ArabicNewSplits7_FineTuningAraBERT_run1_AugV5_k9_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: 0.5240
  • Qwk: 0.5397
  • Mse: 0.5240
  • Rmse: 0.7239

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 2.3861 -0.0262 2.3861 1.5447
No log 0.1667 4 1.0688 0.2508 1.0688 1.0338
No log 0.25 6 1.2133 -0.1162 1.2133 1.1015
No log 0.3333 8 1.5554 -0.1924 1.5554 1.2471
No log 0.4167 10 1.1353 -0.0622 1.1353 1.0655
No log 0.5 12 0.9150 0.0 0.9150 0.9566
No log 0.5833 14 0.9082 0.0541 0.9082 0.9530
No log 0.6667 16 0.9328 0.1416 0.9328 0.9658
No log 0.75 18 1.0116 0.2412 1.0116 1.0058
No log 0.8333 20 0.9747 0.1277 0.9747 0.9873
No log 0.9167 22 0.8724 0.0 0.8724 0.9340
No log 1.0 24 0.9295 0.0 0.9295 0.9641
No log 1.0833 26 0.9671 0.1345 0.9671 0.9834
No log 1.1667 28 1.0301 0.2533 1.0301 1.0149
No log 1.25 30 0.9275 0.2408 0.9275 0.9631
No log 1.3333 32 0.7627 0.0 0.7627 0.8733
No log 1.4167 34 0.7109 0.0 0.7109 0.8432
No log 1.5 36 0.7444 0.0 0.7444 0.8628
No log 1.5833 38 0.7567 0.0 0.7567 0.8699
No log 1.6667 40 0.7563 0.0481 0.7563 0.8697
No log 1.75 42 0.7223 -0.0027 0.7223 0.8499
No log 1.8333 44 0.7282 0.0444 0.7282 0.8533
No log 1.9167 46 0.7845 0.2132 0.7845 0.8857
No log 2.0 48 0.8331 0.3131 0.8331 0.9128
No log 2.0833 50 0.8155 0.2736 0.8155 0.9031
No log 2.1667 52 0.7107 0.1729 0.7107 0.8430
No log 2.25 54 0.6970 0.1365 0.6970 0.8349
No log 2.3333 56 0.6822 0.1321 0.6822 0.8260
No log 2.4167 58 0.6988 0.2118 0.6988 0.8360
No log 2.5 60 0.7431 0.2087 0.7431 0.8620
No log 2.5833 62 0.8569 0.3192 0.8569 0.9257
No log 2.6667 64 0.8842 0.3601 0.8842 0.9403
No log 2.75 66 0.7329 0.2992 0.7329 0.8561
No log 2.8333 68 0.6165 0.3862 0.6165 0.7852
No log 2.9167 70 0.7260 0.3637 0.7260 0.8521
No log 3.0 72 0.7028 0.3637 0.7028 0.8383
No log 3.0833 74 0.6228 0.2744 0.6228 0.7892
No log 3.1667 76 0.9057 0.3892 0.9057 0.9517
No log 3.25 78 0.9123 0.3892 0.9123 0.9552
No log 3.3333 80 0.6551 0.3135 0.6551 0.8094
No log 3.4167 82 0.6258 0.4103 0.6258 0.7911
No log 3.5 84 0.6210 0.3864 0.6210 0.7880
No log 3.5833 86 0.6178 0.3581 0.6178 0.7860
No log 3.6667 88 0.6578 0.3492 0.6578 0.8110
No log 3.75 90 0.6617 0.2973 0.6617 0.8135
No log 3.8333 92 0.7579 0.3384 0.7579 0.8706
No log 3.9167 94 0.9451 0.3538 0.9451 0.9721
No log 4.0 96 0.8619 0.4064 0.8619 0.9284
No log 4.0833 98 0.9659 0.3274 0.9659 0.9828
No log 4.1667 100 1.2205 0.3257 1.2205 1.1048
No log 4.25 102 0.8914 0.3377 0.8914 0.9441
No log 4.3333 104 0.7829 0.3343 0.7829 0.8848
No log 4.4167 106 0.6971 0.4640 0.6971 0.8349
No log 4.5 108 0.6775 0.4137 0.6775 0.8231
No log 4.5833 110 0.7189 0.3746 0.7189 0.8479
No log 4.6667 112 1.0207 0.3460 1.0207 1.0103
No log 4.75 114 0.9866 0.3517 0.9866 0.9933
No log 4.8333 116 0.7478 0.3606 0.7478 0.8648
No log 4.9167 118 0.6217 0.3713 0.6217 0.7885
No log 5.0 120 0.6146 0.4229 0.6146 0.7840
No log 5.0833 122 0.6523 0.3032 0.6523 0.8076
No log 5.1667 124 0.7586 0.4154 0.7586 0.8710
No log 5.25 126 0.7891 0.3287 0.7891 0.8883
No log 5.3333 128 0.8266 0.3228 0.8266 0.9092
No log 5.4167 130 0.7153 0.3473 0.7153 0.8457
No log 5.5 132 0.6211 0.3950 0.6211 0.7881
No log 5.5833 134 0.6230 0.3950 0.6230 0.7893
No log 5.6667 136 0.6130 0.4919 0.6130 0.7829
No log 5.75 138 0.6446 0.4764 0.6446 0.8029
No log 5.8333 140 0.5738 0.4763 0.5738 0.7575
No log 5.9167 142 0.5840 0.4349 0.5840 0.7642
No log 6.0 144 0.5840 0.4486 0.5840 0.7642
No log 6.0833 146 0.5588 0.4934 0.5588 0.7476
No log 6.1667 148 0.5758 0.4234 0.5758 0.7588
No log 6.25 150 0.5744 0.4314 0.5744 0.7579
No log 6.3333 152 0.5929 0.4762 0.5929 0.7700
No log 6.4167 154 0.5972 0.4747 0.5972 0.7728
No log 6.5 156 0.6933 0.3746 0.6933 0.8326
No log 6.5833 158 0.6784 0.3494 0.6784 0.8236
No log 6.6667 160 0.6313 0.3865 0.6313 0.7945
No log 6.75 162 0.6417 0.3475 0.6417 0.8011
No log 6.8333 164 0.6380 0.4482 0.6380 0.7987
No log 6.9167 166 0.6554 0.3833 0.6554 0.8096
No log 7.0 168 0.6561 0.3994 0.6561 0.8100
No log 7.0833 170 0.6338 0.4423 0.6338 0.7961
No log 7.1667 172 0.6704 0.4186 0.6704 0.8188
No log 7.25 174 0.7031 0.4464 0.7031 0.8385
No log 7.3333 176 0.6456 0.4044 0.6456 0.8035
No log 7.4167 178 0.6423 0.4362 0.6423 0.8015
No log 7.5 180 0.6253 0.3859 0.6253 0.7908
No log 7.5833 182 0.6246 0.4724 0.6246 0.7903
No log 7.6667 184 0.6984 0.4502 0.6984 0.8357
No log 7.75 186 0.7873 0.4080 0.7873 0.8873
No log 7.8333 188 0.7232 0.4562 0.7232 0.8504
No log 7.9167 190 0.5980 0.5115 0.5980 0.7733
No log 8.0 192 0.5940 0.4283 0.5940 0.7707
No log 8.0833 194 0.6622 0.5251 0.6622 0.8138
No log 8.1667 196 0.6828 0.5251 0.6828 0.8263
No log 8.25 198 0.6211 0.4892 0.6211 0.7881
No log 8.3333 200 0.5783 0.4753 0.5783 0.7605
No log 8.4167 202 0.5689 0.5201 0.5689 0.7542
No log 8.5 204 0.5808 0.4370 0.5808 0.7621
No log 8.5833 206 0.7074 0.4076 0.7074 0.8411
No log 8.6667 208 0.8153 0.3151 0.8153 0.9029
No log 8.75 210 0.7372 0.3889 0.7372 0.8586
No log 8.8333 212 0.5944 0.4315 0.5944 0.7710
No log 8.9167 214 0.6190 0.4749 0.6190 0.7868
No log 9.0 216 0.8306 0.3657 0.8306 0.9114
No log 9.0833 218 0.9163 0.3963 0.9163 0.9572
No log 9.1667 220 0.7723 0.3665 0.7723 0.8788
No log 9.25 222 0.6372 0.4375 0.6372 0.7982
No log 9.3333 224 0.6205 0.3787 0.6205 0.7877
No log 9.4167 226 0.6123 0.3787 0.6123 0.7825
No log 9.5 228 0.6003 0.3738 0.6003 0.7748
No log 9.5833 230 0.5966 0.4547 0.5966 0.7724
No log 9.6667 232 0.5888 0.4547 0.5888 0.7673
No log 9.75 234 0.5939 0.5079 0.5939 0.7706
No log 9.8333 236 0.5920 0.4516 0.5920 0.7694
No log 9.9167 238 0.6036 0.4942 0.6036 0.7769
No log 10.0 240 0.6253 0.4895 0.6253 0.7908
No log 10.0833 242 0.6084 0.4655 0.6084 0.7800
No log 10.1667 244 0.5911 0.4314 0.5911 0.7688
No log 10.25 246 0.6083 0.4059 0.6083 0.7799
No log 10.3333 248 0.5928 0.4534 0.5928 0.7699
No log 10.4167 250 0.6151 0.3746 0.6151 0.7843
No log 10.5 252 0.6704 0.3918 0.6704 0.8188
No log 10.5833 254 0.7057 0.4030 0.7057 0.8401
No log 10.6667 256 0.6857 0.4030 0.6857 0.8281
No log 10.75 258 0.6240 0.3789 0.6240 0.7900
No log 10.8333 260 0.6103 0.2652 0.6103 0.7812
No log 10.9167 262 0.6378 0.3712 0.6378 0.7987
No log 11.0 264 0.7079 0.3918 0.7079 0.8414
No log 11.0833 266 0.6915 0.4067 0.6915 0.8316
No log 11.1667 268 0.6255 0.3789 0.6255 0.7909
No log 11.25 270 0.6154 0.3196 0.6154 0.7845
No log 11.3333 272 0.6274 0.3355 0.6274 0.7921
No log 11.4167 274 0.6254 0.3942 0.6254 0.7908
No log 11.5 276 0.6113 0.3498 0.6113 0.7819
No log 11.5833 278 0.5881 0.4970 0.5881 0.7669
No log 11.6667 280 0.5745 0.5208 0.5745 0.7580
No log 11.75 282 0.5691 0.4402 0.5691 0.7544
No log 11.8333 284 0.5765 0.5397 0.5765 0.7593
No log 11.9167 286 0.5803 0.5397 0.5803 0.7618
No log 12.0 288 0.5701 0.5208 0.5701 0.7550
No log 12.0833 290 0.5715 0.4949 0.5715 0.7560
No log 12.1667 292 0.5741 0.4949 0.5741 0.7577
No log 12.25 294 0.5827 0.5386 0.5827 0.7634
No log 12.3333 296 0.5889 0.4526 0.5889 0.7674
No log 12.4167 298 0.6442 0.3196 0.6442 0.8026
No log 12.5 300 0.6887 0.3963 0.6887 0.8299
No log 12.5833 302 0.6505 0.3544 0.6505 0.8065
No log 12.6667 304 0.6936 0.3723 0.6936 0.8328
No log 12.75 306 0.8015 0.4366 0.8015 0.8953
No log 12.8333 308 0.7400 0.3586 0.7400 0.8602
No log 12.9167 310 0.6420 0.3261 0.6420 0.8013
No log 13.0 312 0.5969 0.3387 0.5969 0.7726
No log 13.0833 314 0.5952 0.3701 0.5952 0.7715
No log 13.1667 316 0.6092 0.3545 0.6092 0.7805
No log 13.25 318 0.6454 0.4272 0.6454 0.8033
No log 13.3333 320 0.6421 0.4409 0.6421 0.8013
No log 13.4167 322 0.5894 0.3996 0.5894 0.7677
No log 13.5 324 0.5529 0.4681 0.5529 0.7436
No log 13.5833 326 0.5534 0.4660 0.5534 0.7439
No log 13.6667 328 0.5823 0.4664 0.5823 0.7631
No log 13.75 330 0.5974 0.5149 0.5974 0.7729
No log 13.8333 332 0.6155 0.4329 0.6155 0.7846
No log 13.9167 334 0.6524 0.4329 0.6524 0.8077
No log 14.0 336 0.6625 0.4067 0.6625 0.8139
No log 14.0833 338 0.6102 0.4067 0.6102 0.7812
No log 14.1667 340 0.5669 0.4337 0.5669 0.7529
No log 14.25 342 0.5603 0.5003 0.5603 0.7485
No log 14.3333 344 0.5615 0.5550 0.5615 0.7493
No log 14.4167 346 0.5827 0.4945 0.5827 0.7634
No log 14.5 348 0.6272 0.4089 0.6272 0.7920
No log 14.5833 350 0.6013 0.4414 0.6013 0.7755
No log 14.6667 352 0.5775 0.4524 0.5775 0.7599
No log 14.75 354 0.5636 0.5503 0.5636 0.7507
No log 14.8333 356 0.5600 0.5784 0.5600 0.7484
No log 14.9167 358 0.5933 0.4555 0.5933 0.7703
No log 15.0 360 0.6111 0.4684 0.6111 0.7817
No log 15.0833 362 0.5709 0.5141 0.5709 0.7556
No log 15.1667 364 0.5527 0.5488 0.5527 0.7434
No log 15.25 366 0.5581 0.5250 0.5581 0.7470
No log 15.3333 368 0.6125 0.4451 0.6125 0.7826
No log 15.4167 370 0.6496 0.3940 0.6496 0.8060
No log 15.5 372 0.6065 0.3985 0.6065 0.7788
No log 15.5833 374 0.5531 0.4828 0.5531 0.7437
No log 15.6667 376 0.5423 0.5326 0.5423 0.7364
No log 15.75 378 0.5457 0.4576 0.5457 0.7387
No log 15.8333 380 0.5756 0.4618 0.5756 0.7587
No log 15.9167 382 0.6218 0.4224 0.6218 0.7885
No log 16.0 384 0.6044 0.3712 0.6044 0.7774
No log 16.0833 386 0.5892 0.3471 0.5892 0.7676
No log 16.1667 388 0.5777 0.4124 0.5777 0.7600
No log 16.25 390 0.5761 0.4291 0.5761 0.7590
No log 16.3333 392 0.5957 0.3196 0.5957 0.7718
No log 16.4167 394 0.6097 0.3737 0.6097 0.7809
No log 16.5 396 0.6503 0.4144 0.6503 0.8064
No log 16.5833 398 0.6439 0.3894 0.6439 0.8024
No log 16.6667 400 0.6148 0.4420 0.6148 0.7841
No log 16.75 402 0.6069 0.3809 0.6069 0.7791
No log 16.8333 404 0.6189 0.3834 0.6189 0.7867
No log 16.9167 406 0.6129 0.3920 0.6129 0.7829
No log 17.0 408 0.6208 0.3839 0.6208 0.7879
No log 17.0833 410 0.7129 0.3494 0.7129 0.8443
No log 17.1667 412 0.7564 0.3665 0.7564 0.8697
No log 17.25 414 0.6839 0.4251 0.6839 0.8270
No log 17.3333 416 0.5826 0.4729 0.5826 0.7633
No log 17.4167 418 0.5604 0.4402 0.5604 0.7486
No log 17.5 420 0.5546 0.4111 0.5546 0.7447
No log 17.5833 422 0.5564 0.4729 0.5564 0.7459
No log 17.6667 424 0.5594 0.4614 0.5594 0.7480
No log 17.75 426 0.5531 0.4291 0.5531 0.7437
No log 17.8333 428 0.5495 0.4561 0.5495 0.7413
No log 17.9167 430 0.5477 0.4795 0.5477 0.7401
No log 18.0 432 0.5515 0.4677 0.5515 0.7427
No log 18.0833 434 0.5472 0.5152 0.5472 0.7397
No log 18.1667 436 0.5448 0.5133 0.5448 0.7381
No log 18.25 438 0.5419 0.5344 0.5419 0.7361
No log 18.3333 440 0.5376 0.4878 0.5376 0.7332
No log 18.4167 442 0.5438 0.5189 0.5438 0.7374
No log 18.5 444 0.5775 0.3976 0.5775 0.7599
No log 18.5833 446 0.5933 0.3673 0.5933 0.7702
No log 18.6667 448 0.5720 0.4354 0.5720 0.7563
No log 18.75 450 0.5679 0.3863 0.5679 0.7536
No log 18.8333 452 0.5695 0.3863 0.5695 0.7547
No log 18.9167 454 0.5718 0.4314 0.5718 0.7562
No log 19.0 456 0.6071 0.4076 0.6071 0.7791
No log 19.0833 458 0.6702 0.4307 0.6702 0.8186
No log 19.1667 460 0.6890 0.4307 0.6890 0.8301
No log 19.25 462 0.6662 0.4387 0.6662 0.8162
No log 19.3333 464 0.6362 0.4387 0.6362 0.7976
No log 19.4167 466 0.6018 0.4330 0.6018 0.7757
No log 19.5 468 0.6083 0.3972 0.6083 0.7799
No log 19.5833 470 0.6584 0.4307 0.6584 0.8114
No log 19.6667 472 0.6610 0.4307 0.6610 0.8130
No log 19.75 474 0.6019 0.3789 0.6019 0.7758
No log 19.8333 476 0.5505 0.3945 0.5505 0.7420
No log 19.9167 478 0.5333 0.4199 0.5333 0.7302
No log 20.0 480 0.5325 0.4468 0.5325 0.7297
No log 20.0833 482 0.5437 0.4444 0.5437 0.7373
No log 20.1667 484 0.5715 0.3622 0.5715 0.7560
No log 20.25 486 0.5930 0.3518 0.5930 0.7701
No log 20.3333 488 0.6162 0.3637 0.6162 0.7850
No log 20.4167 490 0.6303 0.4144 0.6303 0.7939
No log 20.5 492 0.5868 0.3518 0.5868 0.7660
No log 20.5833 494 0.5667 0.4100 0.5667 0.7528
No log 20.6667 496 0.5402 0.5440 0.5402 0.7350
No log 20.75 498 0.5283 0.5681 0.5283 0.7269
0.314 20.8333 500 0.5274 0.5656 0.5274 0.7262
0.314 20.9167 502 0.5289 0.5768 0.5289 0.7273
0.314 21.0 504 0.5356 0.5765 0.5356 0.7318
0.314 21.0833 506 0.5418 0.5014 0.5418 0.7361
0.314 21.1667 508 0.5474 0.5233 0.5474 0.7398
0.314 21.25 510 0.5485 0.5233 0.5485 0.7406
0.314 21.3333 512 0.5314 0.5283 0.5314 0.7290
0.314 21.4167 514 0.5235 0.6114 0.5235 0.7235
0.314 21.5 516 0.5249 0.5692 0.5249 0.7245
0.314 21.5833 518 0.5776 0.6075 0.5776 0.7600
0.314 21.6667 520 0.6001 0.5026 0.6001 0.7747
0.314 21.75 522 0.5646 0.4827 0.5646 0.7514
0.314 21.8333 524 0.5267 0.4397 0.5267 0.7257
0.314 21.9167 526 0.5137 0.5522 0.5137 0.7168
0.314 22.0 528 0.5240 0.5397 0.5240 0.7239

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0+cu118
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
8
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/ArabicNewSplits7_FineTuningAraBERT_run1_AugV5_k9_task7_organization

Finetuned
(4223)
this model