ArabicNewSplits8_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k4_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: 0.6396
  • Qwk: 0.4892
  • Mse: 0.6396
  • Rmse: 0.7998

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 0.0833 2 4.4371 -0.0170 4.4371 2.1064
No log 0.1667 4 2.6965 0.0233 2.6965 1.6421
No log 0.25 6 1.3448 0.0307 1.3448 1.1597
No log 0.3333 8 1.1271 -0.0565 1.1271 1.0616
No log 0.4167 10 1.0321 0.0186 1.0321 1.0159
No log 0.5 12 1.0993 -0.0822 1.0993 1.0485
No log 0.5833 14 1.1387 -0.0310 1.1387 1.0671
No log 0.6667 16 1.0173 0.0405 1.0173 1.0086
No log 0.75 18 1.0186 -0.0080 1.0186 1.0092
No log 0.8333 20 0.9428 0.0283 0.9428 0.9710
No log 0.9167 22 0.8516 0.2202 0.8516 0.9228
No log 1.0 24 0.7919 0.3000 0.7919 0.8899
No log 1.0833 26 0.7830 0.3169 0.7830 0.8849
No log 1.1667 28 0.8189 0.2097 0.8189 0.9049
No log 1.25 30 0.8603 0.1942 0.8603 0.9275
No log 1.3333 32 0.9227 0.2060 0.9227 0.9606
No log 1.4167 34 1.0138 0.0630 1.0138 1.0069
No log 1.5 36 0.9664 0.0958 0.9664 0.9831
No log 1.5833 38 0.8604 0.2034 0.8604 0.9276
No log 1.6667 40 0.8206 0.2043 0.8206 0.9059
No log 1.75 42 0.8042 0.1527 0.8042 0.8968
No log 1.8333 44 0.7802 0.1417 0.7802 0.8833
No log 1.9167 46 0.7490 0.1842 0.7490 0.8655
No log 2.0 48 0.7340 0.3554 0.7340 0.8567
No log 2.0833 50 0.7349 0.3309 0.7349 0.8573
No log 2.1667 52 0.7318 0.3388 0.7318 0.8555
No log 2.25 54 0.7172 0.3220 0.7172 0.8468
No log 2.3333 56 0.6993 0.3386 0.6993 0.8362
No log 2.4167 58 0.7618 0.2123 0.7618 0.8728
No log 2.5 60 0.9879 0.2795 0.9879 0.9940
No log 2.5833 62 0.8440 0.2709 0.8440 0.9187
No log 2.6667 64 0.6712 0.3357 0.6712 0.8193
No log 2.75 66 0.7070 0.3183 0.7070 0.8408
No log 2.8333 68 0.6352 0.3999 0.6352 0.7970
No log 2.9167 70 0.7372 0.3846 0.7372 0.8586
No log 3.0 72 1.3245 0.2583 1.3245 1.1509
No log 3.0833 74 1.3904 0.2730 1.3904 1.1792
No log 3.1667 76 0.8568 0.3704 0.8568 0.9256
No log 3.25 78 0.6147 0.4198 0.6147 0.7840
No log 3.3333 80 0.6647 0.4164 0.6647 0.8153
No log 3.4167 82 0.6327 0.4248 0.6327 0.7955
No log 3.5 84 0.7387 0.4070 0.7387 0.8595
No log 3.5833 86 0.8702 0.3743 0.8702 0.9328
No log 3.6667 88 0.7319 0.3884 0.7319 0.8555
No log 3.75 90 0.7013 0.4800 0.7013 0.8375
No log 3.8333 92 1.0387 0.2769 1.0387 1.0192
No log 3.9167 94 1.0248 0.2958 1.0248 1.0123
No log 4.0 96 0.7734 0.4081 0.7734 0.8795
No log 4.0833 98 0.6632 0.3434 0.6632 0.8143
No log 4.1667 100 0.7174 0.3838 0.7174 0.8470
No log 4.25 102 0.7545 0.3557 0.7545 0.8686
No log 4.3333 104 0.6698 0.3683 0.6698 0.8184
No log 4.4167 106 0.7193 0.3326 0.7193 0.8481
No log 4.5 108 0.8934 0.3260 0.8934 0.9452
No log 4.5833 110 0.8777 0.3348 0.8777 0.9369
No log 4.6667 112 0.7026 0.3670 0.7026 0.8382
No log 4.75 114 0.6470 0.3960 0.6470 0.8044
No log 4.8333 116 0.9426 0.2568 0.9426 0.9709
No log 4.9167 118 1.0292 0.2978 1.0292 1.0145
No log 5.0 120 0.7823 0.3523 0.7823 0.8845
No log 5.0833 122 0.6448 0.4025 0.6448 0.8030
No log 5.1667 124 0.6465 0.4656 0.6465 0.8041
No log 5.25 126 0.6816 0.5824 0.6816 0.8256
No log 5.3333 128 0.6532 0.5458 0.6532 0.8082
No log 5.4167 130 0.7111 0.4362 0.7111 0.8433
No log 5.5 132 0.8758 0.3795 0.8758 0.9358
No log 5.5833 134 0.7997 0.4247 0.7997 0.8943
No log 5.6667 136 0.7094 0.4194 0.7094 0.8423
No log 5.75 138 0.6399 0.5520 0.6399 0.7999
No log 5.8333 140 0.6549 0.5085 0.6549 0.8093
No log 5.9167 142 0.6549 0.5071 0.6549 0.8093
No log 6.0 144 0.6376 0.5268 0.6376 0.7985
No log 6.0833 146 0.6611 0.5073 0.6611 0.8131
No log 6.1667 148 0.7041 0.4948 0.7042 0.8391
No log 6.25 150 0.6773 0.5330 0.6773 0.8230
No log 6.3333 152 0.8313 0.4357 0.8313 0.9118
No log 6.4167 154 0.9966 0.3724 0.9966 0.9983
No log 6.5 156 0.9042 0.3668 0.9042 0.9509
No log 6.5833 158 0.8096 0.4361 0.8096 0.8998
No log 6.6667 160 0.6611 0.5661 0.6611 0.8131
No log 6.75 162 0.6517 0.5350 0.6517 0.8073
No log 6.8333 164 0.6501 0.5631 0.6501 0.8063
No log 6.9167 166 0.6535 0.5588 0.6535 0.8084
No log 7.0 168 0.6784 0.5486 0.6784 0.8236
No log 7.0833 170 0.7077 0.5231 0.7077 0.8412
No log 7.1667 172 0.7449 0.4898 0.7449 0.8630
No log 7.25 174 0.7574 0.4842 0.7574 0.8703
No log 7.3333 176 0.6732 0.5354 0.6732 0.8205
No log 7.4167 178 0.6446 0.4930 0.6446 0.8029
No log 7.5 180 0.6483 0.4628 0.6483 0.8052
No log 7.5833 182 0.6850 0.4584 0.6850 0.8277
No log 7.6667 184 0.7420 0.4972 0.7420 0.8614
No log 7.75 186 0.7637 0.5148 0.7637 0.8739
No log 7.8333 188 0.6851 0.4852 0.6851 0.8277
No log 7.9167 190 0.7527 0.4697 0.7527 0.8676
No log 8.0 192 0.8888 0.4367 0.8888 0.9427
No log 8.0833 194 0.8187 0.4564 0.8187 0.9048
No log 8.1667 196 0.6425 0.5668 0.6425 0.8016
No log 8.25 198 0.6256 0.5163 0.6256 0.7910
No log 8.3333 200 0.6188 0.5483 0.6188 0.7867
No log 8.4167 202 0.6373 0.5440 0.6373 0.7983
No log 8.5 204 0.6195 0.5483 0.6195 0.7871
No log 8.5833 206 0.6369 0.4946 0.6369 0.7981
No log 8.6667 208 0.6213 0.5132 0.6213 0.7882
No log 8.75 210 0.6422 0.5295 0.6422 0.8014
No log 8.8333 212 0.7376 0.4508 0.7376 0.8589
No log 8.9167 214 0.6829 0.4551 0.6829 0.8264
No log 9.0 216 0.6184 0.5094 0.6184 0.7864
No log 9.0833 218 0.6296 0.4756 0.6296 0.7935
No log 9.1667 220 0.6192 0.5079 0.6192 0.7869
No log 9.25 222 0.6157 0.5079 0.6157 0.7847
No log 9.3333 224 0.6765 0.4367 0.6765 0.8225
No log 9.4167 226 0.6978 0.4496 0.6978 0.8354
No log 9.5 228 0.6441 0.4932 0.6441 0.8026
No log 9.5833 230 0.6879 0.5131 0.6879 0.8294
No log 9.6667 232 0.7189 0.5145 0.7189 0.8479
No log 9.75 234 0.7042 0.4987 0.7042 0.8392
No log 9.8333 236 0.7206 0.4835 0.7206 0.8489
No log 9.9167 238 0.7103 0.4846 0.7103 0.8428
No log 10.0 240 0.6893 0.4682 0.6893 0.8302
No log 10.0833 242 0.6731 0.4979 0.6731 0.8204
No log 10.1667 244 0.6713 0.5455 0.6713 0.8193
No log 10.25 246 0.7021 0.4988 0.7021 0.8379
No log 10.3333 248 0.6555 0.5309 0.6555 0.8096
No log 10.4167 250 0.6166 0.4957 0.6166 0.7852
No log 10.5 252 0.6146 0.4706 0.6146 0.7839
No log 10.5833 254 0.6277 0.4594 0.6277 0.7923
No log 10.6667 256 0.6230 0.4998 0.6230 0.7893
No log 10.75 258 0.6468 0.5747 0.6468 0.8042
No log 10.8333 260 0.6346 0.5632 0.6346 0.7966
No log 10.9167 262 0.6170 0.5373 0.6170 0.7855
No log 11.0 264 0.6176 0.4540 0.6176 0.7858
No log 11.0833 266 0.6031 0.4805 0.6031 0.7766
No log 11.1667 268 0.6315 0.5505 0.6315 0.7947
No log 11.25 270 0.6749 0.4609 0.6749 0.8215
No log 11.3333 272 0.6423 0.5083 0.6423 0.8015
No log 11.4167 274 0.6125 0.4756 0.6125 0.7826
No log 11.5 276 0.6402 0.4878 0.6402 0.8001
No log 11.5833 278 0.6380 0.5029 0.6380 0.7988
No log 11.6667 280 0.6917 0.5048 0.6917 0.8317
No log 11.75 282 0.7263 0.5061 0.7263 0.8522
No log 11.8333 284 0.7102 0.4696 0.7102 0.8427
No log 11.9167 286 0.6538 0.5034 0.6538 0.8086
No log 12.0 288 0.6226 0.4476 0.6226 0.7890
No log 12.0833 290 0.6243 0.4661 0.6243 0.7901
No log 12.1667 292 0.6322 0.4952 0.6322 0.7951
No log 12.25 294 0.6868 0.5139 0.6868 0.8287
No log 12.3333 296 0.8141 0.4239 0.8141 0.9023
No log 12.4167 298 0.7944 0.4320 0.7944 0.8913
No log 12.5 300 0.7237 0.5090 0.7237 0.8507
No log 12.5833 302 0.7030 0.4844 0.7030 0.8384
No log 12.6667 304 0.6155 0.5013 0.6155 0.7846
No log 12.75 306 0.5906 0.4786 0.5906 0.7685
No log 12.8333 308 0.5881 0.5515 0.5881 0.7669
No log 12.9167 310 0.5945 0.5331 0.5945 0.7711
No log 13.0 312 0.6652 0.4830 0.6652 0.8156
No log 13.0833 314 0.6868 0.4809 0.6868 0.8287
No log 13.1667 316 0.6258 0.5416 0.6258 0.7910
No log 13.25 318 0.5892 0.5554 0.5892 0.7676
No log 13.3333 320 0.5909 0.5282 0.5909 0.7687
No log 13.4167 322 0.6036 0.5403 0.6036 0.7769
No log 13.5 324 0.6022 0.5403 0.6022 0.7760
No log 13.5833 326 0.6206 0.4951 0.6206 0.7878
No log 13.6667 328 0.5991 0.4902 0.5991 0.7740
No log 13.75 330 0.5847 0.5284 0.5847 0.7646
No log 13.8333 332 0.5860 0.4941 0.5860 0.7655
No log 13.9167 334 0.5974 0.4806 0.5974 0.7729
No log 14.0 336 0.6033 0.4936 0.6033 0.7767
No log 14.0833 338 0.6070 0.4895 0.6070 0.7791
No log 14.1667 340 0.5942 0.5120 0.5942 0.7708
No log 14.25 342 0.6208 0.4937 0.6208 0.7879
No log 14.3333 344 0.6182 0.4930 0.6182 0.7863
No log 14.4167 346 0.6060 0.5006 0.6060 0.7785
No log 14.5 348 0.6345 0.4794 0.6345 0.7965
No log 14.5833 350 0.6255 0.4772 0.6255 0.7909
No log 14.6667 352 0.6019 0.4547 0.6019 0.7758
No log 14.75 354 0.6017 0.4482 0.6017 0.7757
No log 14.8333 356 0.6041 0.4482 0.6041 0.7772
No log 14.9167 358 0.6086 0.4121 0.6086 0.7801
No log 15.0 360 0.6279 0.4716 0.6279 0.7924
No log 15.0833 362 0.6657 0.4369 0.6657 0.8159
No log 15.1667 364 0.7237 0.4245 0.7237 0.8507
No log 15.25 366 0.7221 0.5038 0.7221 0.8497
No log 15.3333 368 0.6745 0.4572 0.6745 0.8213
No log 15.4167 370 0.6586 0.4775 0.6586 0.8115
No log 15.5 372 0.6841 0.4485 0.6841 0.8271
No log 15.5833 374 0.6993 0.4175 0.6993 0.8363
No log 15.6667 376 0.7489 0.4351 0.7489 0.8654
No log 15.75 378 0.8769 0.3973 0.8769 0.9364
No log 15.8333 380 0.8692 0.3973 0.8692 0.9323
No log 15.9167 382 0.7188 0.4324 0.7188 0.8478
No log 16.0 384 0.5977 0.4529 0.5977 0.7731
No log 16.0833 386 0.6033 0.4608 0.6033 0.7767
No log 16.1667 388 0.5916 0.4799 0.5916 0.7692
No log 16.25 390 0.5981 0.5020 0.5981 0.7733
No log 16.3333 392 0.6713 0.5319 0.6713 0.8193
No log 16.4167 394 0.6705 0.5319 0.6705 0.8188
No log 16.5 396 0.6206 0.5342 0.6206 0.7878
No log 16.5833 398 0.5974 0.5049 0.5974 0.7729
No log 16.6667 400 0.6055 0.4886 0.6055 0.7781
No log 16.75 402 0.5944 0.5164 0.5944 0.7710
No log 16.8333 404 0.6073 0.5538 0.6073 0.7793
No log 16.9167 406 0.6431 0.5359 0.6431 0.8019
No log 17.0 408 0.6304 0.5495 0.6304 0.7940
No log 17.0833 410 0.6013 0.5545 0.6013 0.7754
No log 17.1667 412 0.6097 0.5784 0.6097 0.7808
No log 17.25 414 0.6116 0.5697 0.6116 0.7820
No log 17.3333 416 0.6448 0.5816 0.6448 0.8030
No log 17.4167 418 0.6769 0.5136 0.6769 0.8227
No log 17.5 420 0.6410 0.6066 0.6410 0.8006
No log 17.5833 422 0.5992 0.5842 0.5992 0.7741
No log 17.6667 424 0.5986 0.5493 0.5986 0.7737
No log 17.75 426 0.5863 0.5290 0.5863 0.7657
No log 17.8333 428 0.5822 0.4820 0.5822 0.7630
No log 17.9167 430 0.6010 0.5077 0.6010 0.7752
No log 18.0 432 0.6915 0.4393 0.6915 0.8315
No log 18.0833 434 0.7289 0.4347 0.7289 0.8537
No log 18.1667 436 0.7047 0.4159 0.7047 0.8395
No log 18.25 438 0.6510 0.4993 0.6510 0.8069
No log 18.3333 440 0.5995 0.5345 0.5995 0.7743
No log 18.4167 442 0.6602 0.4139 0.6602 0.8125
No log 18.5 444 0.7078 0.3987 0.7078 0.8413
No log 18.5833 446 0.6590 0.3775 0.6590 0.8118
No log 18.6667 448 0.6223 0.5063 0.6223 0.7889
No log 18.75 450 0.6277 0.5804 0.6277 0.7923
No log 18.8333 452 0.6315 0.5804 0.6315 0.7947
No log 18.9167 454 0.6326 0.5112 0.6326 0.7953
No log 19.0 456 0.6475 0.4677 0.6475 0.8047
No log 19.0833 458 0.6475 0.4408 0.6475 0.8047
No log 19.1667 460 0.6318 0.5112 0.6318 0.7948
No log 19.25 462 0.6240 0.5363 0.6240 0.7899
No log 19.3333 464 0.6182 0.5908 0.6182 0.7862
No log 19.4167 466 0.6467 0.4643 0.6467 0.8042
No log 19.5 468 0.6522 0.4274 0.6522 0.8076
No log 19.5833 470 0.6357 0.4930 0.6357 0.7973
No log 19.6667 472 0.6390 0.4758 0.6390 0.7994
No log 19.75 474 0.6594 0.4334 0.6594 0.8120
No log 19.8333 476 0.6268 0.4765 0.6268 0.7917
No log 19.9167 478 0.5982 0.5090 0.5982 0.7734
No log 20.0 480 0.5908 0.5180 0.5908 0.7686
No log 20.0833 482 0.5887 0.5169 0.5887 0.7673
No log 20.1667 484 0.6102 0.4799 0.6102 0.7811
No log 20.25 486 0.6251 0.4821 0.6251 0.7906
No log 20.3333 488 0.6032 0.5035 0.6032 0.7766
No log 20.4167 490 0.5900 0.5575 0.5900 0.7681
No log 20.5 492 0.5840 0.5575 0.5840 0.7642
No log 20.5833 494 0.5872 0.5403 0.5872 0.7663
No log 20.6667 496 0.6020 0.5325 0.6020 0.7759
No log 20.75 498 0.6166 0.5617 0.6166 0.7852
0.3658 20.8333 500 0.6200 0.5557 0.6200 0.7874
0.3658 20.9167 502 0.6188 0.5600 0.6187 0.7866
0.3658 21.0 504 0.6346 0.5164 0.6346 0.7966
0.3658 21.0833 506 0.6454 0.5153 0.6454 0.8034
0.3658 21.1667 508 0.6409 0.4765 0.6409 0.8006
0.3658 21.25 510 0.6344 0.5126 0.6344 0.7965
0.3658 21.3333 512 0.6396 0.4892 0.6396 0.7998

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
135M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported third-party Inference Providers, and the model is not deployed on the HF Inference API.

Model tree for MayBashendy/ArabicNewSplits8_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k4_task2_organization

Finetuned
(4206)
this model