ArabicNewSplits8_usingALLEssays_FineTuningAraBERT_run1_AugV5_k3_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: 0.6211
  • Qwk: 0.4889
  • Mse: 0.6211
  • Rmse: 0.7881

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 0.125 2 4.0777 -0.0339 4.0777 2.0193
No log 0.25 4 2.4265 0.0741 2.4265 1.5577
No log 0.375 6 1.2042 0.0687 1.2042 1.0974
No log 0.5 8 0.8717 0.0345 0.8717 0.9336
No log 0.625 10 0.8152 0.2082 0.8152 0.9029
No log 0.75 12 0.8354 0.2041 0.8354 0.9140
No log 0.875 14 1.0774 0.0503 1.0774 1.0380
No log 1.0 16 1.6882 0.0791 1.6882 1.2993
No log 1.125 18 1.2247 0.0718 1.2247 1.1066
No log 1.25 20 0.9073 0.1977 0.9073 0.9525
No log 1.375 22 0.7723 0.2314 0.7723 0.8788
No log 1.5 24 0.7425 0.0867 0.7425 0.8617
No log 1.625 26 0.7933 0.0939 0.7933 0.8907
No log 1.75 28 0.7636 0.1337 0.7636 0.8738
No log 1.875 30 0.7093 0.2723 0.7093 0.8422
No log 2.0 32 0.8054 0.2358 0.8054 0.8975
No log 2.125 34 0.7838 0.2834 0.7838 0.8853
No log 2.25 36 0.6867 0.2982 0.6867 0.8287
No log 2.375 38 0.6468 0.3237 0.6468 0.8042
No log 2.5 40 0.6981 0.3655 0.6981 0.8355
No log 2.625 42 0.9500 0.2739 0.9500 0.9747
No log 2.75 44 0.9777 0.2739 0.9777 0.9888
No log 2.875 46 0.8586 0.2983 0.8586 0.9266
No log 3.0 48 0.6876 0.3450 0.6876 0.8292
No log 3.125 50 0.6134 0.3627 0.6134 0.7832
No log 3.25 52 0.6405 0.3670 0.6405 0.8003
No log 3.375 54 0.7789 0.3254 0.7789 0.8825
No log 3.5 56 1.1733 0.2246 1.1733 1.0832
No log 3.625 58 1.5005 0.1689 1.5005 1.2249
No log 3.75 60 1.3461 0.1714 1.3461 1.1602
No log 3.875 62 0.8868 0.3360 0.8868 0.9417
No log 4.0 64 0.6723 0.3749 0.6723 0.8199
No log 4.125 66 0.6132 0.5004 0.6132 0.7831
No log 4.25 68 0.5796 0.4607 0.5796 0.7613
No log 4.375 70 0.5721 0.4481 0.5721 0.7564
No log 4.5 72 0.6503 0.4218 0.6503 0.8064
No log 4.625 74 0.9407 0.3164 0.9407 0.9699
No log 4.75 76 1.2711 0.2311 1.2711 1.1274
No log 4.875 78 1.2661 0.2045 1.2661 1.1252
No log 5.0 80 0.9261 0.3067 0.9261 0.9624
No log 5.125 82 0.5971 0.4850 0.5971 0.7727
No log 5.25 84 0.6009 0.4147 0.6009 0.7752
No log 5.375 86 0.6415 0.4333 0.6415 0.8009
No log 5.5 88 0.6005 0.3801 0.6005 0.7749
No log 5.625 90 0.5541 0.4354 0.5541 0.7444
No log 5.75 92 0.7275 0.3799 0.7275 0.8529
No log 5.875 94 0.9462 0.3874 0.9462 0.9727
No log 6.0 96 0.9967 0.4017 0.9967 0.9983
No log 6.125 98 0.8081 0.4763 0.8081 0.8990
No log 6.25 100 0.7767 0.4915 0.7767 0.8813
No log 6.375 102 0.6220 0.4945 0.6220 0.7887
No log 6.5 104 0.6022 0.5169 0.6022 0.7760
No log 6.625 106 0.6218 0.4639 0.6218 0.7885
No log 6.75 108 0.7791 0.4706 0.7791 0.8827
No log 6.875 110 0.8671 0.3948 0.8671 0.9312
No log 7.0 112 0.7272 0.4610 0.7272 0.8528
No log 7.125 114 0.5521 0.5137 0.5521 0.7431
No log 7.25 116 0.5495 0.5447 0.5495 0.7413
No log 7.375 118 0.5648 0.5124 0.5648 0.7516
No log 7.5 120 0.5216 0.5845 0.5216 0.7222
No log 7.625 122 0.5376 0.5394 0.5376 0.7332
No log 7.75 124 0.6473 0.4982 0.6473 0.8045
No log 7.875 126 0.7727 0.4407 0.7727 0.8790
No log 8.0 128 0.7813 0.4565 0.7813 0.8839
No log 8.125 130 0.7278 0.5177 0.7278 0.8531
No log 8.25 132 0.6022 0.5889 0.6022 0.7760
No log 8.375 134 0.6445 0.5630 0.6445 0.8028
No log 8.5 136 0.7577 0.5018 0.7577 0.8705
No log 8.625 138 0.7679 0.4895 0.7679 0.8763
No log 8.75 140 0.6686 0.5229 0.6686 0.8177
No log 8.875 142 0.5822 0.5768 0.5822 0.7630
No log 9.0 144 0.5804 0.5358 0.5804 0.7618
No log 9.125 146 0.5927 0.5095 0.5927 0.7699
No log 9.25 148 0.6441 0.4966 0.6441 0.8026
No log 9.375 150 0.6322 0.4736 0.6322 0.7951
No log 9.5 152 0.6183 0.4504 0.6183 0.7863
No log 9.625 154 0.6300 0.4404 0.6300 0.7937
No log 9.75 156 0.5889 0.4897 0.5889 0.7674
No log 9.875 158 0.6055 0.5056 0.6055 0.7782
No log 10.0 160 0.5904 0.4825 0.5904 0.7684
No log 10.125 162 0.6276 0.5030 0.6276 0.7922
No log 10.25 164 0.6551 0.5070 0.6551 0.8094
No log 10.375 166 0.6184 0.4880 0.6184 0.7864
No log 10.5 168 0.6745 0.5079 0.6745 0.8213
No log 10.625 170 0.7020 0.5079 0.7020 0.8378
No log 10.75 172 0.7987 0.4940 0.7987 0.8937
No log 10.875 174 0.6727 0.5288 0.6727 0.8202
No log 11.0 176 0.6208 0.5306 0.6208 0.7879
No log 11.125 178 0.7212 0.4832 0.7212 0.8492
No log 11.25 180 0.7059 0.4989 0.7059 0.8402
No log 11.375 182 0.6031 0.4981 0.6031 0.7766
No log 11.5 184 0.6215 0.5425 0.6215 0.7883
No log 11.625 186 0.6714 0.5618 0.6714 0.8194
No log 11.75 188 0.7142 0.5532 0.7142 0.8451
No log 11.875 190 0.6744 0.5602 0.6744 0.8212
No log 12.0 192 0.6185 0.5630 0.6185 0.7865
No log 12.125 194 0.6021 0.5454 0.6021 0.7760
No log 12.25 196 0.6094 0.4921 0.6094 0.7806
No log 12.375 198 0.5839 0.5087 0.5839 0.7641
No log 12.5 200 0.5694 0.5574 0.5694 0.7546
No log 12.625 202 0.5791 0.5295 0.5791 0.7610
No log 12.75 204 0.5874 0.5410 0.5874 0.7664
No log 12.875 206 0.5516 0.5397 0.5516 0.7427
No log 13.0 208 0.5498 0.5212 0.5498 0.7415
No log 13.125 210 0.5536 0.5530 0.5536 0.7440
No log 13.25 212 0.5591 0.5622 0.5591 0.7477
No log 13.375 214 0.5609 0.5469 0.5609 0.7489
No log 13.5 216 0.6166 0.5582 0.6166 0.7852
No log 13.625 218 0.5767 0.5445 0.5767 0.7594
No log 13.75 220 0.5585 0.5322 0.5585 0.7473
No log 13.875 222 0.5490 0.5511 0.5490 0.7409
No log 14.0 224 0.5757 0.5510 0.5757 0.7587
No log 14.125 226 0.5991 0.5026 0.5991 0.7740
No log 14.25 228 0.5497 0.5137 0.5497 0.7414
No log 14.375 230 0.5381 0.5454 0.5381 0.7336
No log 14.5 232 0.5941 0.4886 0.5941 0.7708
No log 14.625 234 0.6377 0.4782 0.6377 0.7985
No log 14.75 236 0.5687 0.5551 0.5687 0.7541
No log 14.875 238 0.5819 0.5784 0.5819 0.7629
No log 15.0 240 0.6453 0.5247 0.6453 0.8033
No log 15.125 242 0.6497 0.5434 0.6497 0.8060
No log 15.25 244 0.5613 0.5668 0.5613 0.7492
No log 15.375 246 0.5833 0.5042 0.5833 0.7637
No log 15.5 248 0.7115 0.4846 0.7115 0.8435
No log 15.625 250 0.6779 0.4804 0.6779 0.8233
No log 15.75 252 0.5673 0.5075 0.5673 0.7532
No log 15.875 254 0.6060 0.5698 0.6060 0.7785
No log 16.0 256 0.6295 0.5423 0.6295 0.7934
No log 16.125 258 0.5773 0.5602 0.5773 0.7598
No log 16.25 260 0.5732 0.4790 0.5732 0.7571
No log 16.375 262 0.7271 0.4201 0.7271 0.8527
No log 16.5 264 0.8290 0.4205 0.8290 0.9105
No log 16.625 266 0.7346 0.4201 0.7346 0.8571
No log 16.75 268 0.5964 0.4399 0.5964 0.7723
No log 16.875 270 0.5679 0.5377 0.5679 0.7536
No log 17.0 272 0.7084 0.4403 0.7084 0.8417
No log 17.125 274 0.7559 0.4392 0.7559 0.8694
No log 17.25 276 0.7097 0.4701 0.7097 0.8424
No log 17.375 278 0.6152 0.4953 0.6152 0.7843
No log 17.5 280 0.5796 0.5705 0.5796 0.7613
No log 17.625 282 0.5723 0.5705 0.5723 0.7565
No log 17.75 284 0.5727 0.5705 0.5727 0.7568
No log 17.875 286 0.5684 0.5511 0.5684 0.7539
No log 18.0 288 0.5762 0.5705 0.5762 0.7591
No log 18.125 290 0.5660 0.5415 0.5660 0.7523
No log 18.25 292 0.5574 0.5171 0.5574 0.7466
No log 18.375 294 0.5625 0.5357 0.5625 0.7500
No log 18.5 296 0.5683 0.5371 0.5683 0.7539
No log 18.625 298 0.5753 0.5327 0.5753 0.7585
No log 18.75 300 0.5839 0.5560 0.5839 0.7641
No log 18.875 302 0.5758 0.5459 0.5758 0.7588
No log 19.0 304 0.5710 0.5660 0.5710 0.7557
No log 19.125 306 0.5686 0.5547 0.5686 0.7541
No log 19.25 308 0.5689 0.5547 0.5689 0.7543
No log 19.375 310 0.5714 0.5363 0.5714 0.7559
No log 19.5 312 0.5851 0.4906 0.5851 0.7649
No log 19.625 314 0.5819 0.5428 0.5819 0.7628
No log 19.75 316 0.5754 0.5721 0.5754 0.7586
No log 19.875 318 0.5887 0.4784 0.5887 0.7673
No log 20.0 320 0.5755 0.5046 0.5755 0.7586
No log 20.125 322 0.5530 0.5595 0.5530 0.7437
No log 20.25 324 0.5528 0.5527 0.5528 0.7435
No log 20.375 326 0.5718 0.5868 0.5718 0.7561
No log 20.5 328 0.5771 0.5344 0.5771 0.7597
No log 20.625 330 0.5551 0.5613 0.5551 0.7450
No log 20.75 332 0.5828 0.4898 0.5828 0.7634
No log 20.875 334 0.5999 0.4882 0.5999 0.7746
No log 21.0 336 0.6166 0.4612 0.6166 0.7852
No log 21.125 338 0.6152 0.4276 0.6152 0.7844
No log 21.25 340 0.5611 0.5491 0.5611 0.7491
No log 21.375 342 0.5527 0.5795 0.5527 0.7434
No log 21.5 344 0.5647 0.5606 0.5647 0.7515
No log 21.625 346 0.5694 0.5489 0.5694 0.7546
No log 21.75 348 0.5733 0.5234 0.5733 0.7571
No log 21.875 350 0.5673 0.5041 0.5673 0.7532
No log 22.0 352 0.5606 0.5175 0.5606 0.7487
No log 22.125 354 0.5576 0.5278 0.5576 0.7467
No log 22.25 356 0.5483 0.5267 0.5483 0.7405
No log 22.375 358 0.5485 0.5434 0.5485 0.7406
No log 22.5 360 0.5727 0.5180 0.5727 0.7567
No log 22.625 362 0.5727 0.5084 0.5727 0.7567
No log 22.75 364 0.5640 0.5724 0.5640 0.7510
No log 22.875 366 0.5769 0.5323 0.5769 0.7595
No log 23.0 368 0.5951 0.5570 0.5951 0.7714
No log 23.125 370 0.6091 0.5013 0.6091 0.7805
No log 23.25 372 0.5956 0.5256 0.5956 0.7718
No log 23.375 374 0.5671 0.5711 0.5671 0.7531
No log 23.5 376 0.5628 0.5592 0.5628 0.7502
No log 23.625 378 0.5716 0.5274 0.5716 0.7560
No log 23.75 380 0.5875 0.4883 0.5875 0.7665
No log 23.875 382 0.5800 0.5086 0.5800 0.7616
No log 24.0 384 0.5767 0.5299 0.5767 0.7594
No log 24.125 386 0.5926 0.5558 0.5926 0.7698
No log 24.25 388 0.6116 0.5113 0.6116 0.7821
No log 24.375 390 0.6252 0.4598 0.6252 0.7907
No log 24.5 392 0.5896 0.5475 0.5896 0.7679
No log 24.625 394 0.5683 0.5228 0.5683 0.7538
No log 24.75 396 0.5717 0.5167 0.5717 0.7561
No log 24.875 398 0.5971 0.5228 0.5971 0.7728
No log 25.0 400 0.6417 0.5481 0.6417 0.8011
No log 25.125 402 0.6579 0.5174 0.6579 0.8111
No log 25.25 404 0.6499 0.5131 0.6499 0.8062
No log 25.375 406 0.6289 0.5216 0.6289 0.7930
No log 25.5 408 0.6186 0.5251 0.6186 0.7865
No log 25.625 410 0.6131 0.5327 0.6131 0.7830
No log 25.75 412 0.6357 0.5379 0.6357 0.7973
No log 25.875 414 0.6631 0.4966 0.6631 0.8143
No log 26.0 416 0.6424 0.4943 0.6424 0.8015
No log 26.125 418 0.6272 0.5666 0.6272 0.7919
No log 26.25 420 0.7122 0.5499 0.7122 0.8439
No log 26.375 422 0.8158 0.5284 0.8158 0.9032
No log 26.5 424 0.8098 0.5070 0.8098 0.8999
No log 26.625 426 0.7654 0.5284 0.7654 0.8749
No log 26.75 428 0.6806 0.5278 0.6806 0.8250
No log 26.875 430 0.6021 0.5149 0.6021 0.7760
No log 27.0 432 0.5900 0.5132 0.5900 0.7681
No log 27.125 434 0.5873 0.5254 0.5873 0.7664
No log 27.25 436 0.5754 0.4977 0.5754 0.7585
No log 27.375 438 0.5749 0.5179 0.5749 0.7582
No log 27.5 440 0.5763 0.5189 0.5763 0.7592
No log 27.625 442 0.5802 0.5521 0.5802 0.7617
No log 27.75 444 0.5926 0.5625 0.5926 0.7698
No log 27.875 446 0.5975 0.5393 0.5975 0.7730
No log 28.0 448 0.5895 0.5539 0.5895 0.7678
No log 28.125 450 0.5789 0.5566 0.5789 0.7608
No log 28.25 452 0.5734 0.5345 0.5734 0.7573
No log 28.375 454 0.5835 0.5711 0.5835 0.7639
No log 28.5 456 0.6212 0.5681 0.6212 0.7882
No log 28.625 458 0.6507 0.5303 0.6507 0.8067
No log 28.75 460 0.6304 0.5602 0.6304 0.7940
No log 28.875 462 0.6101 0.5364 0.6101 0.7811
No log 29.0 464 0.6114 0.5351 0.6114 0.7819
No log 29.125 466 0.5848 0.5397 0.5848 0.7647
No log 29.25 468 0.5835 0.5023 0.5835 0.7639
No log 29.375 470 0.5872 0.5556 0.5872 0.7663
No log 29.5 472 0.5910 0.5348 0.5910 0.7688
No log 29.625 474 0.6072 0.5144 0.6072 0.7792
No log 29.75 476 0.6472 0.5270 0.6472 0.8045
No log 29.875 478 0.6895 0.5191 0.6895 0.8304
No log 30.0 480 0.6606 0.5304 0.6606 0.8128
No log 30.125 482 0.5978 0.5227 0.5978 0.7732
No log 30.25 484 0.5798 0.5511 0.5798 0.7615
No log 30.375 486 0.5768 0.5453 0.5768 0.7595
No log 30.5 488 0.5989 0.5315 0.5989 0.7739
No log 30.625 490 0.6254 0.5535 0.6254 0.7908
No log 30.75 492 0.6301 0.5619 0.6301 0.7938
No log 30.875 494 0.6285 0.5617 0.6285 0.7928
No log 31.0 496 0.6197 0.5513 0.6197 0.7872
No log 31.125 498 0.6068 0.5111 0.6068 0.7790
0.3345 31.25 500 0.6003 0.5090 0.6003 0.7748
0.3345 31.375 502 0.6054 0.5385 0.6054 0.7781
0.3345 31.5 504 0.6096 0.5367 0.6096 0.7807
0.3345 31.625 506 0.6097 0.5254 0.6097 0.7809
0.3345 31.75 508 0.6026 0.5364 0.6026 0.7763
0.3345 31.875 510 0.5979 0.5197 0.5979 0.7732
0.3345 32.0 512 0.5962 0.5438 0.5962 0.7722
0.3345 32.125 514 0.6043 0.5304 0.6043 0.7773
0.3345 32.25 516 0.6141 0.4934 0.6141 0.7836
0.3345 32.375 518 0.6224 0.5042 0.6224 0.7889
0.3345 32.5 520 0.6214 0.4889 0.6214 0.7883
0.3345 32.625 522 0.6204 0.5016 0.6204 0.7877
0.3345 32.75 524 0.6211 0.4889 0.6211 0.7881

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_k3_task2_organization

Finetuned
(4040)
this model