ArabicNewSplits7_FineTuningAraBERT_run1_AugV5_k4_task7_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: 0.4544
  • Qwk: 0.6133
  • Mse: 0.4544
  • Rmse: 0.6741

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 0.1667 2 2.6011 -0.0262 2.6011 1.6128
No log 0.3333 4 1.3154 0.0754 1.3154 1.1469
No log 0.5 6 0.8718 0.0944 0.8718 0.9337
No log 0.6667 8 0.8615 -0.0483 0.8615 0.9282
No log 0.8333 10 0.7831 0.1313 0.7831 0.8849
No log 1.0 12 0.7177 0.0889 0.7177 0.8472
No log 1.1667 14 0.8447 0.2435 0.8447 0.9191
No log 1.3333 16 0.9153 0.2702 0.9153 0.9567
No log 1.5 18 1.0602 0.1737 1.0602 1.0297
No log 1.6667 20 0.7519 0.4246 0.7519 0.8671
No log 1.8333 22 0.5493 0.3745 0.5493 0.7411
No log 2.0 24 0.5618 0.4158 0.5618 0.7496
No log 2.1667 26 0.5797 0.3996 0.5797 0.7614
No log 2.3333 28 0.5338 0.4614 0.5338 0.7306
No log 2.5 30 0.5257 0.4384 0.5257 0.7251
No log 2.6667 32 0.6413 0.5171 0.6413 0.8008
No log 2.8333 34 0.5914 0.4756 0.5914 0.7690
No log 3.0 36 0.5618 0.4979 0.5618 0.7496
No log 3.1667 38 0.5462 0.5195 0.5462 0.7391
No log 3.3333 40 0.5712 0.5195 0.5712 0.7558
No log 3.5 42 0.5502 0.5357 0.5502 0.7418
No log 3.6667 44 0.6169 0.4807 0.6169 0.7854
No log 3.8333 46 0.8249 0.4260 0.8249 0.9083
No log 4.0 48 0.7295 0.4444 0.7295 0.8541
No log 4.1667 50 0.5392 0.5406 0.5392 0.7343
No log 4.3333 52 0.5370 0.5015 0.5370 0.7328
No log 4.5 54 0.5520 0.4920 0.5520 0.7430
No log 4.6667 56 0.5263 0.5561 0.5263 0.7255
No log 4.8333 58 0.6657 0.4961 0.6657 0.8159
No log 5.0 60 0.7027 0.4906 0.7027 0.8383
No log 5.1667 62 0.6820 0.4906 0.6820 0.8259
No log 5.3333 64 0.5220 0.5697 0.5220 0.7225
No log 5.5 66 0.5394 0.4923 0.5394 0.7344
No log 5.6667 68 0.5961 0.5330 0.5961 0.7721
No log 5.8333 70 0.5405 0.5823 0.5405 0.7352
No log 6.0 72 0.5448 0.6130 0.5448 0.7381
No log 6.1667 74 0.5389 0.5024 0.5389 0.7341
No log 6.3333 76 0.5292 0.4901 0.5292 0.7275
No log 6.5 78 0.5634 0.5908 0.5634 0.7506
No log 6.6667 80 0.4995 0.4904 0.4995 0.7068
No log 6.8333 82 0.4983 0.5965 0.4983 0.7059
No log 7.0 84 0.4749 0.5882 0.4749 0.6891
No log 7.1667 86 0.5067 0.5932 0.5067 0.7118
No log 7.3333 88 0.7851 0.4815 0.7851 0.8860
No log 7.5 90 0.7856 0.4993 0.7856 0.8864
No log 7.6667 92 0.5632 0.5874 0.5632 0.7505
No log 7.8333 94 0.4890 0.6307 0.4890 0.6993
No log 8.0 96 0.4868 0.5742 0.4868 0.6977
No log 8.1667 98 0.5075 0.5654 0.5075 0.7124
No log 8.3333 100 0.6534 0.5595 0.6534 0.8083
No log 8.5 102 0.6808 0.5408 0.6808 0.8251
No log 8.6667 104 0.5376 0.5765 0.5376 0.7332
No log 8.8333 106 0.4917 0.5617 0.4917 0.7012
No log 9.0 108 0.5678 0.5495 0.5678 0.7535
No log 9.1667 110 0.5031 0.5559 0.5031 0.7093
No log 9.3333 112 0.5725 0.5683 0.5725 0.7566
No log 9.5 114 0.6775 0.5266 0.6775 0.8231
No log 9.6667 116 0.6063 0.5683 0.6063 0.7787
No log 9.8333 118 0.5102 0.5413 0.5102 0.7143
No log 10.0 120 0.5376 0.5617 0.5376 0.7332
No log 10.1667 122 0.5191 0.6395 0.5191 0.7205
No log 10.3333 124 0.5045 0.4972 0.5045 0.7103
No log 10.5 126 0.5936 0.5298 0.5936 0.7705
No log 10.6667 128 0.6632 0.5200 0.6632 0.8144
No log 10.8333 130 0.6045 0.5378 0.6045 0.7775
No log 11.0 132 0.5067 0.4536 0.5067 0.7118
No log 11.1667 134 0.5103 0.5868 0.5103 0.7144
No log 11.3333 136 0.5105 0.5326 0.5105 0.7145
No log 11.5 138 0.5009 0.5440 0.5009 0.7078
No log 11.6667 140 0.5473 0.5265 0.5473 0.7398
No log 11.8333 142 0.5663 0.5666 0.5663 0.7525
No log 12.0 144 0.5096 0.5539 0.5096 0.7138
No log 12.1667 146 0.4889 0.5379 0.4889 0.6992
No log 12.3333 148 0.5018 0.5633 0.5018 0.7084
No log 12.5 150 0.4820 0.5533 0.4820 0.6943
No log 12.6667 152 0.4824 0.5604 0.4824 0.6945
No log 12.8333 154 0.5330 0.5486 0.5330 0.7300
No log 13.0 156 0.5875 0.5595 0.5875 0.7665
No log 13.1667 158 0.5877 0.5393 0.5877 0.7666
No log 13.3333 160 0.4954 0.5841 0.4954 0.7039
No log 13.5 162 0.4481 0.5815 0.4481 0.6694
No log 13.6667 164 0.4591 0.5974 0.4591 0.6776
No log 13.8333 166 0.4762 0.6518 0.4762 0.6901
No log 14.0 168 0.4586 0.6716 0.4586 0.6772
No log 14.1667 170 0.4692 0.5767 0.4692 0.6850
No log 14.3333 172 0.4634 0.5840 0.4634 0.6807
No log 14.5 174 0.4453 0.6739 0.4453 0.6673
No log 14.6667 176 0.4486 0.7053 0.4486 0.6698
No log 14.8333 178 0.4373 0.6839 0.4373 0.6613
No log 15.0 180 0.4925 0.5567 0.4925 0.7017
No log 15.1667 182 0.5562 0.5595 0.5562 0.7458
No log 15.3333 184 0.6857 0.5093 0.6857 0.8281
No log 15.5 186 0.7068 0.5146 0.7068 0.8407
No log 15.6667 188 0.5838 0.6081 0.5838 0.7641
No log 15.8333 190 0.5012 0.6198 0.5012 0.7080
No log 16.0 192 0.5635 0.6237 0.5635 0.7506
No log 16.1667 194 0.5821 0.5562 0.5821 0.7630
No log 16.3333 196 0.5132 0.5457 0.5132 0.7163
No log 16.5 198 0.4828 0.5662 0.4828 0.6948
No log 16.6667 200 0.5140 0.6118 0.5140 0.7169
No log 16.8333 202 0.5379 0.5596 0.5379 0.7334
No log 17.0 204 0.5083 0.5756 0.5083 0.7130
No log 17.1667 206 0.4918 0.6307 0.4918 0.7013
No log 17.3333 208 0.5050 0.6314 0.5050 0.7106
No log 17.5 210 0.5291 0.6074 0.5291 0.7274
No log 17.6667 212 0.5438 0.6070 0.5438 0.7375
No log 17.8333 214 0.5160 0.5953 0.5160 0.7183
No log 18.0 216 0.4724 0.6006 0.4724 0.6873
No log 18.1667 218 0.4623 0.5974 0.4623 0.6799
No log 18.3333 220 0.4621 0.5951 0.4621 0.6798
No log 18.5 222 0.4480 0.5902 0.4480 0.6693
No log 18.6667 224 0.4655 0.5751 0.4655 0.6823
No log 18.8333 226 0.5566 0.6104 0.5566 0.7461
No log 19.0 228 0.5943 0.5735 0.5943 0.7709
No log 19.1667 230 0.5433 0.6104 0.5433 0.7371
No log 19.3333 232 0.4723 0.5554 0.4723 0.6872
No log 19.5 234 0.4777 0.5840 0.4777 0.6911
No log 19.6667 236 0.5100 0.6690 0.5100 0.7141
No log 19.8333 238 0.4954 0.6282 0.4954 0.7038
No log 20.0 240 0.4743 0.5866 0.4743 0.6887
No log 20.1667 242 0.4738 0.5812 0.4738 0.6884
No log 20.3333 244 0.4693 0.5596 0.4693 0.6850
No log 20.5 246 0.4898 0.5607 0.4898 0.6999
No log 20.6667 248 0.5096 0.5692 0.5096 0.7139
No log 20.8333 250 0.5078 0.5352 0.5078 0.7126
No log 21.0 252 0.4878 0.5522 0.4878 0.6984
No log 21.1667 254 0.4673 0.5114 0.4673 0.6836
No log 21.3333 256 0.4602 0.5665 0.4602 0.6784
No log 21.5 258 0.4526 0.5522 0.4526 0.6727
No log 21.6667 260 0.4844 0.5932 0.4844 0.6960
No log 21.8333 262 0.4923 0.5974 0.4923 0.7016
No log 22.0 264 0.4852 0.6254 0.4852 0.6966
No log 22.1667 266 0.5069 0.5745 0.5069 0.7119
No log 22.3333 268 0.4970 0.5786 0.4970 0.7050
No log 22.5 270 0.4616 0.5937 0.4616 0.6794
No log 22.6667 272 0.4682 0.4924 0.4682 0.6842
No log 22.8333 274 0.4903 0.4538 0.4903 0.7002
No log 23.0 276 0.4817 0.4634 0.4817 0.6940
No log 23.1667 278 0.4714 0.4825 0.4714 0.6866
No log 23.3333 280 0.4492 0.5286 0.4492 0.6702
No log 23.5 282 0.4645 0.5841 0.4645 0.6815
No log 23.6667 284 0.4808 0.6257 0.4808 0.6934
No log 23.8333 286 0.4753 0.6340 0.4753 0.6894
No log 24.0 288 0.4774 0.6340 0.4774 0.6909
No log 24.1667 290 0.4812 0.5979 0.4812 0.6937
No log 24.3333 292 0.4908 0.5809 0.4908 0.7005
No log 24.5 294 0.4890 0.6289 0.4890 0.6993
No log 24.6667 296 0.4722 0.6158 0.4722 0.6872
No log 24.8333 298 0.4831 0.6195 0.4831 0.6950
No log 25.0 300 0.4727 0.6345 0.4727 0.6875
No log 25.1667 302 0.4988 0.5599 0.4988 0.7063
No log 25.3333 304 0.5063 0.5400 0.5063 0.7115
No log 25.5 306 0.4797 0.5527 0.4797 0.6926
No log 25.6667 308 0.4567 0.5114 0.4567 0.6758
No log 25.8333 310 0.4664 0.5267 0.4664 0.6829
No log 26.0 312 0.4713 0.5305 0.4713 0.6865
No log 26.1667 314 0.4952 0.5960 0.4952 0.7037
No log 26.3333 316 0.5123 0.6333 0.5123 0.7158
No log 26.5 318 0.5085 0.6484 0.5085 0.7131
No log 26.6667 320 0.5142 0.6657 0.5142 0.7171
No log 26.8333 322 0.5174 0.6484 0.5174 0.7193
No log 27.0 324 0.5153 0.6040 0.5153 0.7179
No log 27.1667 326 0.5514 0.5942 0.5514 0.7426
No log 27.3333 328 0.6137 0.5521 0.6137 0.7834
No log 27.5 330 0.6145 0.5208 0.6145 0.7839
No log 27.6667 332 0.5887 0.5111 0.5887 0.7673
No log 27.8333 334 0.5088 0.5687 0.5088 0.7133
No log 28.0 336 0.4745 0.5951 0.4745 0.6888
No log 28.1667 338 0.4699 0.5714 0.4699 0.6855
No log 28.3333 340 0.4661 0.6092 0.4661 0.6827
No log 28.5 342 0.4680 0.5889 0.4680 0.6841
No log 28.6667 344 0.4687 0.6001 0.4687 0.6846
No log 28.8333 346 0.4817 0.6709 0.4817 0.6940
No log 29.0 348 0.4879 0.6631 0.4879 0.6985
No log 29.1667 350 0.4979 0.6361 0.4979 0.7056
No log 29.3333 352 0.4894 0.6526 0.4894 0.6996
No log 29.5 354 0.4741 0.6127 0.4741 0.6886
No log 29.6667 356 0.4679 0.5538 0.4679 0.6840
No log 29.8333 358 0.4755 0.5539 0.4755 0.6896
No log 30.0 360 0.4939 0.5404 0.4939 0.7027
No log 30.1667 362 0.5404 0.5499 0.5404 0.7351
No log 30.3333 364 0.5608 0.4845 0.5608 0.7489
No log 30.5 366 0.5286 0.5157 0.5286 0.7271
No log 30.6667 368 0.4972 0.5190 0.4972 0.7051
No log 30.8333 370 0.4928 0.5190 0.4928 0.7020
No log 31.0 372 0.4915 0.5715 0.4915 0.7011
No log 31.1667 374 0.4847 0.5910 0.4847 0.6962
No log 31.3333 376 0.4834 0.5840 0.4834 0.6953
No log 31.5 378 0.4828 0.5902 0.4828 0.6948
No log 31.6667 380 0.4806 0.5902 0.4806 0.6933
No log 31.8333 382 0.4704 0.5982 0.4704 0.6859
No log 32.0 384 0.4582 0.5669 0.4582 0.6769
No log 32.1667 386 0.4520 0.5555 0.4520 0.6723
No log 32.3333 388 0.4463 0.5538 0.4463 0.6680
No log 32.5 390 0.4430 0.5357 0.4430 0.6655
No log 32.6667 392 0.4457 0.5450 0.4457 0.6676
No log 32.8333 394 0.4577 0.5368 0.4577 0.6765
No log 33.0 396 0.4660 0.5578 0.4660 0.6826
No log 33.1667 398 0.4684 0.5368 0.4684 0.6844
No log 33.3333 400 0.4645 0.5217 0.4645 0.6815
No log 33.5 402 0.4598 0.5286 0.4598 0.6781
No log 33.6667 404 0.4573 0.5383 0.4573 0.6763
No log 33.8333 406 0.4610 0.5840 0.4610 0.6789
No log 34.0 408 0.4600 0.5705 0.4600 0.6782
No log 34.1667 410 0.4616 0.5621 0.4616 0.6794
No log 34.3333 412 0.4640 0.5810 0.4640 0.6812
No log 34.5 414 0.4597 0.5701 0.4597 0.6780
No log 34.6667 416 0.4600 0.5434 0.4600 0.6782
No log 34.8333 418 0.4607 0.5419 0.4607 0.6787
No log 35.0 420 0.4568 0.5572 0.4568 0.6759
No log 35.1667 422 0.4618 0.5867 0.4618 0.6796
No log 35.3333 424 0.4683 0.6333 0.4683 0.6843
No log 35.5 426 0.4785 0.6156 0.4785 0.6917
No log 35.6667 428 0.4918 0.6156 0.4918 0.7013
No log 35.8333 430 0.4797 0.6170 0.4797 0.6926
No log 36.0 432 0.4668 0.5753 0.4668 0.6833
No log 36.1667 434 0.4694 0.5853 0.4694 0.6851
No log 36.3333 436 0.4761 0.6028 0.4761 0.6900
No log 36.5 438 0.4710 0.5840 0.4710 0.6863
No log 36.6667 440 0.4707 0.5840 0.4707 0.6861
No log 36.8333 442 0.4812 0.5840 0.4812 0.6937
No log 37.0 444 0.4824 0.6028 0.4824 0.6945
No log 37.1667 446 0.4860 0.6013 0.4860 0.6972
No log 37.3333 448 0.4976 0.6287 0.4976 0.7054
No log 37.5 450 0.5130 0.6137 0.5130 0.7163
No log 37.6667 452 0.4984 0.5941 0.4984 0.7060
No log 37.8333 454 0.4662 0.5628 0.4662 0.6828
No log 38.0 456 0.4529 0.5769 0.4529 0.6730
No log 38.1667 458 0.4466 0.5674 0.4466 0.6683
No log 38.3333 460 0.4420 0.5594 0.4420 0.6649
No log 38.5 462 0.4379 0.5707 0.4379 0.6617
No log 38.6667 464 0.4494 0.5841 0.4494 0.6704
No log 38.8333 466 0.4632 0.6169 0.4632 0.6806
No log 39.0 468 0.4599 0.6169 0.4599 0.6781
No log 39.1667 470 0.4552 0.6248 0.4552 0.6747
No log 39.3333 472 0.4546 0.6349 0.4546 0.6742
No log 39.5 474 0.4498 0.6156 0.4498 0.6707
No log 39.6667 476 0.4604 0.6349 0.4604 0.6785
No log 39.8333 478 0.4599 0.6156 0.4599 0.6782
No log 40.0 480 0.4479 0.5956 0.4479 0.6693
No log 40.1667 482 0.4437 0.5578 0.4437 0.6661
No log 40.3333 484 0.4416 0.5797 0.4416 0.6645
No log 40.5 486 0.4420 0.5866 0.4420 0.6648
No log 40.6667 488 0.4417 0.5915 0.4417 0.6646
No log 40.8333 490 0.4579 0.5918 0.4579 0.6766
No log 41.0 492 0.4746 0.5918 0.4746 0.6889
No log 41.1667 494 0.4645 0.5918 0.4645 0.6815
No log 41.3333 496 0.4429 0.5213 0.4429 0.6655
No log 41.5 498 0.4455 0.5719 0.4455 0.6675
0.2345 41.6667 500 0.4652 0.6599 0.4652 0.6821
0.2345 41.8333 502 0.4575 0.6438 0.4575 0.6764
0.2345 42.0 504 0.4488 0.5742 0.4488 0.6699
0.2345 42.1667 506 0.4536 0.5731 0.4536 0.6735
0.2345 42.3333 508 0.4614 0.6059 0.4614 0.6793
0.2345 42.5 510 0.4544 0.6133 0.4544 0.6741

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0+cu118
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
7
Safetensors
Model size
135M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for MayBashendy/ArabicNewSplits7_FineTuningAraBERT_run1_AugV5_k4_task7_organization

Finetuned
(4204)
this model