ArabicNewSplits8_usingALLEssays_FineTuningAraBERT_run1_AugV5_k20_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: 0.9740
  • Qwk: 0.5327
  • Mse: 0.9740
  • Rmse: 0.9869

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 0.02 2 5.3139 -0.0034 5.3139 2.3052
No log 0.04 4 3.1468 0.0583 3.1468 1.7739
No log 0.06 6 2.0035 0.0531 2.0035 1.4154
No log 0.08 8 2.1891 -0.0327 2.1891 1.4795
No log 0.1 10 2.1307 -0.0761 2.1307 1.4597
No log 0.12 12 1.4639 0.0688 1.4639 1.2099
No log 0.14 14 1.3357 0.1729 1.3357 1.1557
No log 0.16 16 1.3496 0.1613 1.3496 1.1617
No log 0.18 18 1.4729 -0.0753 1.4729 1.2136
No log 0.2 20 1.7438 -0.0622 1.7438 1.3205
No log 0.22 22 1.7243 -0.0543 1.7243 1.3131
No log 0.24 24 1.8283 -0.0348 1.8283 1.3521
No log 0.26 26 1.6265 0.0123 1.6265 1.2753
No log 0.28 28 1.3250 0.1549 1.3250 1.1511
No log 0.3 30 1.3143 0.1118 1.3143 1.1464
No log 0.32 32 1.2858 0.1529 1.2858 1.1339
No log 0.34 34 1.2886 0.1391 1.2886 1.1352
No log 0.36 36 1.3526 0.1359 1.3526 1.1630
No log 0.38 38 1.5452 0.1352 1.5452 1.2431
No log 0.4 40 1.8211 0.1806 1.8211 1.3495
No log 0.42 42 2.0943 0.1334 2.0943 1.4472
No log 0.44 44 2.0226 0.1519 2.0226 1.4222
No log 0.46 46 1.5786 0.1710 1.5786 1.2564
No log 0.48 48 1.2527 0.3063 1.2527 1.1192
No log 0.5 50 1.1543 0.3076 1.1543 1.0744
No log 0.52 52 1.1921 0.3410 1.1921 1.0918
No log 0.54 54 1.2459 0.3356 1.2459 1.1162
No log 0.56 56 1.1812 0.3723 1.1812 1.0868
No log 0.58 58 1.1259 0.4098 1.1259 1.0611
No log 0.6 60 1.0789 0.4053 1.0789 1.0387
No log 0.62 62 0.9947 0.4111 0.9947 0.9974
No log 0.64 64 0.9941 0.4222 0.9941 0.9970
No log 0.66 66 1.0015 0.4082 1.0015 1.0008
No log 0.68 68 1.2656 0.3865 1.2656 1.1250
No log 0.7 70 2.1521 0.2710 2.1521 1.4670
No log 0.72 72 2.4282 0.2203 2.4282 1.5583
No log 0.74 74 1.8986 0.2775 1.8986 1.3779
No log 0.76 76 1.0887 0.4366 1.0887 1.0434
No log 0.78 78 0.8539 0.5424 0.8539 0.9241
No log 0.8 80 0.9016 0.4464 0.9016 0.9495
No log 0.82 82 0.8547 0.5144 0.8547 0.9245
No log 0.84 84 1.0499 0.4252 1.0499 1.0246
No log 0.86 86 1.2250 0.3514 1.2250 1.1068
No log 0.88 88 1.4931 0.3493 1.4931 1.2219
No log 0.9 90 1.3876 0.4246 1.3876 1.1780
No log 0.92 92 1.3102 0.4423 1.3102 1.1446
No log 0.94 94 1.1040 0.4308 1.1040 1.0507
No log 0.96 96 0.9657 0.4594 0.9657 0.9827
No log 0.98 98 0.9053 0.4834 0.9053 0.9515
No log 1.0 100 1.0178 0.4295 1.0178 1.0089
No log 1.02 102 1.0065 0.4051 1.0065 1.0032
No log 1.04 104 0.8596 0.5342 0.8596 0.9271
No log 1.06 106 0.7433 0.5956 0.7433 0.8621
No log 1.08 108 0.7555 0.6226 0.7555 0.8692
No log 1.1 110 0.7632 0.6271 0.7632 0.8736
No log 1.12 112 0.7289 0.5981 0.7289 0.8538
No log 1.1400 114 0.7745 0.5547 0.7745 0.8801
No log 1.16 116 0.8771 0.5210 0.8771 0.9365
No log 1.18 118 0.8456 0.5145 0.8456 0.9196
No log 1.2 120 0.7879 0.5696 0.7879 0.8877
No log 1.22 122 0.7838 0.5827 0.7838 0.8853
No log 1.24 124 0.8913 0.5560 0.8913 0.9441
No log 1.26 126 0.8782 0.5447 0.8782 0.9371
No log 1.28 128 0.8355 0.5493 0.8355 0.9140
No log 1.3 130 0.8372 0.5509 0.8372 0.9150
No log 1.32 132 0.8317 0.5775 0.8317 0.9120
No log 1.34 134 0.8164 0.5834 0.8164 0.9036
No log 1.3600 136 0.7795 0.6421 0.7795 0.8829
No log 1.38 138 0.7645 0.6720 0.7645 0.8744
No log 1.4 140 0.7483 0.6890 0.7483 0.8650
No log 1.42 142 0.7523 0.6272 0.7523 0.8674
No log 1.44 144 0.7858 0.6030 0.7858 0.8865
No log 1.46 146 0.7356 0.6743 0.7356 0.8576
No log 1.48 148 0.7865 0.6034 0.7865 0.8868
No log 1.5 150 0.9166 0.5979 0.9166 0.9574
No log 1.52 152 1.0735 0.5705 1.0735 1.0361
No log 1.54 154 0.8626 0.5815 0.8626 0.9288
No log 1.56 156 0.7620 0.6348 0.7620 0.8729
No log 1.58 158 0.9780 0.5148 0.9780 0.9890
No log 1.6 160 1.2053 0.4492 1.2053 1.0979
No log 1.62 162 1.2979 0.4453 1.2979 1.1392
No log 1.6400 164 1.0841 0.4638 1.0841 1.0412
No log 1.6600 166 0.9735 0.5105 0.9735 0.9866
No log 1.6800 168 0.9033 0.5891 0.9033 0.9504
No log 1.7 170 0.9518 0.5597 0.9518 0.9756
No log 1.72 172 0.8497 0.5914 0.8497 0.9218
No log 1.74 174 0.8221 0.6255 0.8221 0.9067
No log 1.76 176 0.8133 0.6153 0.8133 0.9018
No log 1.78 178 0.8200 0.6251 0.8200 0.9056
No log 1.8 180 0.7950 0.6251 0.7950 0.8916
No log 1.8200 182 0.7398 0.6748 0.7398 0.8601
No log 1.8400 184 0.7168 0.6710 0.7168 0.8466
No log 1.8600 186 0.7172 0.6657 0.7172 0.8469
No log 1.88 188 0.7544 0.5922 0.7544 0.8686
No log 1.9 190 0.7425 0.5663 0.7425 0.8617
No log 1.92 192 0.7203 0.6130 0.7203 0.8487
No log 1.94 194 0.7392 0.6195 0.7392 0.8598
No log 1.96 196 0.7104 0.6621 0.7104 0.8429
No log 1.98 198 0.6839 0.6376 0.6839 0.8270
No log 2.0 200 0.6835 0.6833 0.6835 0.8268
No log 2.02 202 0.7427 0.6565 0.7427 0.8618
No log 2.04 204 0.7873 0.6791 0.7873 0.8873
No log 2.06 206 0.8030 0.6182 0.8030 0.8961
No log 2.08 208 0.7816 0.6155 0.7816 0.8841
No log 2.1 210 0.8424 0.6176 0.8424 0.9178
No log 2.12 212 0.9997 0.5011 0.9997 0.9998
No log 2.14 214 0.9638 0.5268 0.9638 0.9817
No log 2.16 216 0.8595 0.5750 0.8595 0.9271
No log 2.18 218 0.7900 0.6094 0.7900 0.8888
No log 2.2 220 0.7919 0.5842 0.7919 0.8899
No log 2.22 222 0.7754 0.5628 0.7754 0.8806
No log 2.24 224 0.7931 0.5893 0.7931 0.8905
No log 2.26 226 0.8366 0.5793 0.8366 0.9147
No log 2.2800 228 0.7970 0.5977 0.7970 0.8928
No log 2.3 230 0.7490 0.6190 0.7490 0.8655
No log 2.32 232 0.7657 0.6014 0.7657 0.8750
No log 2.34 234 0.7323 0.6331 0.7323 0.8558
No log 2.36 236 0.7460 0.6359 0.7460 0.8637
No log 2.38 238 0.7788 0.6338 0.7788 0.8825
No log 2.4 240 0.7723 0.5989 0.7723 0.8788
No log 2.42 242 0.7890 0.6220 0.7890 0.8882
No log 2.44 244 0.8279 0.5735 0.8279 0.9099
No log 2.46 246 0.7674 0.6170 0.7674 0.8760
No log 2.48 248 0.6982 0.6265 0.6982 0.8356
No log 2.5 250 0.8427 0.5789 0.8427 0.9180
No log 2.52 252 0.9135 0.5549 0.9135 0.9558
No log 2.54 254 0.8158 0.5977 0.8158 0.9032
No log 2.56 256 0.7719 0.6020 0.7719 0.8786
No log 2.58 258 0.7663 0.5979 0.7663 0.8754
No log 2.6 260 0.7291 0.5954 0.7291 0.8539
No log 2.62 262 0.7131 0.6095 0.7131 0.8444
No log 2.64 264 0.7137 0.6310 0.7137 0.8448
No log 2.66 266 0.7144 0.6338 0.7144 0.8452
No log 2.68 268 0.7082 0.6581 0.7082 0.8415
No log 2.7 270 0.7059 0.6571 0.7059 0.8402
No log 2.7200 272 0.6897 0.6610 0.6897 0.8305
No log 2.74 274 0.6803 0.6523 0.6803 0.8248
No log 2.76 276 0.7077 0.6360 0.7077 0.8413
No log 2.7800 278 0.7663 0.6235 0.7663 0.8754
No log 2.8 280 0.7630 0.6415 0.7630 0.8735
No log 2.82 282 0.7466 0.6554 0.7466 0.8641
No log 2.84 284 0.6929 0.6061 0.6929 0.8324
No log 2.86 286 0.7197 0.6146 0.7197 0.8484
No log 2.88 288 0.8197 0.6079 0.8197 0.9054
No log 2.9 290 0.9349 0.5798 0.9349 0.9669
No log 2.92 292 0.8407 0.5876 0.8407 0.9169
No log 2.94 294 0.6946 0.6111 0.6946 0.8334
No log 2.96 296 0.7286 0.6140 0.7286 0.8536
No log 2.98 298 0.8425 0.5751 0.8425 0.9179
No log 3.0 300 0.9327 0.5449 0.9327 0.9658
No log 3.02 302 0.7741 0.5957 0.7741 0.8798
No log 3.04 304 0.6229 0.6537 0.6229 0.7893
No log 3.06 306 0.7205 0.6615 0.7205 0.8488
No log 3.08 308 0.7895 0.6289 0.7895 0.8885
No log 3.1 310 0.6890 0.6613 0.6890 0.8301
No log 3.12 312 0.6361 0.6417 0.6361 0.7975
No log 3.14 314 0.6945 0.6375 0.6945 0.8333
No log 3.16 316 0.7220 0.5997 0.7220 0.8497
No log 3.18 318 0.6978 0.6073 0.6978 0.8353
No log 3.2 320 0.6686 0.5980 0.6686 0.8177
No log 3.22 322 0.6869 0.6050 0.6869 0.8288
No log 3.24 324 0.7050 0.6403 0.7050 0.8397
No log 3.26 326 0.7033 0.6029 0.7033 0.8386
No log 3.2800 328 0.7212 0.6392 0.7212 0.8492
No log 3.3 330 0.7762 0.6248 0.7762 0.8810
No log 3.32 332 0.7753 0.6229 0.7753 0.8805
No log 3.34 334 0.7207 0.5840 0.7207 0.8489
No log 3.36 336 0.7076 0.5790 0.7076 0.8412
No log 3.38 338 0.7051 0.6080 0.7051 0.8397
No log 3.4 340 0.7196 0.6061 0.7196 0.8483
No log 3.42 342 0.7287 0.6251 0.7287 0.8536
No log 3.44 344 0.7115 0.6187 0.7115 0.8435
No log 3.46 346 0.6897 0.6352 0.6897 0.8305
No log 3.48 348 0.6916 0.6134 0.6916 0.8317
No log 3.5 350 0.7085 0.6281 0.7085 0.8417
No log 3.52 352 0.6953 0.6323 0.6953 0.8339
No log 3.54 354 0.6604 0.6850 0.6604 0.8126
No log 3.56 356 0.6566 0.6871 0.6566 0.8103
No log 3.58 358 0.6528 0.6670 0.6528 0.8080
No log 3.6 360 0.6731 0.6894 0.6731 0.8204
No log 3.62 362 0.7555 0.6712 0.7555 0.8692
No log 3.64 364 0.7588 0.6779 0.7588 0.8711
No log 3.66 366 0.6981 0.6638 0.6981 0.8355
No log 3.68 368 0.7144 0.6772 0.7144 0.8452
No log 3.7 370 0.7126 0.6641 0.7126 0.8441
No log 3.7200 372 0.6888 0.6518 0.6888 0.8299
No log 3.74 374 0.7614 0.5994 0.7614 0.8726
No log 3.76 376 0.7680 0.5882 0.7680 0.8763
No log 3.7800 378 0.6907 0.6471 0.6907 0.8311
No log 3.8 380 0.7219 0.6465 0.7219 0.8496
No log 3.82 382 0.7792 0.6446 0.7792 0.8827
No log 3.84 384 0.7713 0.6306 0.7713 0.8782
No log 3.86 386 0.7512 0.6496 0.7512 0.8667
No log 3.88 388 0.6989 0.6545 0.6989 0.8360
No log 3.9 390 0.6786 0.6389 0.6786 0.8238
No log 3.92 392 0.6608 0.6469 0.6608 0.8129
No log 3.94 394 0.6542 0.6888 0.6542 0.8088
No log 3.96 396 0.7679 0.7062 0.7679 0.8763
No log 3.98 398 1.1099 0.6200 1.1099 1.0535
No log 4.0 400 1.1479 0.6097 1.1479 1.0714
No log 4.02 402 0.9085 0.6451 0.9085 0.9532
No log 4.04 404 0.6637 0.7065 0.6637 0.8147
No log 4.06 406 0.6315 0.7068 0.6315 0.7947
No log 4.08 408 0.6629 0.6720 0.6629 0.8142
No log 4.1 410 0.6719 0.6353 0.6719 0.8197
No log 4.12 412 0.6923 0.5570 0.6923 0.8321
No log 4.14 414 0.7934 0.6056 0.7934 0.8907
No log 4.16 416 1.0365 0.5293 1.0365 1.0181
No log 4.18 418 1.1663 0.4895 1.1663 1.0799
No log 4.2 420 1.0673 0.5538 1.0673 1.0331
No log 4.22 422 0.7919 0.6265 0.7919 0.8899
No log 4.24 424 0.6665 0.6769 0.6665 0.8164
No log 4.26 426 0.6440 0.6777 0.6440 0.8025
No log 4.28 428 0.6495 0.6891 0.6495 0.8059
No log 4.3 430 0.6523 0.7001 0.6523 0.8076
No log 4.32 432 0.6634 0.6701 0.6634 0.8145
No log 4.34 434 0.6663 0.6767 0.6663 0.8163
No log 4.36 436 0.6922 0.6237 0.6922 0.8320
No log 4.38 438 0.7008 0.6243 0.7008 0.8371
No log 4.4 440 0.6809 0.6291 0.6809 0.8252
No log 4.42 442 0.6696 0.6250 0.6696 0.8183
No log 4.44 444 0.7001 0.6705 0.7001 0.8367
No log 4.46 446 0.7869 0.6685 0.7869 0.8871
No log 4.48 448 0.7616 0.6785 0.7616 0.8727
No log 4.5 450 0.6682 0.6839 0.6682 0.8174
No log 4.52 452 0.6554 0.6746 0.6554 0.8096
No log 4.54 454 0.7067 0.6633 0.7067 0.8406
No log 4.5600 456 0.8048 0.6198 0.8048 0.8971
No log 4.58 458 0.7998 0.6233 0.7998 0.8943
No log 4.6 460 0.8252 0.6092 0.8252 0.9084
No log 4.62 462 0.7800 0.6366 0.7800 0.8832
No log 4.64 464 0.6843 0.6288 0.6843 0.8272
No log 4.66 466 0.6498 0.6789 0.6498 0.8061
No log 4.68 468 0.6623 0.6394 0.6623 0.8138
No log 4.7 470 0.7332 0.6358 0.7332 0.8562
No log 4.72 472 0.8492 0.6309 0.8492 0.9215
No log 4.74 474 0.9065 0.5717 0.9065 0.9521
No log 4.76 476 0.8183 0.5982 0.8183 0.9046
No log 4.78 478 0.6890 0.6462 0.6890 0.8301
No log 4.8 480 0.6663 0.6476 0.6663 0.8163
No log 4.82 482 0.6722 0.6356 0.6722 0.8198
No log 4.84 484 0.6947 0.6150 0.6947 0.8335
No log 4.86 486 0.7273 0.6239 0.7273 0.8528
No log 4.88 488 0.7676 0.6354 0.7676 0.8762
No log 4.9 490 0.7701 0.6418 0.7701 0.8776
No log 4.92 492 0.7204 0.6446 0.7204 0.8488
No log 4.9400 494 0.7138 0.6427 0.7138 0.8449
No log 4.96 496 0.6946 0.6414 0.6946 0.8334
No log 4.98 498 0.7284 0.6715 0.7284 0.8535
0.4589 5.0 500 0.8391 0.6034 0.8391 0.9160
0.4589 5.02 502 0.9217 0.5740 0.9217 0.9601
0.4589 5.04 504 0.9379 0.5583 0.9379 0.9685
0.4589 5.06 506 0.8488 0.6106 0.8488 0.9213
0.4589 5.08 508 0.7816 0.6288 0.7816 0.8841
0.4589 5.1 510 0.7589 0.6715 0.7589 0.8712
0.4589 5.12 512 0.7987 0.6498 0.7987 0.8937
0.4589 5.14 514 0.8874 0.6240 0.8874 0.9420
0.4589 5.16 516 0.8970 0.6181 0.8970 0.9471
0.4589 5.18 518 0.8700 0.6197 0.8700 0.9327
0.4589 5.2 520 0.9011 0.6087 0.9011 0.9492
0.4589 5.22 522 0.9233 0.6117 0.9233 0.9609
0.4589 5.24 524 0.9383 0.5656 0.9383 0.9686
0.4589 5.26 526 0.9356 0.5606 0.9356 0.9672
0.4589 5.28 528 0.9748 0.5563 0.9748 0.9873
0.4589 5.3 530 0.9621 0.5666 0.9621 0.9809
0.4589 5.32 532 0.9740 0.5327 0.9740 0.9869

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0+cu118
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
4
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/ArabicNewSplits8_usingALLEssays_FineTuningAraBERT_run1_AugV5_k20_task1_organization

Finetuned
(4040)
this model