0.0322 0.7777 0.8819
No log 8.8302 468 0.8174 0.0934 0.8174 0.9041
No log 8.8679 470 0.8522 0.0934 0.8522 0.9232
No log 8.9057 472 0.8801 0.0570 0.8801 0.9381
No log 8.9434 474 0.8699 -0.0284 0.8699 0.9327
No log 8.9811 476 0.8206 0.0976 0.8206 0.9059
No log 9.0189 478 0.8027 0.0976 0.8027 0.8959
No log 9.0566 480 0.8078 0.1027 0.8078 0.8988
No log 9.0943 482 0.8061 -0.0406 0.8061 0.8978
No log 9.1321 484 0.7983 -0.0406 0.7983 0.8935
No log 9.1698 486 0.8121 -0.0406 0.8121 0.9011
No log 9.2075 488 0.8304 -0.0284 0.8304 0.9112
No log 9.2453 490 0.8469 0.0141 0.8469 0.9203
No log 9.2830 492 0.9100 -0.1247 0.9100 0.9539
No log 9.3208 494 0.9083 -0.0393 0.9083 0.9531
No log 9.3585 496 0.8394 0.0545 0.8394 0.9162
No log 9.3962 498 0.8137 0.0545 0.8137 0.9021
0.3721 9.4340 500 0.7904 0.0503 0.7904 0.8891
0.3721 9.4717 502 0.7796 0.0528 0.7796 0.8830
0.3721 9.5094 504 0.8003 0.0240 0.8003 0.8946
0.3721 9.5472 506 0.7841 0.0173 0.7841 0.8855
0.3721 9.5849 508 0.7675 0.0488 0.7675 0.8761
0.3721 9.6226 510 0.7606 0.0503 0.7606 0.8721
0.3721 9.6604 512 0.7808 0.0683 0.7808 0.8836
0.3721 9.6981 514 0.8355 -0.1413 0.8355 0.9141
0.3721 9.7358 516 0.8157 -0.0551 0.8157 0.9032
0.3721 9.7736 518 0.7461 0.0122 0.7461 0.8638
0.3721 9.8113 520 0.6827 0.0922 0.6827 0.8263
0.3721 9.8491 522 0.6832 0.0513 0.6832 0.8266
0.3721 9.8868 524 0.7314 -0.0451 0.7314 0.8552
0.3721 9.9245 526 0.7343 -0.0451 0.7343 0.8569
0.3721 9.9623 528 0.7069 0.0031 0.7069 0.8408
0.3721 10.0 530 0.7132 0.0471 0.7132 0.8445
0.3721 10.0377 532 0.7574 0.0545 0.7574 0.8703
0.3721 10.0755 534 0.8193 0.0682 0.8193 0.9052
0.3721 10.1132 536 0.9152 0.0052 0.9152 0.9566
0.3721 10.1509 538 0.8943 0.0052 0.8943 0.9457
0.3721 10.1887 540 0.8001 0.0944 0.8001 0.8945
0.3721 10.2264 542 0.8500 -0.0291 0.8500 0.9220
0.3721 10.2642 544 0.8808 0.0876 0.8808 0.9385
0.3721 10.3019 546 0.8005 0.0257 0.8005 0.8947
0.3721 10.3396 548 0.7928 0.0545 0.7928 0.8904
0.3721 10.3774 550 0.7996 0.0157 0.7996 0.8942
0.3721 10.4151 552 0.7743 0.0532 0.7743 0.8800
0.3721 10.4528 554 0.7577 0.0926 0.7577 0.8705
0.3721 10.4906 556 0.7803 0.0481 0.7803 0.8834
0.3721 10.5283 558 0.7899 -0.0750 0.7899 0.8887
0.3721 10.5660 560 0.7754 -0.0750 0.7754 0.8806
0.3721 10.6038 562 0.7616 0.0967 0.7616 0.8727
0.3721 10.6415 564 0.7762 -0.0152 0.7762 0.8810
0.3721 10.6792 566 0.8304 -0.0549 0.8304 0.9112
0.3721 10.7170 568 0.8269 -0.0462 0.8269 0.9094
0.3721 10.7547 570 0.8237 0.0179 0.8237 0.9076
0.3721 10.7925 572 0.7868 0.0116 0.7868 0.8870

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_k20_task3_organization

Finetuned
(2729)
this model