ArabicNewSplits7_FineTuningAraBERT_run3_AugV5_k12_task5_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: 0.5713
  • Qwk: 0.6210
  • Mse: 0.5713
  • Rmse: 0.7558

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 3.8998 0.0124 3.8998 1.9748
No log 0.125 4 1.8693 0.0318 1.8693 1.3672
No log 0.1875 6 1.1996 -0.0627 1.1996 1.0953
No log 0.25 8 1.0795 0.2441 1.0795 1.0390
No log 0.3125 10 1.0976 0.1418 1.0976 1.0476
No log 0.375 12 1.2351 0.0249 1.2351 1.1114
No log 0.4375 14 1.4945 -0.0858 1.4945 1.2225
No log 0.5 16 1.6856 -0.0411 1.6856 1.2983
No log 0.5625 18 1.5115 -0.0560 1.5115 1.2294
No log 0.625 20 1.2983 -0.0328 1.2983 1.1394
No log 0.6875 22 1.1399 0.1268 1.1399 1.0676
No log 0.75 24 1.0546 0.2416 1.0546 1.0270
No log 0.8125 26 1.0514 0.0762 1.0514 1.0254
No log 0.875 28 1.0289 0.1076 1.0289 1.0143
No log 0.9375 30 1.0153 0.4051 1.0153 1.0076
No log 1.0 32 1.0239 0.2343 1.0239 1.0119
No log 1.0625 34 1.1216 0.1142 1.1216 1.0591
No log 1.125 36 1.1869 0.0 1.1869 1.0895
No log 1.1875 38 1.1328 0.0996 1.1328 1.0644
No log 1.25 40 0.9713 0.4167 0.9713 0.9855
No log 1.3125 42 0.9117 0.4031 0.9117 0.9548
No log 1.375 44 0.9185 0.4218 0.9185 0.9584
No log 1.4375 46 0.9131 0.4512 0.9131 0.9556
No log 1.5 48 0.9880 0.3790 0.9880 0.9940
No log 1.5625 50 1.1038 0.2513 1.1038 1.0506
No log 1.625 52 1.1056 0.2850 1.1056 1.0515
No log 1.6875 54 0.9419 0.375 0.9419 0.9705
No log 1.75 56 0.9125 0.2314 0.9125 0.9553
No log 1.8125 58 1.0117 0.1799 1.0117 1.0058
No log 1.875 60 1.0029 0.1545 1.0029 1.0015
No log 1.9375 62 0.9630 0.1783 0.9630 0.9813
No log 2.0 64 0.9758 0.3310 0.9758 0.9878
No log 2.0625 66 0.9465 0.4167 0.9465 0.9729
No log 2.125 68 0.8062 0.3435 0.8062 0.8979
No log 2.1875 70 0.7766 0.3652 0.7766 0.8813
No log 2.25 72 0.8179 0.3164 0.8179 0.9044
No log 2.3125 74 0.7901 0.3603 0.7901 0.8889
No log 2.375 76 0.7128 0.4831 0.7128 0.8443
No log 2.4375 78 0.7097 0.5763 0.7097 0.8425
No log 2.5 80 0.7035 0.5559 0.7035 0.8387
No log 2.5625 82 0.6607 0.5153 0.6607 0.8128
No log 2.625 84 0.6674 0.5562 0.6674 0.8170
No log 2.6875 86 0.6472 0.6272 0.6472 0.8045
No log 2.75 88 0.6976 0.6015 0.6976 0.8353
No log 2.8125 90 0.8251 0.5614 0.8251 0.9083
No log 2.875 92 1.0169 0.3942 1.0169 1.0084
No log 2.9375 94 1.0329 0.4073 1.0329 1.0163
No log 3.0 96 0.9518 0.4668 0.9518 0.9756
No log 3.0625 98 0.8173 0.6035 0.8173 0.9040
No log 3.125 100 0.7586 0.5902 0.7586 0.8710
No log 3.1875 102 0.7438 0.5675 0.7438 0.8624
No log 3.25 104 0.7557 0.5521 0.7557 0.8693
No log 3.3125 106 0.6627 0.6071 0.6627 0.8141
No log 3.375 108 0.6591 0.6445 0.6591 0.8119
No log 3.4375 110 0.7110 0.6529 0.7110 0.8432
No log 3.5 112 0.7763 0.5275 0.7763 0.8811
No log 3.5625 114 0.8187 0.5485 0.8187 0.9048
No log 3.625 116 0.7913 0.5239 0.7913 0.8895
No log 3.6875 118 0.8024 0.5968 0.8024 0.8958
No log 3.75 120 0.8101 0.6141 0.8101 0.9001
No log 3.8125 122 0.7107 0.6147 0.7107 0.8430
No log 3.875 124 0.6774 0.5495 0.6774 0.8231
No log 3.9375 126 0.6323 0.5603 0.6323 0.7952
No log 4.0 128 0.6063 0.5934 0.6063 0.7787
No log 4.0625 130 0.5699 0.6886 0.5699 0.7549
No log 4.125 132 0.5839 0.6719 0.5839 0.7642
No log 4.1875 134 0.5388 0.6878 0.5388 0.7340
No log 4.25 136 0.4915 0.7231 0.4915 0.7011
No log 4.3125 138 0.4837 0.7182 0.4837 0.6955
No log 4.375 140 0.5296 0.7483 0.5296 0.7277
No log 4.4375 142 0.6198 0.7469 0.6198 0.7873
No log 4.5 144 0.5382 0.7437 0.5382 0.7336
No log 4.5625 146 0.4707 0.7449 0.4707 0.6860
No log 4.625 148 0.4827 0.7301 0.4827 0.6948
No log 4.6875 150 0.5262 0.7437 0.5262 0.7254
No log 4.75 152 0.6911 0.6653 0.6911 0.8313
No log 4.8125 154 0.7210 0.6238 0.7210 0.8491
No log 4.875 156 0.6373 0.6053 0.6373 0.7983
No log 4.9375 158 0.6038 0.6311 0.6038 0.7771
No log 5.0 160 0.5812 0.6196 0.5812 0.7623
No log 5.0625 162 0.5698 0.6498 0.5698 0.7549
No log 5.125 164 0.6278 0.6137 0.6278 0.7923
No log 5.1875 166 0.8817 0.5394 0.8817 0.9390
No log 5.25 168 1.0177 0.5201 1.0177 1.0088
No log 5.3125 170 0.8755 0.5899 0.8755 0.9357
No log 5.375 172 0.6923 0.5980 0.6923 0.8321
No log 5.4375 174 0.6529 0.5584 0.6529 0.8080
No log 5.5 176 0.7399 0.5800 0.7399 0.8602
No log 5.5625 178 0.7666 0.5952 0.7666 0.8755
No log 5.625 180 0.6478 0.6537 0.6478 0.8049
No log 5.6875 182 0.5709 0.6602 0.5709 0.7556
No log 5.75 184 0.5491 0.6197 0.5491 0.7410
No log 5.8125 186 0.5811 0.6240 0.5811 0.7623
No log 5.875 188 0.6422 0.6190 0.6422 0.8013
No log 5.9375 190 0.6044 0.6592 0.6044 0.7775
No log 6.0 192 0.5460 0.6933 0.5460 0.7389
No log 6.0625 194 0.5185 0.6965 0.5185 0.7201
No log 6.125 196 0.5247 0.7193 0.5247 0.7244
No log 6.1875 198 0.5567 0.7381 0.5567 0.7461
No log 6.25 200 0.6952 0.6776 0.6952 0.8338
No log 6.3125 202 0.5979 0.7517 0.5979 0.7732
No log 6.375 204 0.4494 0.7544 0.4494 0.6704
No log 6.4375 206 0.4880 0.7263 0.4880 0.6986
No log 6.5 208 0.5542 0.6677 0.5542 0.7444
No log 6.5625 210 0.4867 0.6745 0.4867 0.6977
No log 6.625 212 0.5169 0.7109 0.5169 0.7190
No log 6.6875 214 0.5699 0.6305 0.5699 0.7549
No log 6.75 216 0.5967 0.5610 0.5967 0.7725
No log 6.8125 218 0.5763 0.6063 0.5763 0.7592
No log 6.875 220 0.6129 0.6328 0.6129 0.7829
No log 6.9375 222 0.6710 0.5832 0.6710 0.8191
No log 7.0 224 0.7187 0.5810 0.7187 0.8478
No log 7.0625 226 0.7672 0.5811 0.7672 0.8759
No log 7.125 228 0.7284 0.5856 0.7284 0.8535
No log 7.1875 230 0.6659 0.4584 0.6659 0.8160
No log 7.25 232 0.6285 0.4336 0.6285 0.7928
No log 7.3125 234 0.5921 0.5522 0.5921 0.7695
No log 7.375 236 0.5798 0.6133 0.5798 0.7614
No log 7.4375 238 0.6491 0.5856 0.6491 0.8056
No log 7.5 240 0.7345 0.5885 0.7345 0.8570
No log 7.5625 242 0.7009 0.5885 0.7009 0.8372
No log 7.625 244 0.5574 0.7373 0.5574 0.7466
No log 7.6875 246 0.5299 0.7379 0.5299 0.7280
No log 7.75 248 0.5456 0.7634 0.5456 0.7387
No log 7.8125 250 0.6236 0.7135 0.6236 0.7897
No log 7.875 252 0.6491 0.6656 0.6491 0.8057
No log 7.9375 254 0.6820 0.6244 0.6820 0.8258
No log 8.0 256 0.8080 0.5629 0.8080 0.8989
No log 8.0625 258 0.9693 0.4563 0.9693 0.9846
No log 8.125 260 0.8316 0.5640 0.8316 0.9119
No log 8.1875 262 0.5935 0.5966 0.5935 0.7704
No log 8.25 264 0.6554 0.6259 0.6554 0.8096
No log 8.3125 266 0.8910 0.5703 0.8910 0.9440
No log 8.375 268 0.9467 0.5239 0.9467 0.9730
No log 8.4375 270 0.8708 0.5703 0.8708 0.9332
No log 8.5 272 0.7199 0.6296 0.7199 0.8484
No log 8.5625 274 0.5521 0.6630 0.5521 0.7430
No log 8.625 276 0.5078 0.6954 0.5078 0.7126
No log 8.6875 278 0.5215 0.6954 0.5215 0.7222
No log 8.75 280 0.6014 0.5916 0.6014 0.7755
No log 8.8125 282 0.7245 0.5908 0.7245 0.8512
No log 8.875 284 0.7396 0.5885 0.7396 0.8600
No log 8.9375 286 0.6724 0.5856 0.6724 0.8200
No log 9.0 288 0.6029 0.5739 0.6029 0.7765
No log 9.0625 290 0.5744 0.5763 0.5744 0.7579
No log 9.125 292 0.5274 0.6305 0.5274 0.7262
No log 9.1875 294 0.5164 0.7122 0.5164 0.7186
No log 9.25 296 0.5466 0.7093 0.5466 0.7393
No log 9.3125 298 0.5693 0.6892 0.5693 0.7545
No log 9.375 300 0.6427 0.5905 0.6427 0.8017
No log 9.4375 302 0.6654 0.6229 0.6654 0.8157
No log 9.5 304 0.7612 0.5908 0.7612 0.8724
No log 9.5625 306 0.7921 0.5666 0.7921 0.8900
No log 9.625 308 0.8091 0.6061 0.8091 0.8995
No log 9.6875 310 0.7256 0.6209 0.7256 0.8518
No log 9.75 312 0.6493 0.6209 0.6493 0.8058
No log 9.8125 314 0.6075 0.5963 0.6075 0.7794
No log 9.875 316 0.6124 0.5482 0.6124 0.7826
No log 9.9375 318 0.5922 0.6109 0.5922 0.7696
No log 10.0 320 0.5565 0.6429 0.5565 0.7460
No log 10.0625 322 0.5362 0.6781 0.5362 0.7323
No log 10.125 324 0.5129 0.6878 0.5129 0.7162
No log 10.1875 326 0.5422 0.7005 0.5422 0.7363
No log 10.25 328 0.5258 0.7377 0.5258 0.7251
No log 10.3125 330 0.4739 0.7136 0.4739 0.6884
No log 10.375 332 0.4867 0.6995 0.4867 0.6976
No log 10.4375 334 0.5081 0.6766 0.5081 0.7128
No log 10.5 336 0.5184 0.6948 0.5184 0.7200
No log 10.5625 338 0.5105 0.7035 0.5105 0.7145
No log 10.625 340 0.5199 0.7122 0.5199 0.7211
No log 10.6875 342 0.5380 0.6539 0.5380 0.7335
No log 10.75 344 0.5482 0.6539 0.5482 0.7404
No log 10.8125 346 0.5135 0.7043 0.5135 0.7166
No log 10.875 348 0.5012 0.6606 0.5012 0.7080
No log 10.9375 350 0.5085 0.6933 0.5085 0.7131
No log 11.0 352 0.5750 0.6482 0.5750 0.7583
No log 11.0625 354 0.5662 0.6482 0.5662 0.7525
No log 11.125 356 0.5145 0.6925 0.5145 0.7173
No log 11.1875 358 0.5021 0.7051 0.5021 0.7086
No log 11.25 360 0.5282 0.6646 0.5282 0.7268
No log 11.3125 362 0.6271 0.6539 0.6271 0.7919
No log 11.375 364 0.7601 0.5631 0.7601 0.8718
No log 11.4375 366 0.7740 0.5340 0.7740 0.8798
No log 11.5 368 0.6763 0.5579 0.6763 0.8224
No log 11.5625 370 0.5558 0.6646 0.5558 0.7455
No log 11.625 372 0.5332 0.6387 0.5332 0.7302
No log 11.6875 374 0.5342 0.6499 0.5342 0.7309
No log 11.75 376 0.5358 0.6499 0.5358 0.7320
No log 11.8125 378 0.5443 0.6916 0.5443 0.7378
No log 11.875 380 0.5444 0.6798 0.5444 0.7378
No log 11.9375 382 0.5477 0.6798 0.5477 0.7401
No log 12.0 384 0.5496 0.6798 0.5496 0.7413
No log 12.0625 386 0.5384 0.6798 0.5384 0.7337
No log 12.125 388 0.5055 0.6925 0.5055 0.7110
No log 12.1875 390 0.5016 0.6925 0.5016 0.7082
No log 12.25 392 0.5149 0.6824 0.5149 0.7175
No log 12.3125 394 0.5285 0.6824 0.5285 0.7270
No log 12.375 396 0.5547 0.6841 0.5547 0.7448
No log 12.4375 398 0.5876 0.6838 0.5876 0.7666
No log 12.5 400 0.5421 0.6670 0.5421 0.7363
No log 12.5625 402 0.4898 0.6993 0.4898 0.6998
No log 12.625 404 0.4537 0.7298 0.4537 0.6736
No log 12.6875 406 0.4577 0.6978 0.4577 0.6766
No log 12.75 408 0.4776 0.7088 0.4776 0.6911
No log 12.8125 410 0.5237 0.7043 0.5237 0.7237
No log 12.875 412 0.5340 0.6708 0.5340 0.7307
No log 12.9375 414 0.5197 0.7011 0.5197 0.7209
No log 13.0 416 0.5420 0.6605 0.5420 0.7362
No log 13.0625 418 0.5838 0.6262 0.5838 0.7640
No log 13.125 420 0.6276 0.6262 0.6276 0.7922
No log 13.1875 422 0.6449 0.5849 0.6449 0.8030
No log 13.25 424 0.6949 0.5332 0.6949 0.8336
No log 13.3125 426 0.6944 0.5532 0.6944 0.8333
No log 13.375 428 0.6899 0.6079 0.6899 0.8306
No log 13.4375 430 0.6288 0.6333 0.6288 0.7930
No log 13.5 432 0.5484 0.6878 0.5484 0.7405
No log 13.5625 434 0.5124 0.6488 0.5124 0.7158
No log 13.625 436 0.5102 0.6451 0.5102 0.7143
No log 13.6875 438 0.5341 0.6762 0.5341 0.7308
No log 13.75 440 0.6236 0.6592 0.6236 0.7897
No log 13.8125 442 0.7039 0.6513 0.7039 0.8390
No log 13.875 444 0.6305 0.6485 0.6305 0.7940
No log 13.9375 446 0.5191 0.6779 0.5191 0.7205
No log 14.0 448 0.4940 0.6911 0.4940 0.7029
No log 14.0625 450 0.4952 0.6634 0.4953 0.7037
No log 14.125 452 0.4875 0.7033 0.4875 0.6982
No log 14.1875 454 0.4791 0.6866 0.4791 0.6922
No log 14.25 456 0.5695 0.6420 0.5695 0.7546
No log 14.3125 458 0.7497 0.6145 0.7497 0.8659
No log 14.375 460 0.8037 0.5484 0.8037 0.8965
No log 14.4375 462 0.7320 0.5431 0.7320 0.8556
No log 14.5 464 0.6203 0.5998 0.6203 0.7876
No log 14.5625 466 0.5367 0.6588 0.5367 0.7326
No log 14.625 468 0.5064 0.6706 0.5064 0.7116
No log 14.6875 470 0.5434 0.6450 0.5434 0.7371
No log 14.75 472 0.6092 0.6368 0.6092 0.7805
No log 14.8125 474 0.6276 0.64 0.6276 0.7922
No log 14.875 476 0.5827 0.6052 0.5827 0.7633
No log 14.9375 478 0.5566 0.6160 0.5566 0.7460
No log 15.0 480 0.5152 0.6272 0.5152 0.7178
No log 15.0625 482 0.5189 0.6679 0.5189 0.7203
No log 15.125 484 0.5802 0.6052 0.5802 0.7617
No log 15.1875 486 0.6943 0.6095 0.6943 0.8332
No log 15.25 488 0.7230 0.5966 0.7230 0.8503
No log 15.3125 490 0.6536 0.6573 0.6536 0.8084
No log 15.375 492 0.5611 0.6961 0.5611 0.7491
No log 15.4375 494 0.5335 0.6567 0.5335 0.7304
No log 15.5 496 0.5405 0.6740 0.5405 0.7352
No log 15.5625 498 0.5630 0.6420 0.5630 0.7503
0.2956 15.625 500 0.6223 0.6300 0.6223 0.7889
0.2956 15.6875 502 0.7031 0.6044 0.7031 0.8385
0.2956 15.75 504 0.7297 0.5943 0.7297 0.8542
0.2956 15.8125 506 0.6798 0.5734 0.6798 0.8245
0.2956 15.875 508 0.6248 0.5811 0.6248 0.7904
0.2956 15.9375 510 0.5713 0.6210 0.5713 0.7558

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0+cu118
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
8
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/ArabicNewSplits7_FineTuningAraBERT_run3_AugV5_k12_task5_organization

Finetuned
(4223)
this model