ArabicNewSplits6_FineTuningAraBERT_run2_AugV5_k11_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: 0.8017
  • Qwk: 0.6898
  • Mse: 0.8017
  • Rmse: 0.8954

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: 10

Training results

Training Loss Epoch Step Validation Loss Qwk Mse Rmse
No log 0.0488 2 2.4405 0.0285 2.4405 1.5622
No log 0.0976 4 1.6229 0.1548 1.6229 1.2739
No log 0.1463 6 1.5256 0.0987 1.5256 1.2351
No log 0.1951 8 1.5798 0.1766 1.5798 1.2569
No log 0.2439 10 1.4788 0.2350 1.4788 1.2161
No log 0.2927 12 1.6493 0.2655 1.6493 1.2842
No log 0.3415 14 1.6411 0.2848 1.6411 1.2811
No log 0.3902 16 1.4069 0.3448 1.4069 1.1861
No log 0.4390 18 1.1949 0.3287 1.1949 1.0931
No log 0.4878 20 1.1040 0.3787 1.1040 1.0507
No log 0.5366 22 1.0803 0.4117 1.0803 1.0394
No log 0.5854 24 1.0959 0.4181 1.0959 1.0469
No log 0.6341 26 1.1550 0.4580 1.1550 1.0747
No log 0.6829 28 1.1167 0.4997 1.1167 1.0567
No log 0.7317 30 1.0591 0.3972 1.0591 1.0291
No log 0.7805 32 0.9993 0.3985 0.9993 0.9996
No log 0.8293 34 0.9738 0.4625 0.9738 0.9868
No log 0.8780 36 0.9766 0.5263 0.9766 0.9882
No log 0.9268 38 1.0143 0.5651 1.0143 1.0071
No log 0.9756 40 0.9883 0.5141 0.9883 0.9941
No log 1.0244 42 1.0640 0.4783 1.0640 1.0315
No log 1.0732 44 1.1611 0.4841 1.1611 1.0776
No log 1.1220 46 1.2152 0.4901 1.2152 1.1024
No log 1.1707 48 1.1195 0.4757 1.1195 1.0581
No log 1.2195 50 1.0044 0.5041 1.0044 1.0022
No log 1.2683 52 0.9794 0.5318 0.9794 0.9896
No log 1.3171 54 0.9635 0.5951 0.9635 0.9816
No log 1.3659 56 0.9396 0.6192 0.9396 0.9693
No log 1.4146 58 0.9067 0.5883 0.9067 0.9522
No log 1.4634 60 0.9333 0.5486 0.9333 0.9661
No log 1.5122 62 1.0469 0.5360 1.0469 1.0232
No log 1.5610 64 1.0963 0.5551 1.0963 1.0471
No log 1.6098 66 1.1156 0.5664 1.1156 1.0562
No log 1.6585 68 1.0541 0.5316 1.0541 1.0267
No log 1.7073 70 1.0114 0.5123 1.0114 1.0057
No log 1.7561 72 1.0125 0.4867 1.0125 1.0062
No log 1.8049 74 1.0681 0.5134 1.0681 1.0335
No log 1.8537 76 1.1793 0.5399 1.1793 1.0860
No log 1.9024 78 1.1860 0.5318 1.1860 1.0891
No log 1.9512 80 1.0115 0.6056 1.0115 1.0058
No log 2.0 82 0.8683 0.5682 0.8683 0.9318
No log 2.0488 84 0.8137 0.6243 0.8137 0.9020
No log 2.0976 86 0.7924 0.6284 0.7924 0.8902
No log 2.1463 88 0.8094 0.6376 0.8094 0.8996
No log 2.1951 90 0.9142 0.6678 0.9142 0.9561
No log 2.2439 92 1.0692 0.6424 1.0692 1.0340
No log 2.2927 94 1.0409 0.6513 1.0409 1.0202
No log 2.3415 96 1.0123 0.6432 1.0123 1.0061
No log 2.3902 98 0.9733 0.6143 0.9733 0.9866
No log 2.4390 100 0.9434 0.6043 0.9434 0.9713
No log 2.4878 102 0.9042 0.6191 0.9042 0.9509
No log 2.5366 104 0.8993 0.5798 0.8993 0.9483
No log 2.5854 106 0.8813 0.5739 0.8813 0.9388
No log 2.6341 108 0.8072 0.6400 0.8072 0.8984
No log 2.6829 110 0.8376 0.6234 0.8376 0.9152
No log 2.7317 112 1.0638 0.6501 1.0638 1.0314
No log 2.7805 114 1.1967 0.5977 1.1967 1.0939
No log 2.8293 116 1.0761 0.6566 1.0761 1.0374
No log 2.8780 118 0.8415 0.6972 0.8415 0.9173
No log 2.9268 120 0.6858 0.7291 0.6858 0.8281
No log 2.9756 122 0.6601 0.7290 0.6601 0.8125
No log 3.0244 124 0.8000 0.7292 0.8000 0.8944
No log 3.0732 126 1.0615 0.6495 1.0615 1.0303
No log 3.1220 128 1.0856 0.6428 1.0856 1.0419
No log 3.1707 130 0.9763 0.6598 0.9763 0.9881
No log 3.2195 132 0.8531 0.7060 0.8531 0.9236
No log 3.2683 134 0.8712 0.7147 0.8712 0.9334
No log 3.3171 136 0.9950 0.6874 0.9950 0.9975
No log 3.3659 138 1.1842 0.6767 1.1842 1.0882
No log 3.4146 140 1.1983 0.6819 1.1983 1.0947
No log 3.4634 142 0.9763 0.6847 0.9763 0.9881
No log 3.5122 144 0.8502 0.6823 0.8502 0.9220
No log 3.5610 146 0.8546 0.6838 0.8546 0.9244
No log 3.6098 148 0.9020 0.6657 0.9020 0.9497
No log 3.6585 150 1.0714 0.6082 1.0714 1.0351
No log 3.7073 152 1.0448 0.6284 1.0448 1.0221
No log 3.7561 154 0.8997 0.6520 0.8997 0.9485
No log 3.8049 156 0.8063 0.7160 0.8063 0.8979
No log 3.8537 158 0.7162 0.6877 0.7162 0.8463
No log 3.9024 160 0.7204 0.7127 0.7204 0.8488
No log 3.9512 162 0.7571 0.7147 0.7571 0.8701
No log 4.0 164 0.8436 0.6762 0.8436 0.9185
No log 4.0488 166 0.8893 0.6554 0.8893 0.9430
No log 4.0976 168 0.9858 0.6381 0.9858 0.9929
No log 4.1463 170 1.0483 0.6347 1.0483 1.0239
No log 4.1951 172 0.9309 0.6473 0.9309 0.9648
No log 4.2439 174 0.7771 0.6900 0.7771 0.8815
No log 4.2927 176 0.7464 0.6897 0.7464 0.8639
No log 4.3415 178 0.7209 0.7083 0.7209 0.8491
No log 4.3902 180 0.7413 0.7085 0.7413 0.8610
No log 4.4390 182 0.8758 0.6604 0.8758 0.9358
No log 4.4878 184 0.9545 0.6553 0.9545 0.9770
No log 4.5366 186 0.8507 0.6855 0.8507 0.9224
No log 4.5854 188 0.7881 0.7012 0.7881 0.8877
No log 4.6341 190 0.7830 0.6957 0.7830 0.8849
No log 4.6829 192 0.8567 0.7037 0.8567 0.9256
No log 4.7317 194 0.9097 0.7040 0.9097 0.9538
No log 4.7805 196 0.8319 0.7123 0.8319 0.9121
No log 4.8293 198 0.7354 0.7499 0.7354 0.8576
No log 4.8780 200 0.7385 0.7496 0.7385 0.8594
No log 4.9268 202 0.8070 0.7359 0.8070 0.8984
No log 4.9756 204 0.8430 0.7173 0.8430 0.9182
No log 5.0244 206 0.9564 0.6793 0.9564 0.9780
No log 5.0732 208 1.0679 0.6264 1.0679 1.0334
No log 5.1220 210 1.0744 0.6239 1.0744 1.0365
No log 5.1707 212 0.9642 0.6817 0.9642 0.9819
No log 5.2195 214 0.9023 0.6884 0.9023 0.9499
No log 5.2683 216 0.8363 0.6979 0.8363 0.9145
No log 5.3171 218 0.8114 0.6979 0.8114 0.9008
No log 5.3659 220 0.7950 0.6989 0.7950 0.8916
No log 5.4146 222 0.7866 0.6797 0.7866 0.8869
No log 5.4634 224 0.7367 0.6979 0.7367 0.8583
No log 5.5122 226 0.7457 0.6944 0.7457 0.8636
No log 5.5610 228 0.8096 0.6712 0.8096 0.8998
No log 5.6098 230 0.9338 0.6460 0.9338 0.9663
No log 5.6585 232 1.1643 0.6178 1.1643 1.0790
No log 5.7073 234 1.3837 0.6008 1.3837 1.1763
No log 5.7561 236 1.3902 0.6008 1.3902 1.1790
No log 5.8049 238 1.2329 0.6232 1.2329 1.1104
No log 5.8537 240 1.0534 0.6364 1.0534 1.0263
No log 5.9024 242 0.8775 0.6720 0.8775 0.9368
No log 5.9512 244 0.8198 0.6600 0.8198 0.9055
No log 6.0 246 0.8106 0.6721 0.8106 0.9003
No log 6.0488 248 0.8556 0.6812 0.8556 0.9250
No log 6.0976 250 0.9316 0.6500 0.9316 0.9652
No log 6.1463 252 1.0218 0.6415 1.0218 1.0109
No log 6.1951 254 1.0075 0.6415 1.0075 1.0037
No log 6.2439 256 0.9557 0.6331 0.9557 0.9776
No log 6.2927 258 0.8442 0.6830 0.8442 0.9188
No log 6.3415 260 0.8197 0.7114 0.8197 0.9054
No log 6.3902 262 0.8910 0.6609 0.8910 0.9439
No log 6.4390 264 0.9408 0.6609 0.9408 0.9700
No log 6.4878 266 0.9412 0.6609 0.9412 0.9702
No log 6.5366 268 0.9222 0.6841 0.9222 0.9603
No log 6.5854 270 0.9526 0.6704 0.9526 0.9760
No log 6.6341 272 1.0019 0.6491 1.0019 1.0009
No log 6.6829 274 0.9952 0.6538 0.9952 0.9976
No log 6.7317 276 0.9799 0.6680 0.9799 0.9899
No log 6.7805 278 0.8870 0.6625 0.8870 0.9418
No log 6.8293 280 0.8050 0.6641 0.8050 0.8972
No log 6.8780 282 0.7534 0.6617 0.7534 0.8680
No log 6.9268 284 0.7573 0.6667 0.7573 0.8702
No log 6.9756 286 0.8019 0.6624 0.8019 0.8955
No log 7.0244 288 0.8071 0.6624 0.8071 0.8984
No log 7.0732 290 0.7806 0.6605 0.7806 0.8835
No log 7.1220 292 0.8014 0.6561 0.8014 0.8952
No log 7.1707 294 0.8675 0.6722 0.8675 0.9314
No log 7.2195 296 0.9354 0.6720 0.9354 0.9672
No log 7.2683 298 0.9901 0.6704 0.9901 0.9951
No log 7.3171 300 0.9775 0.6704 0.9775 0.9887
No log 7.3659 302 0.9420 0.6720 0.9420 0.9705
No log 7.4146 304 0.8898 0.6551 0.8898 0.9433
No log 7.4634 306 0.8090 0.6755 0.8090 0.8994
No log 7.5122 308 0.7761 0.6883 0.7761 0.8809
No log 7.5610 310 0.7538 0.6860 0.7538 0.8682
No log 7.6098 312 0.7369 0.7077 0.7369 0.8584
No log 7.6585 314 0.7379 0.7080 0.7379 0.8590
No log 7.7073 316 0.7707 0.6883 0.7707 0.8779
No log 7.7561 318 0.8225 0.6770 0.8225 0.9069
No log 7.8049 320 0.8493 0.6919 0.8493 0.9216
No log 7.8537 322 0.8794 0.6944 0.8794 0.9378
No log 7.9024 324 0.9032 0.6883 0.9032 0.9504
No log 7.9512 326 0.8897 0.6944 0.8897 0.9432
No log 8.0 328 0.8412 0.6988 0.8412 0.9171
No log 8.0488 330 0.7886 0.6929 0.7886 0.8881
No log 8.0976 332 0.7625 0.7008 0.7625 0.8732
No log 8.1463 334 0.7714 0.7065 0.7714 0.8783
No log 8.1951 336 0.8037 0.7002 0.8037 0.8965
No log 8.2439 338 0.8655 0.6884 0.8655 0.9303
No log 8.2927 340 0.9201 0.6815 0.9201 0.9592
No log 8.3415 342 0.9234 0.6815 0.9234 0.9610
No log 8.3902 344 0.9402 0.6727 0.9402 0.9696
No log 8.4390 346 0.9669 0.6585 0.9669 0.9833
No log 8.4878 348 0.9587 0.6585 0.9587 0.9791
No log 8.5366 350 0.9211 0.6877 0.9211 0.9597
No log 8.5854 352 0.8782 0.6813 0.8782 0.9371
No log 8.6341 354 0.8565 0.6813 0.8565 0.9255
No log 8.6829 356 0.8467 0.6907 0.8467 0.9201
No log 8.7317 358 0.8411 0.6907 0.8411 0.9171
No log 8.7805 360 0.8379 0.6772 0.8379 0.9154
No log 8.8293 362 0.8442 0.6972 0.8442 0.9188
No log 8.8780 364 0.8442 0.6972 0.8442 0.9188
No log 8.9268 366 0.8415 0.6972 0.8415 0.9173
No log 8.9756 368 0.8405 0.6972 0.8405 0.9168
No log 9.0244 370 0.8488 0.6972 0.8488 0.9213
No log 9.0732 372 0.8509 0.6972 0.8509 0.9225
No log 9.1220 374 0.8552 0.6972 0.8552 0.9248
No log 9.1707 376 0.8637 0.6878 0.8637 0.9293
No log 9.2195 378 0.8779 0.6878 0.8779 0.9370
No log 9.2683 380 0.8892 0.6878 0.8892 0.9430
No log 9.3171 382 0.8923 0.6884 0.8923 0.9446
No log 9.3659 384 0.8998 0.6884 0.8998 0.9486
No log 9.4146 386 0.8939 0.6884 0.8939 0.9455
No log 9.4634 388 0.8890 0.6884 0.8890 0.9428
No log 9.5122 390 0.8742 0.6878 0.8742 0.9350
No log 9.5610 392 0.8521 0.6745 0.8521 0.9231
No log 9.6098 394 0.8372 0.6904 0.8372 0.9150
No log 9.6585 396 0.8282 0.6904 0.8282 0.9100
No log 9.7073 398 0.8236 0.6904 0.8236 0.9075
No log 9.7561 400 0.8193 0.6898 0.8193 0.9051
No log 9.8049 402 0.8135 0.6898 0.8135 0.9019
No log 9.8537 404 0.8089 0.6898 0.8089 0.8994
No log 9.9024 406 0.8051 0.6898 0.8051 0.8973
No log 9.9512 408 0.8026 0.6898 0.8026 0.8959
No log 10.0 410 0.8017 0.6898 0.8017 0.8954

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/ArabicNewSplits6_FineTuningAraBERT_run2_AugV5_k11_task5_organization

Finetuned
(4040)
this model