ArabicNewSplits7_usingALLEssays_FineTuningAraBERT_run2_AugV5_k7_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: 0.8414
  • Qwk: 0.6906
  • Mse: 0.8414
  • Rmse: 0.9173

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 0.0625 2 6.8119 0.0303 6.8119 2.6100
No log 0.125 4 4.3192 0.0675 4.3192 2.0783
No log 0.1875 6 3.3926 0.0357 3.3926 1.8419
No log 0.25 8 3.1998 -0.0119 3.1998 1.7888
No log 0.3125 10 2.3061 0.1449 2.3061 1.5186
No log 0.375 12 2.3599 -0.0426 2.3599 1.5362
No log 0.4375 14 2.3636 -0.0438 2.3636 1.5374
No log 0.5 16 1.9265 0.2655 1.9265 1.3880
No log 0.5625 18 1.7729 0.2342 1.7729 1.3315
No log 0.625 20 1.8019 0.2281 1.8019 1.3424
No log 0.6875 22 2.2640 0.1029 2.2640 1.5046
No log 0.75 24 2.7416 0.0414 2.7416 1.6558
No log 0.8125 26 2.3351 0.1295 2.3351 1.5281
No log 0.875 28 1.8244 0.2615 1.8244 1.3507
No log 0.9375 30 1.6211 0.25 1.6211 1.2732
No log 1.0 32 1.5094 0.3478 1.5094 1.2286
No log 1.0625 34 1.4188 0.3509 1.4188 1.1911
No log 1.125 36 1.4078 0.3826 1.4078 1.1865
No log 1.1875 38 1.4970 0.3967 1.4970 1.2235
No log 1.25 40 1.6682 0.3636 1.6682 1.2916
No log 1.3125 42 1.8614 0.3008 1.8614 1.3643
No log 1.375 44 1.9723 0.2727 1.9723 1.4044
No log 1.4375 46 1.9390 0.2727 1.9390 1.3925
No log 1.5 48 1.8694 0.3407 1.8694 1.3673
No log 1.5625 50 1.9611 0.2815 1.9611 1.4004
No log 1.625 52 2.0558 0.2778 2.0558 1.4338
No log 1.6875 54 2.2828 0.2763 2.2828 1.5109
No log 1.75 56 2.3024 0.2987 2.3024 1.5174
No log 1.8125 58 1.8635 0.3916 1.8635 1.3651
No log 1.875 60 1.4586 0.4242 1.4586 1.2077
No log 1.9375 62 1.3650 0.4462 1.3650 1.1683
No log 2.0 64 1.3294 0.4615 1.3294 1.1530
No log 2.0625 66 1.4152 0.4662 1.4152 1.1896
No log 2.125 68 1.5666 0.4143 1.5666 1.2516
No log 2.1875 70 1.6757 0.3553 1.6757 1.2945
No log 2.25 72 1.7535 0.4286 1.7535 1.3242
No log 2.3125 74 1.9179 0.3694 1.9179 1.3849
No log 2.375 76 2.4670 0.2485 2.4670 1.5707
No log 2.4375 78 2.8799 0.1297 2.8799 1.6970
No log 2.5 80 2.3407 0.2706 2.3407 1.5299
No log 2.5625 82 2.1810 0.3765 2.1810 1.4768
No log 2.625 84 1.9895 0.3484 1.9895 1.4105
No log 2.6875 86 1.7837 0.3472 1.7837 1.3356
No log 2.75 88 1.5906 0.3817 1.5906 1.2612
No log 2.8125 90 1.6110 0.3182 1.6110 1.2693
No log 2.875 92 1.7612 0.2879 1.7612 1.3271
No log 2.9375 94 1.8913 0.2443 1.8913 1.3752
No log 3.0 96 1.7796 0.2879 1.7796 1.3340
No log 3.0625 98 1.7818 0.2879 1.7818 1.3348
No log 3.125 100 1.4776 0.4154 1.4776 1.2156
No log 3.1875 102 1.2273 0.5649 1.2273 1.1078
No log 3.25 104 1.1325 0.5496 1.1325 1.0642
No log 3.3125 106 1.1523 0.5882 1.1523 1.0735
No log 3.375 108 1.2154 0.5612 1.2154 1.1025
No log 3.4375 110 1.3320 0.5036 1.3320 1.1541
No log 3.5 112 1.7684 0.3521 1.7684 1.3298
No log 3.5625 114 1.9869 0.2759 1.9869 1.4096
No log 3.625 116 1.6999 0.4286 1.6999 1.3038
No log 3.6875 118 1.3108 0.4932 1.3108 1.1449
No log 3.75 120 1.2221 0.5139 1.2221 1.1055
No log 3.8125 122 1.2171 0.5503 1.2171 1.1032
No log 3.875 124 1.2230 0.5733 1.2230 1.1059
No log 3.9375 126 1.4969 0.5270 1.4969 1.2235
No log 4.0 128 2.1568 0.2994 2.1568 1.4686
No log 4.0625 130 2.6435 0.2570 2.6435 1.6259
No log 4.125 132 2.3102 0.2959 2.3102 1.5199
No log 4.1875 134 1.4902 0.4861 1.4902 1.2207
No log 4.25 136 1.0394 0.5926 1.0394 1.0195
No log 4.3125 138 0.9844 0.5970 0.9844 0.9922
No log 4.375 140 1.0273 0.5303 1.0273 1.0135
No log 4.4375 142 0.9915 0.6029 0.9915 0.9957
No log 4.5 144 0.9406 0.6993 0.9406 0.9698
No log 4.5625 146 1.0241 0.6531 1.0241 1.0120
No log 4.625 148 1.1958 0.5638 1.1958 1.0935
No log 4.6875 150 1.2165 0.5658 1.2165 1.1030
No log 4.75 152 1.0938 0.6536 1.0938 1.0459
No log 4.8125 154 1.0377 0.6358 1.0377 1.0187
No log 4.875 156 1.0834 0.6358 1.0834 1.0409
No log 4.9375 158 1.3259 0.5490 1.3259 1.1515
No log 5.0 160 1.6752 0.5062 1.6752 1.2943
No log 5.0625 162 2.1241 0.4211 2.1241 1.4574
No log 5.125 164 2.1051 0.4211 2.1051 1.4509
No log 5.1875 166 1.6350 0.5030 1.6350 1.2787
No log 5.25 168 1.1739 0.5714 1.1739 1.0835
No log 5.3125 170 0.9832 0.5970 0.9832 0.9916
No log 5.375 172 0.9679 0.6212 0.9679 0.9838
No log 5.4375 174 0.9750 0.6466 0.9750 0.9874
No log 5.5 176 0.9874 0.5984 0.9874 0.9937
No log 5.5625 178 1.0157 0.5669 1.0157 1.0078
No log 5.625 180 1.0523 0.5238 1.0523 1.0258
No log 5.6875 182 1.0738 0.5496 1.0738 1.0362
No log 5.75 184 1.1300 0.5693 1.1300 1.0630
No log 5.8125 186 1.0999 0.6040 1.0999 1.0488
No log 5.875 188 1.1425 0.6104 1.1425 1.0689
No log 5.9375 190 1.3552 0.6012 1.3552 1.1641
No log 6.0 192 1.5209 0.5269 1.5209 1.2332
No log 6.0625 194 1.4864 0.5122 1.4864 1.2192
No log 6.125 196 1.3978 0.5786 1.3978 1.1823
No log 6.1875 198 1.2578 0.5987 1.2578 1.1215
No log 6.25 200 1.2107 0.5987 1.2107 1.1003
No log 6.3125 202 1.2956 0.5987 1.2956 1.1383
No log 6.375 204 1.2957 0.5875 1.2957 1.1383
No log 6.4375 206 1.3035 0.575 1.3035 1.1417
No log 6.5 208 1.2101 0.6076 1.2101 1.1000
No log 6.5625 210 1.0974 0.6364 1.0974 1.0476
No log 6.625 212 1.0392 0.6753 1.0392 1.0194
No log 6.6875 214 0.9717 0.6622 0.9717 0.9858
No log 6.75 216 0.9214 0.6763 0.9214 0.9599
No log 6.8125 218 0.8753 0.7153 0.8753 0.9356
No log 6.875 220 0.8883 0.6569 0.8883 0.9425
No log 6.9375 222 0.9290 0.6269 0.9290 0.9639
No log 7.0 224 0.9430 0.6412 0.9430 0.9711
No log 7.0625 226 0.9185 0.6515 0.9185 0.9584
No log 7.125 228 0.8725 0.7206 0.8725 0.9341
No log 7.1875 230 0.8755 0.6912 0.8755 0.9357
No log 7.25 232 0.9627 0.6277 0.9627 0.9812
No log 7.3125 234 1.0071 0.6232 1.0071 1.0035
No log 7.375 236 0.9570 0.6232 0.9570 0.9783
No log 7.4375 238 0.8978 0.6618 0.8978 0.9475
No log 7.5 240 0.8936 0.6861 0.8936 0.9453
No log 7.5625 242 0.8976 0.6815 0.8976 0.9474
No log 7.625 244 0.9597 0.6383 0.9597 0.9796
No log 7.6875 246 1.0940 0.5753 1.0940 1.0459
No log 7.75 248 1.2170 0.5753 1.2170 1.1032
No log 7.8125 250 1.2775 0.6 1.2775 1.1302
No log 7.875 252 1.2414 0.5946 1.2414 1.1142
No log 7.9375 254 1.1897 0.5793 1.1897 1.0907
No log 8.0 256 1.1875 0.5793 1.1875 1.0897
No log 8.0625 258 1.1962 0.5850 1.1962 1.0937
No log 8.125 260 1.1661 0.5793 1.1661 1.0798
No log 8.1875 262 1.1458 0.5594 1.1458 1.0704
No log 8.25 264 1.2283 0.5429 1.2283 1.1083
No log 8.3125 266 1.4303 0.4539 1.4303 1.1960
No log 8.375 268 1.6669 0.3514 1.6669 1.2911
No log 8.4375 270 1.6338 0.4774 1.6338 1.2782
No log 8.5 272 1.4681 0.5033 1.4681 1.2117
No log 8.5625 274 1.3375 0.5467 1.3375 1.1565
No log 8.625 276 1.1405 0.5772 1.1405 1.0679
No log 8.6875 278 0.9507 0.6569 0.9507 0.9751
No log 8.75 280 0.9202 0.6277 0.9202 0.9593
No log 8.8125 282 0.9466 0.6569 0.9466 0.9729
No log 8.875 284 0.9600 0.6316 0.9600 0.9798
No log 8.9375 286 0.9930 0.5692 0.9930 0.9965
No log 9.0 288 1.1073 0.5191 1.1073 1.0523
No log 9.0625 290 1.2286 0.5113 1.2286 1.1084
No log 9.125 292 1.2475 0.5113 1.2475 1.1169
No log 9.1875 294 1.1706 0.5 1.1706 1.0820
No log 9.25 296 1.1035 0.5373 1.1035 1.0505
No log 9.3125 298 1.0173 0.5891 1.0173 1.0086
No log 9.375 300 0.9532 0.6316 0.9532 0.9763
No log 9.4375 302 0.9527 0.6316 0.9527 0.9761
No log 9.5 304 1.0194 0.6222 1.0194 1.0096
No log 9.5625 306 1.1628 0.5564 1.1628 1.0783
No log 9.625 308 1.4936 0.4755 1.4936 1.2221
No log 9.6875 310 1.6713 0.4557 1.6713 1.2928
No log 9.75 312 1.7787 0.4304 1.7787 1.3337
No log 9.8125 314 1.6656 0.4459 1.6656 1.2906
No log 9.875 316 1.5135 0.4626 1.5135 1.2303
No log 9.9375 318 1.3769 0.5106 1.3769 1.1734
No log 10.0 320 1.3942 0.4823 1.3942 1.1807
No log 10.0625 322 1.3938 0.4604 1.3938 1.1806
No log 10.125 324 1.2423 0.5286 1.2423 1.1146
No log 10.1875 326 1.1290 0.5441 1.1290 1.0626
No log 10.25 328 1.1217 0.5152 1.1217 1.0591
No log 10.3125 330 1.1440 0.5263 1.1440 1.0696
No log 10.375 332 1.2454 0.4925 1.2454 1.1160
No log 10.4375 334 1.4623 0.4672 1.4623 1.2092
No log 10.5 336 1.7003 0.3852 1.7003 1.3040
No log 10.5625 338 1.8366 0.3333 1.8366 1.3552
No log 10.625 340 1.7161 0.3688 1.7161 1.3100
No log 10.6875 342 1.5138 0.4397 1.5138 1.2304
No log 10.75 344 1.4964 0.4539 1.4964 1.2233
No log 10.8125 346 1.3425 0.5106 1.3425 1.1587
No log 10.875 348 1.1612 0.5286 1.1612 1.0776
No log 10.9375 350 1.0325 0.6176 1.0325 1.0161
No log 11.0 352 0.9966 0.6963 0.9966 0.9983
No log 11.0625 354 1.0022 0.6519 1.0022 1.0011
No log 11.125 356 1.0142 0.6222 1.0142 1.0071
No log 11.1875 358 0.9868 0.6269 0.9868 0.9934
No log 11.25 360 0.9572 0.6316 0.9572 0.9784
No log 11.3125 362 0.9400 0.6466 0.9400 0.9695
No log 11.375 364 0.9526 0.6715 0.9526 0.9760
No log 11.4375 366 1.0211 0.5926 1.0211 1.0105
No log 11.5 368 1.1009 0.5373 1.1009 1.0492
No log 11.5625 370 1.0978 0.5263 1.0978 1.0478
No log 11.625 372 1.0887 0.5630 1.0887 1.0434
No log 11.6875 374 1.0972 0.5547 1.0972 1.0475
No log 11.75 376 1.2133 0.5 1.2133 1.1015
No log 11.8125 378 1.4893 0.4476 1.4893 1.2204
No log 11.875 380 1.4744 0.4571 1.4744 1.2143
No log 11.9375 382 1.3676 0.4812 1.3676 1.1695
No log 12.0 384 1.1045 0.5373 1.1045 1.0510
No log 12.0625 386 0.8884 0.6522 0.8884 0.9425
No log 12.125 388 0.8279 0.7092 0.8279 0.9099
No log 12.1875 390 0.7780 0.7092 0.7780 0.8820
No log 12.25 392 0.7585 0.7376 0.7585 0.8709
No log 12.3125 394 0.7615 0.7222 0.7615 0.8727
No log 12.375 396 0.7978 0.7397 0.7978 0.8932
No log 12.4375 398 0.8100 0.7273 0.8100 0.9000
No log 12.5 400 0.8517 0.7183 0.8517 0.9229
No log 12.5625 402 0.9017 0.7050 0.9017 0.9496
No log 12.625 404 0.8926 0.6815 0.8926 0.9448
No log 12.6875 406 0.9113 0.6815 0.9113 0.9546
No log 12.75 408 0.9429 0.6364 0.9429 0.9710
No log 12.8125 410 0.9749 0.5802 0.9749 0.9874
No log 12.875 412 1.0136 0.5426 1.0136 1.0068
No log 12.9375 414 1.0215 0.5630 1.0215 1.0107
No log 13.0 416 1.0263 0.5630 1.0263 1.0131
No log 13.0625 418 1.0399 0.5630 1.0399 1.0198
No log 13.125 420 1.1443 0.5362 1.1443 1.0697
No log 13.1875 422 1.1907 0.5362 1.1907 1.0912
No log 13.25 424 1.2032 0.5362 1.2032 1.0969
No log 13.3125 426 1.2905 0.5655 1.2905 1.1360
No log 13.375 428 1.3812 0.5034 1.3812 1.1752
No log 13.4375 430 1.2718 0.5734 1.2718 1.1277
No log 13.5 432 1.1571 0.5481 1.1571 1.0757
No log 13.5625 434 1.1101 0.5401 1.1101 1.0536
No log 13.625 436 1.0582 0.5821 1.0582 1.0287
No log 13.6875 438 1.0730 0.5564 1.0730 1.0358
No log 13.75 440 1.0980 0.5821 1.0980 1.0479
No log 13.8125 442 1.1041 0.5714 1.1041 1.0507
No log 13.875 444 1.0766 0.5821 1.0766 1.0376
No log 13.9375 446 1.0200 0.5802 1.0200 1.0100
No log 14.0 448 0.9938 0.6061 0.9938 0.9969
No log 14.0625 450 0.9786 0.6269 0.9786 0.9893
No log 14.125 452 0.9675 0.6119 0.9675 0.9836
No log 14.1875 454 0.9387 0.6715 0.9387 0.9688
No log 14.25 456 0.8996 0.6519 0.8996 0.9485
No log 14.3125 458 0.8826 0.6765 0.8826 0.9395
No log 14.375 460 0.8764 0.6957 0.8764 0.9361
No log 14.4375 462 0.8777 0.6667 0.8777 0.9368
No log 14.5 464 0.8650 0.7101 0.8650 0.9301
No log 14.5625 466 0.8426 0.7194 0.8426 0.9179
No log 14.625 468 0.8276 0.7183 0.8276 0.9097
No log 14.6875 470 0.8230 0.7397 0.8230 0.9072
No log 14.75 472 0.8211 0.7172 0.8211 0.9061
No log 14.8125 474 0.8210 0.7183 0.8210 0.9061
No log 14.875 476 0.8348 0.7101 0.8348 0.9137
No log 14.9375 478 0.9171 0.6471 0.9171 0.9577
No log 15.0 480 1.0443 0.5899 1.0443 1.0219
No log 15.0625 482 1.1870 0.5441 1.1870 1.0895
No log 15.125 484 1.2603 0.4853 1.2603 1.1226
No log 15.1875 486 1.3428 0.4593 1.3428 1.1588
No log 15.25 488 1.2204 0.4853 1.2204 1.1047
No log 15.3125 490 1.0610 0.5649 1.0610 1.0300
No log 15.375 492 0.9575 0.6316 0.9575 0.9785
No log 15.4375 494 0.8743 0.6861 0.8743 0.9351
No log 15.5 496 0.8441 0.6861 0.8441 0.9187
No log 15.5625 498 0.8262 0.7 0.8262 0.9089
0.3558 15.625 500 0.8229 0.7376 0.8229 0.9071
0.3558 15.6875 502 0.8223 0.7376 0.8223 0.9068
0.3558 15.75 504 0.8138 0.7376 0.8138 0.9021
0.3558 15.8125 506 0.8075 0.7143 0.8075 0.8986
0.3558 15.875 508 0.8300 0.6906 0.8300 0.9110
0.3558 15.9375 510 0.8414 0.6906 0.8414 0.9173

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0+cu118
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
2
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_usingALLEssays_FineTuningAraBERT_run2_AugV5_k7_task1_organization

Finetuned
(4019)
this model