ArabicNewSplits6_FineTuningAraBERTFreeze_run1_AugV5_k16_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: 0.7776
  • Qwk: 0.4882
  • Mse: 0.7776
  • Rmse: 0.8818

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 0.05 2 6.4769 -0.0278 6.4769 2.5450
No log 0.1 4 4.3481 -0.0299 4.3481 2.0852
No log 0.15 6 3.0631 0.0124 3.0631 1.7502
No log 0.2 8 2.2538 0.0246 2.2538 1.5013
No log 0.25 10 1.6411 -0.0188 1.6411 1.2811
No log 0.3 12 1.3742 -0.0448 1.3742 1.1723
No log 0.35 14 1.3858 -0.1083 1.3858 1.1772
No log 0.4 16 1.4077 -0.0974 1.4077 1.1865
No log 0.45 18 1.4918 -0.0974 1.4918 1.2214
No log 0.5 20 1.5067 -0.1174 1.5067 1.2275
No log 0.55 22 1.3231 -0.0476 1.3231 1.1502
No log 0.6 24 1.2901 -0.0433 1.2901 1.1358
No log 0.65 26 1.2365 0.0580 1.2365 1.1120
No log 0.7 28 1.0687 0.0923 1.0687 1.0338
No log 0.75 30 0.9828 0.1360 0.9828 0.9913
No log 0.8 32 0.9082 0.1668 0.9082 0.9530
No log 0.85 34 0.9454 0.1675 0.9454 0.9723
No log 0.9 36 0.9770 0.1173 0.9770 0.9884
No log 0.95 38 1.0598 0.0861 1.0598 1.0295
No log 1.0 40 1.1523 0.1104 1.1523 1.0735
No log 1.05 42 1.2984 0.1539 1.2984 1.1395
No log 1.1 44 1.3852 0.1613 1.3852 1.1770
No log 1.15 46 1.3003 0.1754 1.3003 1.1403
No log 1.2 48 1.0849 0.1429 1.0849 1.0416
No log 1.25 50 0.8493 0.2700 0.8493 0.9216
No log 1.3 52 0.7653 0.3279 0.7653 0.8748
No log 1.35 54 0.7767 0.3176 0.7767 0.8813
No log 1.4 56 0.8036 0.3212 0.8036 0.8965
No log 1.45 58 0.8259 0.3231 0.8259 0.9088
No log 1.5 60 0.7943 0.3581 0.7943 0.8912
No log 1.55 62 0.7836 0.3776 0.7836 0.8852
No log 1.6 64 0.7501 0.4250 0.7501 0.8661
No log 1.65 66 0.7400 0.4466 0.7400 0.8603
No log 1.7 68 0.7612 0.4273 0.7612 0.8725
No log 1.75 70 0.8506 0.4042 0.8506 0.9223
No log 1.8 72 0.8032 0.4441 0.8032 0.8962
No log 1.85 74 0.6830 0.4447 0.6830 0.8264
No log 1.9 76 0.6253 0.4605 0.6253 0.7907
No log 1.95 78 0.5947 0.4567 0.5947 0.7711
No log 2.0 80 0.6053 0.4841 0.6053 0.7780
No log 2.05 82 0.6174 0.4409 0.6174 0.7857
No log 2.1 84 0.6439 0.4291 0.6439 0.8024
No log 2.15 86 0.6992 0.4623 0.6992 0.8362
No log 2.2 88 0.7370 0.4836 0.7370 0.8585
No log 2.25 90 0.7894 0.4435 0.7894 0.8885
No log 2.3 92 0.7873 0.4751 0.7873 0.8873
No log 2.35 94 0.7967 0.4686 0.7967 0.8926
No log 2.4 96 0.7628 0.4804 0.7628 0.8734
No log 2.45 98 0.7155 0.5016 0.7155 0.8459
No log 2.5 100 0.6568 0.5103 0.6568 0.8104
No log 2.55 102 0.6633 0.5361 0.6633 0.8145
No log 2.6 104 0.6773 0.5218 0.6773 0.8230
No log 2.65 106 0.7989 0.5092 0.7989 0.8938
No log 2.7 108 1.0771 0.4010 1.0771 1.0378
No log 2.75 110 1.1492 0.3819 1.1492 1.0720
No log 2.8 112 0.9638 0.4146 0.9638 0.9817
No log 2.85 114 0.7458 0.5068 0.7458 0.8636
No log 2.9 116 0.6373 0.4864 0.6373 0.7983
No log 2.95 118 0.6015 0.4649 0.6015 0.7756
No log 3.0 120 0.6061 0.4718 0.6061 0.7785
No log 3.05 122 0.6339 0.4769 0.6339 0.7962
No log 3.1 124 0.7343 0.4853 0.7343 0.8569
No log 3.15 126 0.7859 0.4880 0.7859 0.8865
No log 3.2 128 0.7952 0.4880 0.7952 0.8918
No log 3.25 130 0.7375 0.4796 0.7375 0.8588
No log 3.3 132 0.6614 0.4631 0.6614 0.8133
No log 3.35 134 0.6286 0.4646 0.6286 0.7928
No log 3.4 136 0.6282 0.4791 0.6282 0.7926
No log 3.45 138 0.6470 0.4538 0.6470 0.8044
No log 3.5 140 0.7268 0.4972 0.7268 0.8525
No log 3.55 142 0.8698 0.4389 0.8698 0.9326
No log 3.6 144 0.8607 0.4565 0.8607 0.9277
No log 3.65 146 0.7567 0.5069 0.7567 0.8699
No log 3.7 148 0.6884 0.4770 0.6884 0.8297
No log 3.75 150 0.6808 0.4521 0.6808 0.8251
No log 3.8 152 0.6758 0.4466 0.6758 0.8221
No log 3.85 154 0.6870 0.4789 0.6870 0.8289
No log 3.9 156 0.6729 0.4547 0.6729 0.8203
No log 3.95 158 0.6785 0.4426 0.6785 0.8237
No log 4.0 160 0.6595 0.4766 0.6595 0.8121
No log 4.05 162 0.6712 0.4690 0.6712 0.8193
No log 4.1 164 0.7035 0.4407 0.7035 0.8387
No log 4.15 166 0.7055 0.4217 0.7055 0.8400
No log 4.2 168 0.7457 0.4953 0.7457 0.8635
No log 4.25 170 0.8108 0.4920 0.8108 0.9004
No log 4.3 172 0.8110 0.4913 0.8110 0.9006
No log 4.35 174 0.7681 0.5040 0.7681 0.8764
No log 4.4 176 0.7365 0.4565 0.7365 0.8582
No log 4.45 178 0.7546 0.4454 0.7546 0.8687
No log 4.5 180 0.7475 0.4613 0.7475 0.8646
No log 4.55 182 0.7649 0.5059 0.7649 0.8746
No log 4.6 184 0.8233 0.4795 0.8233 0.9073
No log 4.65 186 0.8290 0.4742 0.8290 0.9105
No log 4.7 188 0.7957 0.4630 0.7957 0.8920
No log 4.75 190 0.7672 0.4335 0.7672 0.8759
No log 4.8 192 0.7691 0.4413 0.7691 0.8770
No log 4.85 194 0.7857 0.4452 0.7857 0.8864
No log 4.9 196 0.7789 0.4247 0.7789 0.8825
No log 4.95 198 0.7592 0.4247 0.7592 0.8713
No log 5.0 200 0.7236 0.4358 0.7236 0.8506
No log 5.05 202 0.7168 0.4239 0.7168 0.8467
No log 5.1 204 0.7468 0.4132 0.7468 0.8642
No log 5.15 206 0.7689 0.4493 0.7689 0.8769
No log 5.2 208 0.7304 0.4341 0.7304 0.8546
No log 5.25 210 0.7113 0.4637 0.7113 0.8434
No log 5.3 212 0.7177 0.4394 0.7177 0.8471
No log 5.35 214 0.7334 0.4539 0.7334 0.8564
No log 5.4 216 0.7507 0.4296 0.7507 0.8664
No log 5.45 218 0.7643 0.4224 0.7643 0.8742
No log 5.5 220 0.7820 0.4487 0.7820 0.8843
No log 5.55 222 0.8234 0.4616 0.8234 0.9074
No log 5.6 224 0.8012 0.4464 0.8012 0.8951
No log 5.65 226 0.7771 0.4562 0.7771 0.8816
No log 5.7 228 0.7802 0.4816 0.7802 0.8833
No log 5.75 230 0.7880 0.5028 0.7880 0.8877
No log 5.8 232 0.7647 0.4931 0.7647 0.8745
No log 5.85 234 0.7488 0.5052 0.7488 0.8653
No log 5.9 236 0.7668 0.4979 0.7668 0.8757
No log 5.95 238 0.7773 0.4995 0.7773 0.8816
No log 6.0 240 0.7477 0.5181 0.7477 0.8647
No log 6.05 242 0.7191 0.4816 0.7191 0.8480
No log 6.1 244 0.7372 0.4980 0.7372 0.8586
No log 6.15 246 0.7424 0.5006 0.7424 0.8616
No log 6.2 248 0.7643 0.4829 0.7643 0.8742
No log 6.25 250 0.8090 0.4625 0.8090 0.8995
No log 6.3 252 0.8352 0.4574 0.8352 0.9139
No log 6.35 254 0.8533 0.4711 0.8533 0.9238
No log 6.4 256 0.8617 0.4504 0.8617 0.9283
No log 6.45 258 0.8558 0.4524 0.8558 0.9251
No log 6.5 260 0.8633 0.4454 0.8633 0.9291
No log 6.55 262 0.8742 0.4351 0.8742 0.9350
No log 6.6 264 0.8727 0.4454 0.8727 0.9342
No log 6.65 266 0.8622 0.4471 0.8622 0.9286
No log 6.7 268 0.8602 0.4538 0.8602 0.9275
No log 6.75 270 0.8343 0.4518 0.8343 0.9134
No log 6.8 272 0.7939 0.4867 0.7939 0.8910
No log 6.85 274 0.7694 0.5019 0.7694 0.8772
No log 6.9 276 0.7484 0.5083 0.7484 0.8651
No log 6.95 278 0.8001 0.4459 0.8001 0.8945
No log 7.0 280 0.8678 0.4439 0.8678 0.9316
No log 7.05 282 0.8564 0.4326 0.8564 0.9254
No log 7.1 284 0.7825 0.4445 0.7825 0.8846
No log 7.15 286 0.7343 0.4713 0.7343 0.8569
No log 7.2 288 0.7300 0.4738 0.7300 0.8544
No log 7.25 290 0.7382 0.4572 0.7382 0.8592
No log 7.3 292 0.7622 0.4493 0.7622 0.8730
No log 7.35 294 0.7617 0.4490 0.7617 0.8728
No log 7.4 296 0.7797 0.4493 0.7797 0.8830
No log 7.45 298 0.7743 0.4549 0.7743 0.8800
No log 7.5 300 0.7575 0.4755 0.7575 0.8703
No log 7.55 302 0.7565 0.4718 0.7565 0.8698
No log 7.6 304 0.7726 0.4448 0.7726 0.8790
No log 7.65 306 0.8097 0.4338 0.8097 0.8998
No log 7.7 308 0.8084 0.4268 0.8084 0.8991
No log 7.75 310 0.7974 0.4541 0.7974 0.8930
No log 7.8 312 0.7940 0.4541 0.7940 0.8911
No log 7.85 314 0.8358 0.4303 0.8358 0.9142
No log 7.9 316 0.8403 0.4271 0.8403 0.9167
No log 7.95 318 0.7801 0.4198 0.7801 0.8832
No log 8.0 320 0.7314 0.4715 0.7314 0.8552
No log 8.05 322 0.7269 0.4045 0.7269 0.8526
No log 8.1 324 0.7758 0.4506 0.7758 0.8808
No log 8.15 326 0.7821 0.4398 0.7821 0.8844
No log 8.2 328 0.7566 0.4686 0.7566 0.8698
No log 8.25 330 0.7079 0.4508 0.7079 0.8413
No log 8.3 332 0.6834 0.4821 0.6834 0.8267
No log 8.35 334 0.6719 0.4836 0.6719 0.8197
No log 8.4 336 0.6791 0.4932 0.6791 0.8241
No log 8.45 338 0.7443 0.4704 0.7443 0.8627
No log 8.5 340 0.7918 0.4381 0.7918 0.8898
No log 8.55 342 0.7745 0.4521 0.7745 0.8800
No log 8.6 344 0.7258 0.4492 0.7258 0.8519
No log 8.65 346 0.7164 0.5077 0.7164 0.8464
No log 8.7 348 0.7343 0.4910 0.7343 0.8569
No log 8.75 350 0.7925 0.4338 0.7925 0.8902
No log 8.8 352 0.9319 0.4342 0.9319 0.9653
No log 8.85 354 0.9558 0.4374 0.9558 0.9776
No log 8.9 356 0.8651 0.4357 0.8651 0.9301
No log 8.95 358 0.7981 0.4530 0.7981 0.8934
No log 9.0 360 0.7908 0.4715 0.7908 0.8893
No log 9.05 362 0.8154 0.4613 0.8154 0.9030
No log 9.1 364 0.8125 0.4550 0.8125 0.9014
No log 9.15 366 0.7973 0.4621 0.7973 0.8929
No log 9.2 368 0.7866 0.4287 0.7866 0.8869
No log 9.25 370 0.7562 0.4813 0.7562 0.8696
No log 9.3 372 0.7399 0.4834 0.7399 0.8602
No log 9.35 374 0.7304 0.4710 0.7304 0.8546
No log 9.4 376 0.7313 0.4970 0.7313 0.8551
No log 9.45 378 0.7427 0.4914 0.7427 0.8618
No log 9.5 380 0.7757 0.4698 0.7757 0.8807
No log 9.55 382 0.8530 0.4273 0.8530 0.9236
No log 9.6 384 0.9924 0.4280 0.9924 0.9962
No log 9.65 386 1.0115 0.4054 1.0115 1.0057
No log 9.7 388 0.9150 0.4371 0.9150 0.9566
No log 9.75 390 0.8504 0.4305 0.8504 0.9222
No log 9.8 392 0.7995 0.4797 0.7995 0.8941
No log 9.85 394 0.8047 0.4992 0.8047 0.8971
No log 9.9 396 0.8191 0.4926 0.8191 0.9050
No log 9.95 398 0.8177 0.4866 0.8177 0.9043
No log 10.0 400 0.8312 0.4509 0.8312 0.9117
No log 10.05 402 0.8315 0.4405 0.8315 0.9118
No log 10.1 404 0.7920 0.5028 0.7920 0.8899
No log 10.15 406 0.7708 0.5123 0.7708 0.8780
No log 10.2 408 0.7770 0.4926 0.7770 0.8814
No log 10.25 410 0.7799 0.4710 0.7799 0.8831
No log 10.3 412 0.7993 0.4440 0.7993 0.8940
No log 10.35 414 0.7841 0.5183 0.7841 0.8855
No log 10.4 416 0.7909 0.5139 0.7909 0.8893
No log 10.45 418 0.8180 0.5099 0.8180 0.9044
No log 10.5 420 0.8816 0.4456 0.8816 0.9389
No log 10.55 422 0.9676 0.4322 0.9676 0.9837
No log 10.6 424 0.9730 0.4258 0.9730 0.9864
No log 10.65 426 0.9331 0.4400 0.9331 0.9660
No log 10.7 428 0.8868 0.5030 0.8868 0.9417
No log 10.75 430 0.8652 0.5313 0.8652 0.9301
No log 10.8 432 0.8356 0.5221 0.8356 0.9141
No log 10.85 434 0.8056 0.5231 0.8056 0.8975
No log 10.9 436 0.8053 0.5042 0.8053 0.8974
No log 10.95 438 0.8838 0.4591 0.8838 0.9401
No log 11.0 440 0.9437 0.4474 0.9437 0.9714
No log 11.05 442 0.8763 0.4620 0.8763 0.9361
No log 11.1 444 0.7613 0.4761 0.7613 0.8725
No log 11.15 446 0.7149 0.4820 0.7149 0.8455
No log 11.2 448 0.7149 0.5099 0.7149 0.8455
No log 11.25 450 0.7174 0.5106 0.7174 0.8470
No log 11.3 452 0.7284 0.5003 0.7284 0.8535
No log 11.35 454 0.7683 0.5129 0.7683 0.8765
No log 11.4 456 0.7913 0.5026 0.7913 0.8895
No log 11.45 458 0.8005 0.5055 0.8005 0.8947
No log 11.5 460 0.8193 0.5018 0.8193 0.9051
No log 11.55 462 0.8347 0.5018 0.8347 0.9136
No log 11.6 464 0.8915 0.4537 0.8915 0.9442
No log 11.65 466 0.9137 0.4665 0.9137 0.9559
No log 11.7 468 0.9097 0.4618 0.9097 0.9538
No log 11.75 470 0.8626 0.4591 0.8626 0.9287
No log 11.8 472 0.7941 0.5166 0.7941 0.8911
No log 11.85 474 0.7728 0.5151 0.7728 0.8791
No log 11.9 476 0.7607 0.5203 0.7607 0.8722
No log 11.95 478 0.7625 0.5281 0.7625 0.8732
No log 12.0 480 0.7569 0.5331 0.7569 0.8700
No log 12.05 482 0.7670 0.5170 0.7670 0.8758
No log 12.1 484 0.7932 0.4981 0.7932 0.8906
No log 12.15 486 0.7848 0.5117 0.7848 0.8859
No log 12.2 488 0.7718 0.5164 0.7718 0.8785
No log 12.25 490 0.7508 0.5275 0.7508 0.8665
No log 12.3 492 0.7420 0.5275 0.7420 0.8614
No log 12.35 494 0.7332 0.5275 0.7332 0.8563
No log 12.4 496 0.7180 0.5104 0.7180 0.8473
No log 12.45 498 0.7433 0.5086 0.7433 0.8621
0.5116 12.5 500 0.7594 0.5071 0.7594 0.8715
0.5116 12.55 502 0.7433 0.5237 0.7433 0.8621
0.5116 12.6 504 0.7287 0.5376 0.7287 0.8536
0.5116 12.65 506 0.7456 0.5283 0.7456 0.8635
0.5116 12.7 508 0.7570 0.5341 0.7570 0.8700
0.5116 12.75 510 0.7829 0.5106 0.7829 0.8848
0.5116 12.8 512 0.8634 0.4827 0.8634 0.9292
0.5116 12.85 514 0.8739 0.4600 0.8739 0.9348
0.5116 12.9 516 0.8202 0.5016 0.8202 0.9057
0.5116 12.95 518 0.7878 0.5095 0.7878 0.8876
0.5116 13.0 520 0.7826 0.5018 0.7826 0.8846
0.5116 13.05 522 0.7801 0.4936 0.7801 0.8832
0.5116 13.1 524 0.7765 0.4936 0.7765 0.8812
0.5116 13.15 526 0.7752 0.4997 0.7752 0.8805
0.5116 13.2 528 0.7776 0.4882 0.7776 0.8818

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.1+cu121
  • Datasets 3.2.0
  • Tokenizers 0.19.1
Downloads last month
1
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_run1_AugV5_k16_task2_organization

Finetuned
(4024)
this model