ArabicNewSplits6_FineTuningAraBERTFreeze_run2_AugV5_k4_task2_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: 0.6735
  • Qwk: 0.5577
  • Mse: 0.6735
  • Rmse: 0.8207

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: 16
  • eval_batch_size: 16
  • 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 0.1667 2 6.4020 -0.0278 6.4020 2.5302
No log 0.3333 4 4.2829 -0.0284 4.2829 2.0695
No log 0.5 6 2.9516 0.0047 2.9516 1.7180
No log 0.6667 8 2.0860 0.0084 2.0860 1.4443
No log 0.8333 10 1.4756 -0.0227 1.4756 1.2148
No log 1.0 12 1.0913 0.0128 1.0913 1.0446
No log 1.1667 14 0.9951 -0.0136 0.9951 0.9975
No log 1.3333 16 0.9209 0.0466 0.9209 0.9596
No log 1.5 18 0.8336 0.0873 0.8336 0.9130
No log 1.6667 20 0.7412 0.1919 0.7412 0.8609
No log 1.8333 22 0.7366 0.1631 0.7366 0.8582
No log 2.0 24 0.6775 0.3824 0.6775 0.8231
No log 2.1667 26 0.6421 0.3575 0.6421 0.8013
No log 2.3333 28 0.6633 0.3211 0.6633 0.8144
No log 2.5 30 0.6799 0.2456 0.6799 0.8246
No log 2.6667 32 0.6636 0.4111 0.6636 0.8146
No log 2.8333 34 0.6411 0.4079 0.6411 0.8007
No log 3.0 36 0.6251 0.3837 0.6251 0.7906
No log 3.1667 38 0.6159 0.4046 0.6159 0.7848
No log 3.3333 40 0.6018 0.3646 0.6018 0.7758
No log 3.5 42 0.5961 0.3646 0.5961 0.7721
No log 3.6667 44 0.5998 0.3892 0.5998 0.7745
No log 3.8333 46 0.6227 0.3857 0.6227 0.7891
No log 4.0 48 0.6621 0.3643 0.6621 0.8137
No log 4.1667 50 0.7007 0.3637 0.7007 0.8371
No log 4.3333 52 0.6997 0.3515 0.6997 0.8365
No log 4.5 54 0.6629 0.3471 0.6629 0.8142
No log 4.6667 56 0.6446 0.3664 0.6446 0.8029
No log 4.8333 58 0.6321 0.3276 0.6321 0.7950
No log 5.0 60 0.6231 0.3639 0.6231 0.7894
No log 5.1667 62 0.5836 0.4144 0.5836 0.7639
No log 5.3333 64 0.5571 0.4260 0.5571 0.7464
No log 5.5 66 0.5556 0.4942 0.5556 0.7454
No log 5.6667 68 0.5794 0.5082 0.5794 0.7612
No log 5.8333 70 0.5838 0.4840 0.5838 0.7641
No log 6.0 72 0.5550 0.4817 0.5550 0.7450
No log 6.1667 74 0.5365 0.4285 0.5365 0.7325
No log 6.3333 76 0.5481 0.4224 0.5481 0.7404
No log 6.5 78 0.5517 0.4022 0.5517 0.7428
No log 6.6667 80 0.5474 0.3905 0.5474 0.7398
No log 6.8333 82 0.5341 0.4018 0.5341 0.7309
No log 7.0 84 0.5262 0.4312 0.5262 0.7254
No log 7.1667 86 0.5455 0.4992 0.5455 0.7386
No log 7.3333 88 0.5776 0.5184 0.5776 0.7600
No log 7.5 90 0.5862 0.5205 0.5862 0.7657
No log 7.6667 92 0.5579 0.5205 0.5579 0.7469
No log 7.8333 94 0.5279 0.5082 0.5279 0.7266
No log 8.0 96 0.5238 0.5082 0.5238 0.7237
No log 8.1667 98 0.5324 0.4992 0.5324 0.7296
No log 8.3333 100 0.5536 0.5361 0.5536 0.7440
No log 8.5 102 0.5660 0.5270 0.5660 0.7523
No log 8.6667 104 0.5510 0.5264 0.5510 0.7423
No log 8.8333 106 0.5451 0.5485 0.5451 0.7383
No log 9.0 108 0.5298 0.5124 0.5298 0.7279
No log 9.1667 110 0.5326 0.5370 0.5326 0.7298
No log 9.3333 112 0.5431 0.5205 0.5431 0.7369
No log 9.5 114 0.5532 0.5070 0.5532 0.7437
No log 9.6667 116 0.5531 0.5272 0.5531 0.7437
No log 9.8333 118 0.5290 0.5537 0.5290 0.7273
No log 10.0 120 0.5081 0.4832 0.5081 0.7128
No log 10.1667 122 0.5097 0.4745 0.5097 0.7140
No log 10.3333 124 0.5175 0.5044 0.5175 0.7193
No log 10.5 126 0.5162 0.5271 0.5162 0.7185
No log 10.6667 128 0.5116 0.5259 0.5116 0.7153
No log 10.8333 130 0.5166 0.5351 0.5166 0.7188
No log 11.0 132 0.5267 0.5656 0.5267 0.7258
No log 11.1667 134 0.5392 0.5806 0.5392 0.7343
No log 11.3333 136 0.5421 0.5822 0.5421 0.7363
No log 11.5 138 0.5370 0.5718 0.5370 0.7328
No log 11.6667 140 0.5332 0.5721 0.5332 0.7302
No log 11.8333 142 0.5367 0.6034 0.5367 0.7326
No log 12.0 144 0.5408 0.5635 0.5408 0.7354
No log 12.1667 146 0.5496 0.5433 0.5496 0.7414
No log 12.3333 148 0.5680 0.5138 0.5680 0.7537
No log 12.5 150 0.5808 0.4843 0.5808 0.7621
No log 12.6667 152 0.6015 0.5194 0.6015 0.7755
No log 12.8333 154 0.6067 0.5251 0.6067 0.7789
No log 13.0 156 0.5944 0.5598 0.5944 0.7709
No log 13.1667 158 0.5784 0.5647 0.5784 0.7605
No log 13.3333 160 0.5717 0.5506 0.5717 0.7561
No log 13.5 162 0.5676 0.5486 0.5676 0.7534
No log 13.6667 164 0.5811 0.5194 0.5811 0.7623
No log 13.8333 166 0.6025 0.5455 0.6025 0.7762
No log 14.0 168 0.6045 0.5085 0.6045 0.7775
No log 14.1667 170 0.6075 0.5357 0.6075 0.7794
No log 14.3333 172 0.6091 0.5688 0.6091 0.7804
No log 14.5 174 0.5975 0.5709 0.5975 0.7730
No log 14.6667 176 0.5851 0.5278 0.5851 0.7649
No log 14.8333 178 0.5931 0.5215 0.5931 0.7701
No log 15.0 180 0.6001 0.4892 0.6001 0.7747
No log 15.1667 182 0.6177 0.5047 0.6177 0.7859
No log 15.3333 184 0.6443 0.5208 0.6443 0.8027
No log 15.5 186 0.6426 0.5282 0.6426 0.8016
No log 15.6667 188 0.6388 0.5522 0.6388 0.7993
No log 15.8333 190 0.6480 0.5483 0.6480 0.8050
No log 16.0 192 0.6400 0.5709 0.6400 0.8000
No log 16.1667 194 0.6306 0.5156 0.6306 0.7941
No log 16.3333 196 0.6268 0.5137 0.6268 0.7917
No log 16.5 198 0.6230 0.5097 0.6230 0.7893
No log 16.6667 200 0.6247 0.5328 0.6247 0.7904
No log 16.8333 202 0.6368 0.5324 0.6368 0.7980
No log 17.0 204 0.6451 0.5380 0.6451 0.8032
No log 17.1667 206 0.6467 0.5617 0.6467 0.8042
No log 17.3333 208 0.6468 0.5493 0.6468 0.8043
No log 17.5 210 0.6552 0.5534 0.6552 0.8095
No log 17.6667 212 0.6599 0.5752 0.6599 0.8123
No log 17.8333 214 0.6572 0.5392 0.6572 0.8107
No log 18.0 216 0.6718 0.5783 0.6718 0.8196
No log 18.1667 218 0.6751 0.5586 0.6751 0.8217
No log 18.3333 220 0.6648 0.5561 0.6648 0.8153
No log 18.5 222 0.6654 0.5729 0.6654 0.8157
No log 18.6667 224 0.6751 0.5862 0.6751 0.8216
No log 18.8333 226 0.6740 0.5738 0.6740 0.8210
No log 19.0 228 0.6839 0.5630 0.6839 0.8270
No log 19.1667 230 0.7130 0.5750 0.7130 0.8444
No log 19.3333 232 0.7293 0.5263 0.7293 0.8540
No log 19.5 234 0.7084 0.5102 0.7084 0.8417
No log 19.6667 236 0.6922 0.5188 0.6922 0.8320
No log 19.8333 238 0.6843 0.5238 0.6843 0.8272
No log 20.0 240 0.6818 0.5628 0.6818 0.8257
No log 20.1667 242 0.6607 0.5321 0.6607 0.8128
No log 20.3333 244 0.6538 0.5374 0.6538 0.8086
No log 20.5 246 0.6655 0.5273 0.6655 0.8158
No log 20.6667 248 0.6789 0.5224 0.6789 0.8239
No log 20.8333 250 0.6713 0.5279 0.6713 0.8193
No log 21.0 252 0.6645 0.5310 0.6645 0.8152
No log 21.1667 254 0.6596 0.5150 0.6596 0.8121
No log 21.3333 256 0.6529 0.5603 0.6529 0.8080
No log 21.5 258 0.6525 0.5493 0.6525 0.8077
No log 21.6667 260 0.6515 0.5928 0.6515 0.8072
No log 21.8333 262 0.6559 0.6108 0.6559 0.8099
No log 22.0 264 0.6606 0.5840 0.6606 0.8128
No log 22.1667 266 0.6681 0.5452 0.6681 0.8174
No log 22.3333 268 0.6632 0.5423 0.6632 0.8144
No log 22.5 270 0.6635 0.5395 0.6635 0.8146
No log 22.6667 272 0.6688 0.5762 0.6688 0.8178
No log 22.8333 274 0.6696 0.5719 0.6696 0.8183
No log 23.0 276 0.6764 0.5886 0.6764 0.8224
No log 23.1667 278 0.6722 0.5790 0.6722 0.8199
No log 23.3333 280 0.6715 0.5663 0.6715 0.8194
No log 23.5 282 0.6767 0.5520 0.6767 0.8226
No log 23.6667 284 0.6884 0.5424 0.6884 0.8297
No log 23.8333 286 0.6993 0.5615 0.6993 0.8363
No log 24.0 288 0.6914 0.5659 0.6914 0.8315
No log 24.1667 290 0.6920 0.5589 0.6920 0.8319
No log 24.3333 292 0.6886 0.5437 0.6886 0.8298
No log 24.5 294 0.6980 0.5412 0.6980 0.8355
No log 24.6667 296 0.6884 0.5589 0.6884 0.8297
No log 24.8333 298 0.6764 0.5840 0.6764 0.8224
No log 25.0 300 0.6660 0.5806 0.6660 0.8161
No log 25.1667 302 0.6498 0.5731 0.6498 0.8061
No log 25.3333 304 0.6536 0.5450 0.6536 0.8085
No log 25.5 306 0.6743 0.5295 0.6743 0.8212
No log 25.6667 308 0.7006 0.5112 0.7006 0.8370
No log 25.8333 310 0.6983 0.5631 0.6983 0.8357
No log 26.0 312 0.6808 0.5592 0.6808 0.8251
No log 26.1667 314 0.6793 0.5855 0.6793 0.8242
No log 26.3333 316 0.6797 0.5707 0.6797 0.8245
No log 26.5 318 0.6852 0.5634 0.6852 0.8277
No log 26.6667 320 0.6902 0.5461 0.6902 0.8308
No log 26.8333 322 0.7031 0.5353 0.7031 0.8385
No log 27.0 324 0.7029 0.5353 0.7029 0.8384
No log 27.1667 326 0.6924 0.5504 0.6924 0.8321
No log 27.3333 328 0.6899 0.5646 0.6899 0.8306
No log 27.5 330 0.6922 0.5390 0.6922 0.8320
No log 27.6667 332 0.6802 0.5532 0.6802 0.8248
No log 27.8333 334 0.6660 0.5717 0.6660 0.8161
No log 28.0 336 0.6680 0.5759 0.6680 0.8173
No log 28.1667 338 0.6739 0.5866 0.6739 0.8209
No log 28.3333 340 0.6591 0.5784 0.6591 0.8118
No log 28.5 342 0.6539 0.5560 0.6539 0.8087
No log 28.6667 344 0.6588 0.5346 0.6588 0.8116
No log 28.8333 346 0.6592 0.5418 0.6592 0.8119
No log 29.0 348 0.6500 0.5546 0.6500 0.8062
No log 29.1667 350 0.6490 0.5670 0.6490 0.8056
No log 29.3333 352 0.6540 0.5670 0.6540 0.8087
No log 29.5 354 0.6653 0.5531 0.6653 0.8156
No log 29.6667 356 0.6739 0.5601 0.6739 0.8209
No log 29.8333 358 0.6720 0.5601 0.6720 0.8198
No log 30.0 360 0.6689 0.5532 0.6689 0.8179
No log 30.1667 362 0.6694 0.5422 0.6694 0.8182
No log 30.3333 364 0.6711 0.5670 0.6711 0.8192
No log 30.5 366 0.6792 0.5754 0.6792 0.8241
No log 30.6667 368 0.6936 0.5842 0.6936 0.8328
No log 30.8333 370 0.6990 0.5887 0.6990 0.8361
No log 31.0 372 0.6937 0.5822 0.6937 0.8329
No log 31.1667 374 0.6920 0.5851 0.6920 0.8319
No log 31.3333 376 0.7049 0.5475 0.7049 0.8396
No log 31.5 378 0.7058 0.5311 0.7058 0.8401
No log 31.6667 380 0.6868 0.5153 0.6868 0.8287
No log 31.8333 382 0.6704 0.4989 0.6704 0.8188
No log 32.0 384 0.6748 0.5124 0.6748 0.8215
No log 32.1667 386 0.6833 0.5810 0.6833 0.8266
No log 32.3333 388 0.6965 0.5884 0.6965 0.8346
No log 32.5 390 0.7200 0.5703 0.7200 0.8485
No log 32.6667 392 0.7552 0.5231 0.7552 0.8690
No log 32.8333 394 0.7791 0.5265 0.7791 0.8827
No log 33.0 396 0.7843 0.5281 0.7843 0.8856
No log 33.1667 398 0.7769 0.5354 0.7769 0.8814
No log 33.3333 400 0.7491 0.5517 0.7491 0.8655
No log 33.5 402 0.7436 0.5655 0.7436 0.8623
No log 33.6667 404 0.7468 0.5586 0.7468 0.8642
No log 33.8333 406 0.7568 0.5226 0.7568 0.8699
No log 34.0 408 0.7676 0.5272 0.7676 0.8761
No log 34.1667 410 0.7759 0.5016 0.7759 0.8809
No log 34.3333 412 0.7570 0.5311 0.7570 0.8700
No log 34.5 414 0.7284 0.5435 0.7284 0.8534
No log 34.6667 416 0.7108 0.5615 0.7108 0.8431
No log 34.8333 418 0.7017 0.5159 0.7017 0.8377
No log 35.0 420 0.6961 0.5229 0.6961 0.8343
No log 35.1667 422 0.7073 0.5475 0.7073 0.8410
No log 35.3333 424 0.7223 0.5403 0.7223 0.8499
No log 35.5 426 0.7103 0.5632 0.7103 0.8428
No log 35.6667 428 0.6965 0.5532 0.6965 0.8346
No log 35.8333 430 0.6900 0.5604 0.6900 0.8307
No log 36.0 432 0.6914 0.5547 0.6914 0.8315
No log 36.1667 434 0.7085 0.5334 0.7085 0.8417
No log 36.3333 436 0.7216 0.5211 0.7216 0.8495
No log 36.5 438 0.7194 0.5265 0.7194 0.8481
No log 36.6667 440 0.7082 0.5236 0.7082 0.8416
No log 36.8333 442 0.6961 0.5688 0.6961 0.8343
No log 37.0 444 0.6775 0.5691 0.6775 0.8231
No log 37.1667 446 0.6782 0.5702 0.6782 0.8235
No log 37.3333 448 0.6859 0.5809 0.6859 0.8282
No log 37.5 450 0.7001 0.5463 0.7001 0.8367
No log 37.6667 452 0.7065 0.5534 0.7065 0.8405
No log 37.8333 454 0.7039 0.5618 0.7039 0.8390
No log 38.0 456 0.7102 0.5463 0.7102 0.8427
No log 38.1667 458 0.7254 0.5601 0.7254 0.8517
No log 38.3333 460 0.7469 0.5201 0.7469 0.8642
No log 38.5 462 0.7515 0.5185 0.7515 0.8669
No log 38.6667 464 0.7330 0.5405 0.7330 0.8561
No log 38.8333 466 0.7201 0.5560 0.7201 0.8486
No log 39.0 468 0.7127 0.5603 0.7127 0.8442
No log 39.1667 470 0.7116 0.5659 0.7116 0.8436
No log 39.3333 472 0.7169 0.5517 0.7169 0.8467
No log 39.5 474 0.7158 0.5431 0.7158 0.8461
No log 39.6667 476 0.7276 0.5544 0.7276 0.8530
No log 39.8333 478 0.7108 0.5559 0.7108 0.8431
No log 40.0 480 0.6869 0.5589 0.6869 0.8288
No log 40.1667 482 0.6792 0.5562 0.6792 0.8241
No log 40.3333 484 0.6822 0.5604 0.6822 0.8259
No log 40.5 486 0.6908 0.5517 0.6908 0.8312
No log 40.6667 488 0.6922 0.5663 0.6922 0.8320
No log 40.8333 490 0.6867 0.5675 0.6867 0.8287
No log 41.0 492 0.6847 0.5675 0.6847 0.8274
No log 41.1667 494 0.6826 0.5532 0.6826 0.8262
No log 41.3333 496 0.6892 0.5559 0.6892 0.8302
No log 41.5 498 0.6963 0.5408 0.6963 0.8344
0.566 41.6667 500 0.6889 0.5693 0.6889 0.8300
0.566 41.8333 502 0.6866 0.5559 0.6866 0.8286
0.566 42.0 504 0.6735 0.5531 0.6735 0.8207
0.566 42.1667 506 0.6671 0.5488 0.6671 0.8168
0.566 42.3333 508 0.6747 0.5411 0.6747 0.8214
0.566 42.5 510 0.6735 0.5577 0.6735 0.8207

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.1+cu121
  • Datasets 3.2.0
  • 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/ArabicNewSplits6_FineTuningAraBERTFreeze_run2_AugV5_k4_task2_organization

Finetuned
(4040)
this model