ArabicNewSplits6_FineTuningAraBERTFreeze_run2_AugV5_k6_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: 0.6901
  • Qwk: 0.5255
  • Mse: 0.6901
  • Rmse: 0.8307

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 0.125 2 6.3156 -0.0169 6.3156 2.5131
No log 0.25 4 4.2756 -0.0288 4.2756 2.0678
No log 0.375 6 3.1662 -0.0070 3.1662 1.7794
No log 0.5 8 2.3416 0.0193 2.3416 1.5302
No log 0.625 10 1.7284 0.0263 1.7284 1.3147
No log 0.75 12 1.3072 0.0046 1.3072 1.1433
No log 0.875 14 1.1099 0.0119 1.1099 1.0535
No log 1.0 16 0.9063 0.1169 0.9063 0.9520
No log 1.125 18 0.7398 0.0883 0.7398 0.8601
No log 1.25 20 0.6908 0.2375 0.6908 0.8312
No log 1.375 22 0.6998 0.2048 0.6998 0.8366
No log 1.5 24 0.7257 0.1306 0.7257 0.8519
No log 1.625 26 0.8001 0.1449 0.8001 0.8945
No log 1.75 28 0.9208 0.0936 0.9208 0.9596
No log 1.875 30 1.0053 0.0364 1.0053 1.0026
No log 2.0 32 0.9970 0.0683 0.9970 0.9985
No log 2.125 34 0.9094 0.2178 0.9094 0.9536
No log 2.25 36 0.7967 0.2101 0.7967 0.8926
No log 2.375 38 0.7103 0.1908 0.7103 0.8428
No log 2.5 40 0.6593 0.3149 0.6593 0.8119
No log 2.625 42 0.6520 0.3392 0.6520 0.8075
No log 2.75 44 0.6473 0.3230 0.6472 0.8045
No log 2.875 46 0.6408 0.3114 0.6408 0.8005
No log 3.0 48 0.6589 0.2752 0.6589 0.8117
No log 3.125 50 0.7143 0.2830 0.7143 0.8451
No log 3.25 52 0.7476 0.2681 0.7476 0.8646
No log 3.375 54 0.7801 0.2667 0.7801 0.8832
No log 3.5 56 0.7606 0.2169 0.7606 0.8721
No log 3.625 58 0.7356 0.2425 0.7356 0.8577
No log 3.75 60 0.7126 0.3014 0.7126 0.8441
No log 3.875 62 0.7138 0.2572 0.7138 0.8449
No log 4.0 64 0.6941 0.2830 0.6941 0.8331
No log 4.125 66 0.6365 0.3736 0.6365 0.7978
No log 4.25 68 0.6004 0.4349 0.6004 0.7749
No log 4.375 70 0.5820 0.4346 0.5820 0.7629
No log 4.5 72 0.6297 0.4346 0.6297 0.7935
No log 4.625 74 0.6941 0.3839 0.6941 0.8331
No log 4.75 76 0.6611 0.4123 0.6611 0.8131
No log 4.875 78 0.5996 0.4510 0.5996 0.7744
No log 5.0 80 0.6556 0.3218 0.6556 0.8097
No log 5.125 82 0.8232 0.2452 0.8232 0.9073
No log 5.25 84 0.8898 0.2443 0.8898 0.9433
No log 5.375 86 0.7865 0.2773 0.7865 0.8869
No log 5.5 88 0.6095 0.3984 0.6095 0.7807
No log 5.625 90 0.5311 0.4519 0.5311 0.7288
No log 5.75 92 0.5304 0.4510 0.5304 0.7283
No log 5.875 94 0.5609 0.4236 0.5609 0.7489
No log 6.0 96 0.6415 0.4295 0.6415 0.8009
No log 6.125 98 0.7412 0.3666 0.7412 0.8610
No log 6.25 100 0.7311 0.3727 0.7311 0.8550
No log 6.375 102 0.7009 0.4250 0.7009 0.8372
No log 6.5 104 0.6380 0.4761 0.6380 0.7987
No log 6.625 106 0.5759 0.4756 0.5759 0.7589
No log 6.75 108 0.5177 0.4681 0.5177 0.7195
No log 6.875 110 0.4995 0.4895 0.4995 0.7068
No log 7.0 112 0.4953 0.5039 0.4953 0.7038
No log 7.125 114 0.5081 0.4789 0.5081 0.7128
No log 7.25 116 0.5423 0.5093 0.5423 0.7364
No log 7.375 118 0.6084 0.4817 0.6084 0.7800
No log 7.5 120 0.6857 0.4789 0.6857 0.8280
No log 7.625 122 0.6730 0.4541 0.6730 0.8204
No log 7.75 124 0.5946 0.5123 0.5946 0.7711
No log 7.875 126 0.5598 0.4849 0.5598 0.7482
No log 8.0 128 0.5688 0.4560 0.5688 0.7542
No log 8.125 130 0.5786 0.4647 0.5786 0.7607
No log 8.25 132 0.6110 0.3754 0.6110 0.7817
No log 8.375 134 0.5705 0.4777 0.5705 0.7553
No log 8.5 136 0.5389 0.4502 0.5389 0.7341
No log 8.625 138 0.5185 0.4536 0.5185 0.7201
No log 8.75 140 0.5181 0.5350 0.5181 0.7198
No log 8.875 142 0.5431 0.4739 0.5431 0.7369
No log 9.0 144 0.5782 0.5437 0.5782 0.7604
No log 9.125 146 0.6210 0.5306 0.6210 0.7880
No log 9.25 148 0.6011 0.5373 0.6011 0.7753
No log 9.375 150 0.5572 0.5 0.5572 0.7464
No log 9.5 152 0.5399 0.5051 0.5399 0.7348
No log 9.625 154 0.5402 0.5078 0.5402 0.7350
No log 9.75 156 0.5425 0.5078 0.5425 0.7366
No log 9.875 158 0.5428 0.5208 0.5428 0.7368
No log 10.0 160 0.5446 0.5274 0.5446 0.7380
No log 10.125 162 0.5452 0.5121 0.5452 0.7384
No log 10.25 164 0.5469 0.4993 0.5469 0.7395
No log 10.375 166 0.5498 0.5232 0.5498 0.7415
No log 10.5 168 0.5499 0.5498 0.5499 0.7415
No log 10.625 170 0.5641 0.5007 0.5641 0.7510
No log 10.75 172 0.5761 0.5085 0.5761 0.7590
No log 10.875 174 0.5604 0.4929 0.5604 0.7486
No log 11.0 176 0.5500 0.5304 0.5500 0.7416
No log 11.125 178 0.5622 0.5073 0.5622 0.7498
No log 11.25 180 0.5741 0.5010 0.5741 0.7577
No log 11.375 182 0.5731 0.5292 0.5731 0.7570
No log 11.5 184 0.5712 0.5618 0.5712 0.7558
No log 11.625 186 0.5725 0.4879 0.5725 0.7567
No log 11.75 188 0.5748 0.5033 0.5748 0.7582
No log 11.875 190 0.5784 0.4987 0.5784 0.7605
No log 12.0 192 0.5840 0.5394 0.5840 0.7642
No log 12.125 194 0.5858 0.5013 0.5858 0.7654
No log 12.25 196 0.5917 0.5221 0.5917 0.7692
No log 12.375 198 0.5912 0.5308 0.5912 0.7689
No log 12.5 200 0.5910 0.5630 0.5910 0.7688
No log 12.625 202 0.5894 0.5098 0.5894 0.7677
No log 12.75 204 0.6002 0.5260 0.6002 0.7747
No log 12.875 206 0.6420 0.4640 0.6420 0.8012
No log 13.0 208 0.6317 0.4926 0.6317 0.7948
No log 13.125 210 0.6062 0.5487 0.6062 0.7786
No log 13.25 212 0.5984 0.5116 0.5984 0.7735
No log 13.375 214 0.5911 0.5318 0.5911 0.7688
No log 13.5 216 0.5980 0.5208 0.5980 0.7733
No log 13.625 218 0.6145 0.5211 0.6145 0.7839
No log 13.75 220 0.6436 0.5490 0.6436 0.8023
No log 13.875 222 0.6385 0.5144 0.6385 0.7991
No log 14.0 224 0.6262 0.5155 0.6262 0.7913
No log 14.125 226 0.6319 0.5173 0.6319 0.7949
No log 14.25 228 0.6413 0.5197 0.6413 0.8008
No log 14.375 230 0.6589 0.5177 0.6589 0.8117
No log 14.5 232 0.6584 0.5147 0.6584 0.8114
No log 14.625 234 0.6483 0.5519 0.6483 0.8052
No log 14.75 236 0.6556 0.5577 0.6556 0.8097
No log 14.875 238 0.6642 0.5368 0.6642 0.8150
No log 15.0 240 0.6584 0.5052 0.6584 0.8114
No log 15.125 242 0.6419 0.5363 0.6419 0.8012
No log 15.25 244 0.6375 0.5197 0.6375 0.7984
No log 15.375 246 0.6392 0.4913 0.6392 0.7995
No log 15.5 248 0.6301 0.5029 0.6301 0.7938
No log 15.625 250 0.6283 0.5566 0.6283 0.7926
No log 15.75 252 0.6402 0.5328 0.6402 0.8001
No log 15.875 254 0.6358 0.5397 0.6358 0.7974
No log 16.0 256 0.6149 0.5520 0.6149 0.7841
No log 16.125 258 0.6052 0.5520 0.6052 0.7779
No log 16.25 260 0.6060 0.5460 0.6060 0.7785
No log 16.375 262 0.6106 0.5595 0.6106 0.7814
No log 16.5 264 0.6200 0.5683 0.6200 0.7874
No log 16.625 266 0.6308 0.5669 0.6308 0.7942
No log 16.75 268 0.6524 0.5448 0.6524 0.8077
No log 16.875 270 0.6619 0.5203 0.6619 0.8136
No log 17.0 272 0.6700 0.5217 0.6700 0.8186
No log 17.125 274 0.6770 0.5489 0.6770 0.8228
No log 17.25 276 0.6845 0.5429 0.6845 0.8273
No log 17.375 278 0.6862 0.5358 0.6862 0.8284
No log 17.5 280 0.6832 0.5338 0.6832 0.8265
No log 17.625 282 0.6691 0.5607 0.6691 0.8180
No log 17.75 284 0.6621 0.5319 0.6621 0.8137
No log 17.875 286 0.6569 0.5189 0.6569 0.8105
No log 18.0 288 0.6531 0.5506 0.6531 0.8081
No log 18.125 290 0.6563 0.5709 0.6563 0.8101
No log 18.25 292 0.6765 0.5729 0.6765 0.8225
No log 18.375 294 0.6953 0.5900 0.6953 0.8338
No log 18.5 296 0.6837 0.5913 0.6837 0.8268
No log 18.625 298 0.6601 0.5774 0.6601 0.8124
No log 18.75 300 0.6533 0.5260 0.6533 0.8083
No log 18.875 302 0.6525 0.5116 0.6525 0.8078
No log 19.0 304 0.6507 0.5179 0.6507 0.8067
No log 19.125 306 0.6468 0.5652 0.6468 0.8043
No log 19.25 308 0.6469 0.5837 0.6469 0.8043
No log 19.375 310 0.6533 0.5823 0.6533 0.8083
No log 19.5 312 0.6579 0.5562 0.6579 0.8111
No log 19.625 314 0.6765 0.5620 0.6765 0.8225
No log 19.75 316 0.6935 0.5260 0.6935 0.8328
No log 19.875 318 0.7124 0.5167 0.7124 0.8440
No log 20.0 320 0.7147 0.5058 0.7147 0.8454
No log 20.125 322 0.7104 0.5362 0.7104 0.8428
No log 20.25 324 0.7131 0.5775 0.7131 0.8445
No log 20.375 326 0.7027 0.5545 0.7027 0.8383
No log 20.5 328 0.6952 0.5882 0.6952 0.8338
No log 20.625 330 0.6887 0.5727 0.6887 0.8299
No log 20.75 332 0.6774 0.5787 0.6774 0.8231
No log 20.875 334 0.6794 0.5768 0.6794 0.8242
No log 21.0 336 0.6918 0.5176 0.6918 0.8317
No log 21.125 338 0.7170 0.4850 0.7170 0.8468
No log 21.25 340 0.7338 0.4850 0.7338 0.8566
No log 21.375 342 0.7577 0.5788 0.7577 0.8705
No log 21.5 344 0.7508 0.5941 0.7508 0.8665
No log 21.625 346 0.7339 0.5575 0.7339 0.8567
No log 21.75 348 0.7299 0.5168 0.7299 0.8543
No log 21.875 350 0.7320 0.5177 0.7320 0.8556
No log 22.0 352 0.7439 0.5575 0.7439 0.8625
No log 22.125 354 0.7735 0.5504 0.7735 0.8795
No log 22.25 356 0.7775 0.5301 0.7775 0.8817
No log 22.375 358 0.7661 0.5265 0.7661 0.8753
No log 22.5 360 0.7371 0.5028 0.7371 0.8585
No log 22.625 362 0.7179 0.4800 0.7179 0.8473
No log 22.75 364 0.7015 0.5264 0.7015 0.8375
No log 22.875 366 0.6878 0.5423 0.6878 0.8293
No log 23.0 368 0.6822 0.5589 0.6822 0.8259
No log 23.125 370 0.6895 0.5672 0.6895 0.8303
No log 23.25 372 0.6995 0.5835 0.6995 0.8364
No log 23.375 374 0.6922 0.5727 0.6922 0.8320
No log 23.5 376 0.7005 0.5545 0.7005 0.8370
No log 23.625 378 0.7074 0.5530 0.7074 0.8411
No log 23.75 380 0.7160 0.5697 0.7160 0.8462
No log 23.875 382 0.7082 0.5657 0.7082 0.8415
No log 24.0 384 0.6939 0.5671 0.6939 0.8330
No log 24.125 386 0.6881 0.5717 0.6881 0.8295
No log 24.25 388 0.6850 0.5659 0.6850 0.8277
No log 24.375 390 0.6809 0.5659 0.6809 0.8252
No log 24.5 392 0.6830 0.5699 0.6830 0.8264
No log 24.625 394 0.6976 0.5503 0.6976 0.8352
No log 24.75 396 0.7219 0.5572 0.7219 0.8497
No log 24.875 398 0.7484 0.5157 0.7484 0.8651
No log 25.0 400 0.7565 0.5216 0.7565 0.8698
No log 25.125 402 0.7476 0.5275 0.7476 0.8647
No log 25.25 404 0.7303 0.5117 0.7303 0.8546
No log 25.375 406 0.7219 0.4833 0.7219 0.8496
No log 25.5 408 0.7185 0.5166 0.7185 0.8476
No log 25.625 410 0.7213 0.5615 0.7213 0.8493
No log 25.75 412 0.7325 0.5747 0.7325 0.8559
No log 25.875 414 0.7444 0.5687 0.7444 0.8628
No log 26.0 416 0.7361 0.5687 0.7361 0.8580
No log 26.125 418 0.7301 0.5692 0.7301 0.8545
No log 26.25 420 0.7312 0.5533 0.7312 0.8551
No log 26.375 422 0.7504 0.5465 0.7504 0.8662
No log 26.5 424 0.7846 0.5169 0.7846 0.8858
No log 26.625 426 0.8038 0.5168 0.8038 0.8965
No log 26.75 428 0.7820 0.5718 0.7820 0.8843
No log 26.875 430 0.7382 0.5848 0.7382 0.8592
No log 27.0 432 0.7229 0.5865 0.7229 0.8502
No log 27.125 434 0.7364 0.5783 0.7364 0.8581
No log 27.25 436 0.7400 0.5819 0.7400 0.8602
No log 27.375 438 0.7266 0.5677 0.7266 0.8524
No log 27.5 440 0.7053 0.5452 0.7053 0.8398
No log 27.625 442 0.7104 0.5192 0.7104 0.8429
No log 27.75 444 0.7089 0.5046 0.7089 0.8420
No log 27.875 446 0.7089 0.5282 0.7089 0.8420
No log 28.0 448 0.7159 0.5532 0.7159 0.8461
No log 28.125 450 0.7261 0.5806 0.7261 0.8521
No log 28.25 452 0.7016 0.5856 0.7016 0.8376
No log 28.375 454 0.6794 0.5831 0.6794 0.8243
No log 28.5 456 0.6652 0.5877 0.6652 0.8156
No log 28.625 458 0.6693 0.5900 0.6693 0.8181
No log 28.75 460 0.6875 0.5505 0.6875 0.8292
No log 28.875 462 0.7221 0.5412 0.7221 0.8497
No log 29.0 464 0.7771 0.5255 0.7771 0.8815
No log 29.125 466 0.8035 0.5122 0.8035 0.8964
No log 29.25 468 0.8056 0.5106 0.8056 0.8976
No log 29.375 470 0.8081 0.5429 0.8081 0.8990
No log 29.5 472 0.7765 0.5481 0.7765 0.8812
No log 29.625 474 0.7258 0.5668 0.7258 0.8520
No log 29.75 476 0.7106 0.5727 0.7106 0.8430
No log 29.875 478 0.7188 0.5468 0.7188 0.8478
No log 30.0 480 0.7402 0.5388 0.7402 0.8603
No log 30.125 482 0.7879 0.5362 0.7879 0.8877
No log 30.25 484 0.8549 0.5296 0.8549 0.9246
No log 30.375 486 0.8955 0.5031 0.8955 0.9463
No log 30.5 488 0.9082 0.4969 0.9082 0.9530
No log 30.625 490 0.8640 0.5279 0.8640 0.9295
No log 30.75 492 0.8095 0.5300 0.8095 0.8997
No log 30.875 494 0.7659 0.4810 0.7659 0.8752
No log 31.0 496 0.7348 0.5312 0.7348 0.8572
No log 31.125 498 0.7119 0.5589 0.7119 0.8437
0.5967 31.25 500 0.7070 0.5825 0.7070 0.8408
0.5967 31.375 502 0.7180 0.5476 0.7180 0.8474
0.5967 31.5 504 0.7396 0.5530 0.7396 0.8600
0.5967 31.625 506 0.7301 0.5651 0.7301 0.8545
0.5967 31.75 508 0.7008 0.5685 0.7008 0.8372
0.5967 31.875 510 0.6901 0.5255 0.6901 0.8307

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.1+cu121
  • Datasets 3.2.0
  • Tokenizers 0.19.1
Downloads last month
3
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_k6_task2_organization

Finetuned
(4040)
this model