ArabicNewSplits8_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k12_task1_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: 0.8421
  • Qwk: 0.6112
  • Mse: 0.8421
  • Rmse: 0.9177

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 0.0333 2 5.4206 -0.0309 5.4206 2.3282
No log 0.0667 4 3.0732 0.0696 3.0732 1.7530
No log 0.1 6 2.3083 -0.1419 2.3083 1.5193
No log 0.1333 8 1.6530 0.0802 1.6530 1.2857
No log 0.1667 10 1.3717 0.0954 1.3717 1.1712
No log 0.2 12 1.4686 -0.0123 1.4686 1.2119
No log 0.2333 14 1.5602 0.0756 1.5602 1.2491
No log 0.2667 16 1.7732 0.0279 1.7732 1.3316
No log 0.3 18 1.8151 0.0468 1.8151 1.3473
No log 0.3333 20 1.4932 0.0524 1.4932 1.2220
No log 0.3667 22 1.4233 0.0932 1.4233 1.1930
No log 0.4 24 1.3719 0.1198 1.3719 1.1713
No log 0.4333 26 1.3419 0.1803 1.3419 1.1584
No log 0.4667 28 1.2588 0.2773 1.2588 1.1220
No log 0.5 30 1.2431 0.3535 1.2431 1.1150
No log 0.5333 32 1.2497 0.3135 1.2497 1.1179
No log 0.5667 34 1.2994 0.2289 1.2994 1.1399
No log 0.6 36 1.3301 0.1655 1.3301 1.1533
No log 0.6333 38 1.2442 0.2386 1.2442 1.1154
No log 0.6667 40 1.1894 0.3051 1.1894 1.0906
No log 0.7 42 1.1969 0.1322 1.1969 1.0940
No log 0.7333 44 1.1730 0.1944 1.1730 1.0831
No log 0.7667 46 1.1679 0.3482 1.1679 1.0807
No log 0.8 48 1.3012 0.1336 1.3012 1.1407
No log 0.8333 50 1.3225 0.1361 1.3225 1.1500
No log 0.8667 52 1.4026 0.1536 1.4026 1.1843
No log 0.9 54 1.3569 0.1794 1.3569 1.1649
No log 0.9333 56 1.1547 0.2835 1.1547 1.0746
No log 0.9667 58 1.1480 0.1771 1.1480 1.0715
No log 1.0 60 1.1873 0.1223 1.1873 1.0896
No log 1.0333 62 1.3114 0.2166 1.3114 1.1452
No log 1.0667 64 1.2995 0.2542 1.2995 1.1399
No log 1.1 66 1.1967 0.1460 1.1967 1.0940
No log 1.1333 68 1.1683 0.1334 1.1683 1.0809
No log 1.1667 70 1.1647 0.0787 1.1647 1.0792
No log 1.2 72 1.1131 0.1822 1.1131 1.0550
No log 1.2333 74 1.1118 0.3643 1.1118 1.0544
No log 1.2667 76 1.0785 0.3745 1.0785 1.0385
No log 1.3 78 1.0838 0.2854 1.0838 1.0410
No log 1.3333 80 1.1795 0.2216 1.1795 1.0860
No log 1.3667 82 1.1195 0.2098 1.1195 1.0581
No log 1.4 84 0.9949 0.2969 0.9949 0.9974
No log 1.4333 86 0.9334 0.3823 0.9334 0.9661
No log 1.4667 88 0.9166 0.3800 0.9166 0.9574
No log 1.5 90 0.9121 0.3781 0.9121 0.9550
No log 1.5333 92 0.9504 0.3648 0.9504 0.9749
No log 1.5667 94 0.9680 0.3643 0.9680 0.9839
No log 1.6 96 0.9260 0.3507 0.9260 0.9623
No log 1.6333 98 0.8708 0.3620 0.8708 0.9332
No log 1.6667 100 0.8368 0.5146 0.8368 0.9148
No log 1.7 102 0.8490 0.5179 0.8490 0.9214
No log 1.7333 104 0.9220 0.5351 0.9220 0.9602
No log 1.7667 106 1.1769 0.4902 1.1769 1.0848
No log 1.8 108 1.3392 0.3499 1.3392 1.1572
No log 1.8333 110 1.3459 0.3582 1.3459 1.1601
No log 1.8667 112 1.2639 0.3890 1.2639 1.1242
No log 1.9 114 1.0078 0.5165 1.0078 1.0039
No log 1.9333 116 0.9172 0.4955 0.9172 0.9577
No log 1.9667 118 0.9968 0.5489 0.9968 0.9984
No log 2.0 120 1.3805 0.3592 1.3805 1.1750
No log 2.0333 122 1.7242 0.3284 1.7242 1.3131
No log 2.0667 124 1.4835 0.3269 1.4835 1.2180
No log 2.1 126 1.1816 0.2868 1.1816 1.0870
No log 2.1333 128 1.0766 0.3639 1.0766 1.0376
No log 2.1667 130 1.0187 0.4932 1.0187 1.0093
No log 2.2 132 1.0126 0.5544 1.0126 1.0063
No log 2.2333 134 1.0652 0.5473 1.0652 1.0321
No log 2.2667 136 1.0708 0.5232 1.0708 1.0348
No log 2.3 138 0.9536 0.4986 0.9536 0.9765
No log 2.3333 140 0.9227 0.5192 0.9227 0.9606
No log 2.3667 142 1.0213 0.5981 1.0213 1.0106
No log 2.4 144 1.3437 0.4317 1.3437 1.1592
No log 2.4333 146 1.4838 0.3601 1.4838 1.2181
No log 2.4667 148 1.4558 0.3834 1.4558 1.2066
No log 2.5 150 1.3063 0.4379 1.3063 1.1429
No log 2.5333 152 1.0724 0.4794 1.0724 1.0356
No log 2.5667 154 1.0855 0.4968 1.0855 1.0419
No log 2.6 156 1.2751 0.4166 1.2751 1.1292
No log 2.6333 158 1.5146 0.3409 1.5146 1.2307
No log 2.6667 160 1.3629 0.3364 1.3629 1.1674
No log 2.7 162 1.0351 0.4809 1.0351 1.0174
No log 2.7333 164 0.9430 0.5261 0.9430 0.9711
No log 2.7667 166 1.0407 0.4756 1.0407 1.0202
No log 2.8 168 1.1493 0.4782 1.1493 1.0720
No log 2.8333 170 1.4511 0.3867 1.4511 1.2046
No log 2.8667 172 1.4887 0.3786 1.4887 1.2201
No log 2.9 174 1.2131 0.4424 1.2131 1.1014
No log 2.9333 176 0.9996 0.5730 0.9996 0.9998
No log 2.9667 178 1.0164 0.5681 1.0164 1.0082
No log 3.0 180 1.2498 0.4197 1.2498 1.1179
No log 3.0333 182 1.5239 0.3433 1.5239 1.2345
No log 3.0667 184 1.4198 0.3884 1.4198 1.1916
No log 3.1 186 1.1511 0.5287 1.1511 1.0729
No log 3.1333 188 0.9329 0.5873 0.9329 0.9659
No log 3.1667 190 0.9559 0.5950 0.9559 0.9777
No log 3.2 192 1.1101 0.5695 1.1101 1.0536
No log 3.2333 194 1.2932 0.4704 1.2932 1.1372
No log 3.2667 196 1.1947 0.5086 1.1947 1.0930
No log 3.3 198 0.9310 0.5833 0.9310 0.9649
No log 3.3333 200 0.7757 0.5926 0.7757 0.8808
No log 3.3667 202 0.8036 0.6080 0.8036 0.8964
No log 3.4 204 0.9857 0.6119 0.9857 0.9928
No log 3.4333 206 1.2832 0.4990 1.2832 1.1328
No log 3.4667 208 1.3832 0.4267 1.3832 1.1761
No log 3.5 210 1.1748 0.5305 1.1748 1.0839
No log 3.5333 212 0.9092 0.5982 0.9092 0.9535
No log 3.5667 214 0.7854 0.6261 0.7854 0.8862
No log 3.6 216 0.7280 0.6280 0.7280 0.8532
No log 3.6333 218 0.7599 0.6163 0.7599 0.8717
No log 3.6667 220 0.7681 0.6129 0.7681 0.8764
No log 3.7 222 0.7346 0.6215 0.7346 0.8571
No log 3.7333 224 0.7604 0.6108 0.7604 0.8720
No log 3.7667 226 0.7860 0.6234 0.7860 0.8866
No log 3.8 228 0.9375 0.6279 0.9375 0.9682
No log 3.8333 230 1.0911 0.5595 1.0911 1.0446
No log 3.8667 232 1.0867 0.5313 1.0867 1.0424
No log 3.9 234 1.0683 0.5372 1.0683 1.0336
No log 3.9333 236 0.9269 0.5985 0.9269 0.9627
No log 3.9667 238 0.7910 0.6328 0.7910 0.8894
No log 4.0 240 0.8488 0.6117 0.8488 0.9213
No log 4.0333 242 1.0313 0.5242 1.0313 1.0155
No log 4.0667 244 1.1834 0.4762 1.1834 1.0879
No log 4.1 246 1.1739 0.4687 1.1739 1.0835
No log 4.1333 248 1.2305 0.4546 1.2305 1.1093
No log 4.1667 250 1.1573 0.4766 1.1573 1.0758
No log 4.2 252 1.0161 0.5568 1.0161 1.0080
No log 4.2333 254 1.0273 0.5571 1.0273 1.0135
No log 4.2667 256 1.0871 0.5308 1.0871 1.0427
No log 4.3 258 1.0777 0.5444 1.0777 1.0381
No log 4.3333 260 1.0815 0.5572 1.0815 1.0400
No log 4.3667 262 1.1779 0.5454 1.1779 1.0853
No log 4.4 264 1.0771 0.5351 1.0771 1.0378
No log 4.4333 266 0.9342 0.5679 0.9342 0.9666
No log 4.4667 268 1.0237 0.5481 1.0237 1.0118
No log 4.5 270 1.1131 0.5218 1.1131 1.0551
No log 4.5333 272 1.0421 0.5052 1.0421 1.0208
No log 4.5667 274 0.8425 0.6367 0.8425 0.9179
No log 4.6 276 0.6759 0.6727 0.6759 0.8221
No log 4.6333 278 0.6648 0.6694 0.6648 0.8154
No log 4.6667 280 0.7601 0.6707 0.7601 0.8718
No log 4.7 282 1.1026 0.5687 1.1026 1.0500
No log 4.7333 284 1.3906 0.4982 1.3906 1.1792
No log 4.7667 286 1.3067 0.5280 1.3067 1.1431
No log 4.8 288 0.9763 0.5991 0.9763 0.9881
No log 4.8333 290 0.7871 0.6466 0.7871 0.8872
No log 4.8667 292 0.7166 0.6580 0.7166 0.8465
No log 4.9 294 0.7141 0.6580 0.7141 0.8451
No log 4.9333 296 0.7542 0.6485 0.7542 0.8685
No log 4.9667 298 0.9079 0.5903 0.9079 0.9528
No log 5.0 300 1.0334 0.5637 1.0334 1.0166
No log 5.0333 302 0.9846 0.5767 0.9846 0.9923
No log 5.0667 304 0.8355 0.6285 0.8355 0.9141
No log 5.1 306 0.7376 0.6051 0.7376 0.8588
No log 5.1333 308 0.7125 0.6714 0.7125 0.8441
No log 5.1667 310 0.7414 0.6675 0.7414 0.8610
No log 5.2 312 0.8174 0.6422 0.8174 0.9041
No log 5.2333 314 0.9029 0.6436 0.9029 0.9502
No log 5.2667 316 0.9812 0.5911 0.9812 0.9906
No log 5.3 318 0.8901 0.6399 0.8901 0.9434
No log 5.3333 320 0.8213 0.6291 0.8213 0.9063
No log 5.3667 322 0.7058 0.6696 0.7058 0.8401
No log 5.4 324 0.7438 0.6636 0.7438 0.8624
No log 5.4333 326 0.8167 0.6508 0.8167 0.9037
No log 5.4667 328 0.9608 0.6171 0.9608 0.9802
No log 5.5 330 0.9591 0.6051 0.9591 0.9793
No log 5.5333 332 0.8478 0.6268 0.8478 0.9208
No log 5.5667 334 0.7205 0.6614 0.7205 0.8488
No log 5.6 336 0.7547 0.6634 0.7547 0.8687
No log 5.6333 338 0.9220 0.5681 0.9220 0.9602
No log 5.6667 340 1.0088 0.5337 1.0088 1.0044
No log 5.7 342 0.8923 0.6066 0.8923 0.9446
No log 5.7333 344 0.8110 0.6321 0.8110 0.9005
No log 5.7667 346 0.7806 0.6376 0.7806 0.8835
No log 5.8 348 0.8641 0.6372 0.8641 0.9296
No log 5.8333 350 1.0505 0.5334 1.0505 1.0250
No log 5.8667 352 1.0715 0.5320 1.0715 1.0351
No log 5.9 354 0.9185 0.5767 0.9185 0.9584
No log 5.9333 356 0.8203 0.6322 0.8203 0.9057
No log 5.9667 358 0.8554 0.6292 0.8554 0.9249
No log 6.0 360 1.0208 0.5143 1.0208 1.0103
No log 6.0333 362 1.2432 0.5210 1.2432 1.1150
No log 6.0667 364 1.2019 0.5300 1.2019 1.0963
No log 6.1 366 0.9497 0.5780 0.9497 0.9745
No log 6.1333 368 0.7521 0.6449 0.7521 0.8673
No log 6.1667 370 0.7246 0.6491 0.7246 0.8512
No log 6.2 372 0.7926 0.6411 0.7926 0.8903
No log 6.2333 374 0.9614 0.5590 0.9614 0.9805
No log 6.2667 376 0.9796 0.5414 0.9796 0.9898
No log 6.3 378 0.8477 0.6143 0.8477 0.9207
No log 6.3333 380 0.7378 0.6278 0.7378 0.8590
No log 6.3667 382 0.7379 0.6291 0.7379 0.8590
No log 6.4 384 0.8279 0.6419 0.8279 0.9099
No log 6.4333 386 0.9796 0.5416 0.9796 0.9897
No log 6.4667 388 1.1012 0.5196 1.1012 1.0494
No log 6.5 390 1.0693 0.5305 1.0693 1.0341
No log 6.5333 392 0.9143 0.5816 0.9143 0.9562
No log 6.5667 394 0.7540 0.6266 0.7540 0.8683
No log 6.6 396 0.7221 0.6394 0.7221 0.8498
No log 6.6333 398 0.7683 0.6624 0.7683 0.8765
No log 6.6667 400 0.9189 0.5655 0.9189 0.9586
No log 6.7 402 1.1753 0.5395 1.1753 1.0841
No log 6.7333 404 1.2228 0.5304 1.2228 1.1058
No log 6.7667 406 1.0810 0.5764 1.0810 1.0397
No log 6.8 408 0.8532 0.6388 0.8532 0.9237
No log 6.8333 410 0.6884 0.6557 0.6884 0.8297
No log 6.8667 412 0.6662 0.6876 0.6662 0.8162
No log 6.9 414 0.7218 0.6676 0.7218 0.8496
No log 6.9333 416 0.8838 0.6461 0.8838 0.9401
No log 6.9667 418 0.9793 0.6152 0.9793 0.9896
No log 7.0 420 0.9476 0.6184 0.9476 0.9735
No log 7.0333 422 0.8192 0.6241 0.8192 0.9051
No log 7.0667 424 0.7529 0.6251 0.7529 0.8677
No log 7.1 426 0.7292 0.6279 0.7292 0.8539
No log 7.1333 428 0.7413 0.6346 0.7413 0.8610
No log 7.1667 430 0.7661 0.6262 0.7661 0.8753
No log 7.2 432 0.8090 0.6373 0.8090 0.8994
No log 7.2333 434 0.8173 0.6334 0.8173 0.9041
No log 7.2667 436 0.8711 0.6126 0.8711 0.9333
No log 7.3 438 0.8487 0.6175 0.8487 0.9213
No log 7.3333 440 0.8737 0.5755 0.8737 0.9347
No log 7.3667 442 0.8260 0.6053 0.8260 0.9089
No log 7.4 444 0.8157 0.6321 0.8157 0.9032
No log 7.4333 446 0.8163 0.5899 0.8163 0.9035
No log 7.4667 448 0.8916 0.5717 0.8916 0.9443
No log 7.5 450 1.1023 0.5445 1.1023 1.0499
No log 7.5333 452 1.2670 0.5167 1.2670 1.1256
No log 7.5667 454 1.2438 0.5141 1.2438 1.1153
No log 7.6 456 1.0884 0.5467 1.0884 1.0433
No log 7.6333 458 0.9397 0.5791 0.9397 0.9694
No log 7.6667 460 0.9355 0.5792 0.9355 0.9672
No log 7.7 462 1.0376 0.5407 1.0376 1.0186
No log 7.7333 464 1.2930 0.4821 1.2930 1.1371
No log 7.7667 466 1.4321 0.4474 1.4321 1.1967
No log 7.8 468 1.3260 0.4769 1.3260 1.1515
No log 7.8333 470 1.0475 0.5388 1.0475 1.0235
No log 7.8667 472 0.8318 0.6224 0.8318 0.9120
No log 7.9 474 0.7833 0.6389 0.7833 0.8851
No log 7.9333 476 0.8482 0.6214 0.8482 0.9210
No log 7.9667 478 0.9486 0.5608 0.9486 0.9740
No log 8.0 480 0.9623 0.5570 0.9623 0.9810
No log 8.0333 482 0.8899 0.6324 0.8899 0.9433
No log 8.0667 484 0.8370 0.6368 0.8370 0.9149
No log 8.1 486 0.8970 0.6149 0.8970 0.9471
No log 8.1333 488 0.9924 0.5779 0.9924 0.9962
No log 8.1667 490 1.0544 0.5342 1.0544 1.0269
No log 8.2 492 1.0303 0.5454 1.0303 1.0150
No log 8.2333 494 0.9302 0.6201 0.9302 0.9645
No log 8.2667 496 0.7964 0.6465 0.7964 0.8924
No log 8.3 498 0.7236 0.6411 0.7236 0.8507
0.519 8.3333 500 0.7858 0.6337 0.7858 0.8864
0.519 8.3667 502 0.9571 0.5954 0.9571 0.9783
0.519 8.4 504 1.0149 0.6095 1.0149 1.0074
0.519 8.4333 506 0.9268 0.6351 0.9268 0.9627
0.519 8.4667 508 0.7975 0.6268 0.7975 0.8930
0.519 8.5 510 0.7521 0.6355 0.7521 0.8672
0.519 8.5333 512 0.7155 0.6348 0.7155 0.8459
0.519 8.5667 514 0.7380 0.6296 0.7380 0.8591
0.519 8.6 516 0.7242 0.6337 0.7242 0.8510
0.519 8.6333 518 0.7061 0.6425 0.7061 0.8403
0.519 8.6667 520 0.6864 0.6665 0.6864 0.8285
0.519 8.7 522 0.7313 0.6475 0.7313 0.8552
0.519 8.7333 524 0.8127 0.6319 0.8127 0.9015
0.519 8.7667 526 0.8495 0.6432 0.8495 0.9217
0.519 8.8 528 0.7418 0.6973 0.7418 0.8613
0.519 8.8333 530 0.6696 0.6931 0.6696 0.8183
0.519 8.8667 532 0.6897 0.6684 0.6897 0.8305
0.519 8.9 534 0.8456 0.6278 0.8456 0.9196
0.519 8.9333 536 1.0504 0.5879 1.0504 1.0249
0.519 8.9667 538 1.0502 0.5834 1.0502 1.0248
0.519 9.0 540 0.9101 0.6142 0.9101 0.9540
0.519 9.0333 542 0.8067 0.6390 0.8067 0.8982
0.519 9.0667 544 0.8431 0.6339 0.8431 0.9182
0.519 9.1 546 0.9106 0.6237 0.9106 0.9542
0.519 9.1333 548 1.0217 0.6124 1.0217 1.0108
0.519 9.1667 550 0.9338 0.6094 0.9338 0.9664
0.519 9.2 552 0.7774 0.6267 0.7774 0.8817
0.519 9.2333 554 0.7045 0.6532 0.7045 0.8394
0.519 9.2667 556 0.7227 0.6532 0.7227 0.8501
0.519 9.3 558 0.8449 0.6196 0.8449 0.9192
0.519 9.3333 560 0.9978 0.5743 0.9978 0.9989
0.519 9.3667 562 1.1264 0.5687 1.1264 1.0613
0.519 9.4 564 1.0228 0.5635 1.0228 1.0113
0.519 9.4333 566 0.8421 0.6112 0.8421 0.9177

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0+cu118
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
6
Safetensors
Model size
135M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported third-party Inference Providers, and the model is not deployed on the HF Inference API.

Model tree for MayBashendy/ArabicNewSplits8_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k12_task1_organization

Finetuned
(4222)
this model