ArabicNewSplits7_OSS_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k1_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: 0.8404
  • Qwk: 0.5057
  • Mse: 0.8404
  • Rmse: 0.9167

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 0.3333 2 4.5524 -0.0132 4.5524 2.1336
No log 0.6667 4 3.5235 0.0165 3.5235 1.8771
No log 1.0 6 1.7990 0.1273 1.7990 1.3413
No log 1.3333 8 1.2878 0.0651 1.2878 1.1348
No log 1.6667 10 1.1413 0.3258 1.1413 1.0683
No log 2.0 12 1.2046 0.1753 1.2046 1.0975
No log 2.3333 14 1.2226 0.0796 1.2226 1.1057
No log 2.6667 16 1.2065 0.1691 1.2065 1.0984
No log 3.0 18 1.2018 0.1792 1.2018 1.0963
No log 3.3333 20 1.2635 0.2492 1.2635 1.1240
No log 3.6667 22 1.1952 0.2049 1.1952 1.0932
No log 4.0 24 1.1400 0.2596 1.1400 1.0677
No log 4.3333 26 1.1218 0.3292 1.1218 1.0592
No log 4.6667 28 1.1095 0.3053 1.1095 1.0534
No log 5.0 30 1.3822 0.3845 1.3822 1.1757
No log 5.3333 32 1.5571 0.3384 1.5571 1.2478
No log 5.6667 34 1.2146 0.4142 1.2146 1.1021
No log 6.0 36 1.1047 0.4487 1.1047 1.0510
No log 6.3333 38 1.0749 0.4493 1.0749 1.0368
No log 6.6667 40 1.1195 0.3517 1.1195 1.0581
No log 7.0 42 1.6989 0.3334 1.6989 1.3034
No log 7.3333 44 1.6383 0.3334 1.6383 1.2800
No log 7.6667 46 1.0951 0.4203 1.0951 1.0465
No log 8.0 48 0.9710 0.4599 0.9710 0.9854
No log 8.3333 50 1.0557 0.5088 1.0557 1.0275
No log 8.6667 52 0.9166 0.5142 0.9166 0.9574
No log 9.0 54 1.1935 0.3590 1.1935 1.0925
No log 9.3333 56 1.4661 0.3550 1.4661 1.2108
No log 9.6667 58 1.2139 0.3590 1.2139 1.1018
No log 10.0 60 0.9279 0.5557 0.9279 0.9633
No log 10.3333 62 0.9348 0.4707 0.9348 0.9668
No log 10.6667 64 0.9525 0.5647 0.9525 0.9759
No log 11.0 66 1.1260 0.3204 1.1260 1.0611
No log 11.3333 68 1.2309 0.3110 1.2309 1.1094
No log 11.6667 70 1.0255 0.5090 1.0255 1.0127
No log 12.0 72 0.9942 0.4660 0.9942 0.9971
No log 12.3333 74 0.9997 0.4668 0.9997 0.9999
No log 12.6667 76 0.9682 0.5110 0.9682 0.9840
No log 13.0 78 1.2375 0.3306 1.2375 1.1124
No log 13.3333 80 1.3444 0.3372 1.3444 1.1595
No log 13.6667 82 1.1124 0.3665 1.1124 1.0547
No log 14.0 84 0.9372 0.5159 0.9372 0.9681
No log 14.3333 86 1.0618 0.4573 1.0618 1.0305
No log 14.6667 88 1.0966 0.4556 1.0966 1.0472
No log 15.0 90 0.9863 0.4783 0.9863 0.9931
No log 15.3333 92 1.0139 0.4994 1.0139 1.0069
No log 15.6667 94 1.2809 0.3234 1.2809 1.1318
No log 16.0 96 1.2312 0.3423 1.2312 1.1096
No log 16.3333 98 1.1606 0.4170 1.1606 1.0773
No log 16.6667 100 0.9992 0.5620 0.9992 0.9996
No log 17.0 102 0.9286 0.5198 0.9286 0.9636
No log 17.3333 104 0.8988 0.5376 0.8988 0.9480
No log 17.6667 106 0.8981 0.4613 0.8981 0.9477
No log 18.0 108 0.8686 0.5214 0.8686 0.9320
No log 18.3333 110 0.9392 0.5545 0.9392 0.9691
No log 18.6667 112 0.9289 0.5367 0.9289 0.9638
No log 19.0 114 0.8732 0.5249 0.8732 0.9344
No log 19.3333 116 0.8963 0.4853 0.8963 0.9467
No log 19.6667 118 0.9244 0.4741 0.9244 0.9615
No log 20.0 120 0.9632 0.4485 0.9632 0.9814
No log 20.3333 122 0.9811 0.4485 0.9811 0.9905
No log 20.6667 124 0.9560 0.4715 0.9560 0.9777
No log 21.0 126 0.9699 0.4723 0.9699 0.9849
No log 21.3333 128 0.9935 0.4732 0.9935 0.9968
No log 21.6667 130 0.9602 0.5283 0.9602 0.9799
No log 22.0 132 0.9393 0.5507 0.9393 0.9692
No log 22.3333 134 0.9606 0.4812 0.9606 0.9801
No log 22.6667 136 0.9344 0.5335 0.9344 0.9667
No log 23.0 138 0.9380 0.4727 0.9380 0.9685
No log 23.3333 140 0.9513 0.4472 0.9513 0.9753
No log 23.6667 142 0.9178 0.4812 0.9178 0.9580
No log 24.0 144 0.9148 0.5582 0.9148 0.9564
No log 24.3333 146 0.9723 0.5549 0.9723 0.9861
No log 24.6667 148 0.9850 0.5283 0.9850 0.9925
No log 25.0 150 0.9430 0.5311 0.9430 0.9711
No log 25.3333 152 0.9634 0.5116 0.9634 0.9815
No log 25.6667 154 0.9746 0.4969 0.9746 0.9872
No log 26.0 156 0.9677 0.4982 0.9677 0.9837
No log 26.3333 158 0.9545 0.4747 0.9545 0.9770
No log 26.6667 160 0.9327 0.4597 0.9327 0.9658
No log 27.0 162 0.9342 0.4236 0.9342 0.9665
No log 27.3333 164 0.9460 0.4236 0.9460 0.9726
No log 27.6667 166 0.9194 0.4716 0.9194 0.9588
No log 28.0 168 0.9143 0.4767 0.9143 0.9562
No log 28.3333 170 0.9166 0.5009 0.9166 0.9574
No log 28.6667 172 0.9185 0.4716 0.9185 0.9584
No log 29.0 174 0.9519 0.5430 0.9519 0.9756
No log 29.3333 176 0.9286 0.4983 0.9286 0.9636
No log 29.6667 178 0.8910 0.4694 0.8910 0.9439
No log 30.0 180 0.8973 0.4846 0.8973 0.9473
No log 30.3333 182 0.9073 0.4846 0.9073 0.9525
No log 30.6667 184 0.9136 0.4864 0.9136 0.9558
No log 31.0 186 0.9305 0.4812 0.9305 0.9646
No log 31.3333 188 1.0298 0.5232 1.0298 1.0148
No log 31.6667 190 1.0114 0.5393 1.0114 1.0057
No log 32.0 192 0.9126 0.5706 0.9126 0.9553
No log 32.3333 194 0.8436 0.5036 0.8436 0.9185
No log 32.6667 196 0.8764 0.5283 0.8764 0.9362
No log 33.0 198 0.9034 0.4910 0.9034 0.9505
No log 33.3333 200 0.8591 0.5242 0.8591 0.9269
No log 33.6667 202 0.8438 0.5176 0.8438 0.9186
No log 34.0 204 0.8630 0.5186 0.8630 0.9290
No log 34.3333 206 0.8900 0.5010 0.8900 0.9434
No log 34.6667 208 0.9362 0.4404 0.9362 0.9676
No log 35.0 210 0.9267 0.4404 0.9267 0.9626
No log 35.3333 212 0.8629 0.4818 0.8629 0.9289
No log 35.6667 214 0.8407 0.4933 0.8407 0.9169
No log 36.0 216 0.8312 0.5044 0.8312 0.9117
No log 36.3333 218 0.8226 0.4637 0.8226 0.9070
No log 36.6667 220 0.8512 0.4454 0.8512 0.9226
No log 37.0 222 0.8569 0.4774 0.8569 0.9257
No log 37.3333 224 0.8632 0.5185 0.8632 0.9291
No log 37.6667 226 0.8456 0.5238 0.8456 0.9196
No log 38.0 228 0.8443 0.5231 0.8443 0.9189
No log 38.3333 230 0.8789 0.4843 0.8789 0.9375
No log 38.6667 232 0.8845 0.5132 0.8845 0.9405
No log 39.0 234 0.8540 0.5294 0.8540 0.9241
No log 39.3333 236 0.8414 0.4966 0.8414 0.9173
No log 39.6667 238 0.8549 0.4747 0.8549 0.9246
No log 40.0 240 0.8698 0.4747 0.8698 0.9326
No log 40.3333 242 0.8858 0.5023 0.8858 0.9412
No log 40.6667 244 0.8860 0.5023 0.8860 0.9413
No log 41.0 246 0.8906 0.4604 0.8906 0.9437
No log 41.3333 248 0.9311 0.5125 0.9311 0.9649
No log 41.6667 250 0.9432 0.5227 0.9432 0.9712
No log 42.0 252 0.9129 0.5042 0.9129 0.9555
No log 42.3333 254 0.8731 0.4572 0.8731 0.9344
No log 42.6667 256 0.8460 0.4920 0.8460 0.9198
No log 43.0 258 0.8410 0.4981 0.8410 0.9171
No log 43.3333 260 0.8416 0.4746 0.8416 0.9174
No log 43.6667 262 0.8503 0.4694 0.8503 0.9221
No log 44.0 264 0.9027 0.4449 0.9027 0.9501
No log 44.3333 266 0.9658 0.5532 0.9658 0.9827
No log 44.6667 268 0.9736 0.5412 0.9736 0.9867
No log 45.0 270 0.9075 0.4444 0.9075 0.9526
No log 45.3333 272 0.8508 0.4846 0.8508 0.9224
No log 45.6667 274 0.8561 0.5621 0.8561 0.9252
No log 46.0 276 0.8584 0.5532 0.8584 0.9265
No log 46.3333 278 0.8426 0.5408 0.8426 0.9180
No log 46.6667 280 0.8505 0.4572 0.8505 0.9222
No log 47.0 282 0.8500 0.4796 0.8500 0.9220
No log 47.3333 284 0.8412 0.4704 0.8412 0.9171
No log 47.6667 286 0.8441 0.5738 0.8441 0.9187
No log 48.0 288 0.9029 0.4723 0.9029 0.9502
No log 48.3333 290 0.9091 0.4752 0.9091 0.9535
No log 48.6667 292 0.8705 0.4703 0.8705 0.9330
No log 49.0 294 0.8406 0.5738 0.8406 0.9168
No log 49.3333 296 0.8377 0.5408 0.8377 0.9152
No log 49.6667 298 0.8300 0.5408 0.8300 0.9110
No log 50.0 300 0.8240 0.5528 0.8240 0.9078
No log 50.3333 302 0.8172 0.5408 0.8172 0.9040
No log 50.6667 304 0.8144 0.5408 0.8144 0.9024
No log 51.0 306 0.8207 0.5408 0.8207 0.9059
No log 51.3333 308 0.8381 0.5439 0.8381 0.9155
No log 51.6667 310 0.8850 0.5507 0.8850 0.9407
No log 52.0 312 0.8965 0.5428 0.8965 0.9468
No log 52.3333 314 0.9023 0.5228 0.9023 0.9499
No log 52.6667 316 0.8588 0.5825 0.8588 0.9267
No log 53.0 318 0.8086 0.5621 0.8086 0.8992
No log 53.3333 320 0.7987 0.5287 0.7987 0.8937
No log 53.6667 322 0.8528 0.4449 0.8528 0.9235
No log 54.0 324 0.9083 0.4201 0.9083 0.9530
No log 54.3333 326 0.9098 0.4111 0.9098 0.9538
No log 54.6667 328 0.8651 0.4672 0.8651 0.9301
No log 55.0 330 0.8333 0.5287 0.8333 0.9128
No log 55.3333 332 0.8204 0.5408 0.8204 0.9057
No log 55.6667 334 0.8250 0.5408 0.8250 0.9083
No log 56.0 336 0.8277 0.5408 0.8277 0.9098
No log 56.3333 338 0.8342 0.5287 0.8342 0.9133
No log 56.6667 340 0.8304 0.5391 0.8304 0.9113
No log 57.0 342 0.8165 0.5391 0.8165 0.9036
No log 57.3333 344 0.7966 0.5621 0.7966 0.8925
No log 57.6667 346 0.7816 0.5621 0.7816 0.8841
No log 58.0 348 0.7715 0.5057 0.7715 0.8783
No log 58.3333 350 0.7739 0.4694 0.7739 0.8797
No log 58.6667 352 0.7724 0.5043 0.7724 0.8789
No log 59.0 354 0.7717 0.5107 0.7717 0.8785
No log 59.3333 356 0.7773 0.5381 0.7773 0.8816
No log 59.6667 358 0.7748 0.5233 0.7748 0.8802
No log 60.0 360 0.7740 0.5287 0.7740 0.8798
No log 60.3333 362 0.7943 0.5028 0.7943 0.8912
No log 60.6667 364 0.8466 0.5359 0.8466 0.9201
No log 61.0 366 0.8800 0.5290 0.8800 0.9381
No log 61.3333 368 0.8764 0.5041 0.8764 0.9362
No log 61.6667 370 0.8485 0.4816 0.8485 0.9211
No log 62.0 372 0.8209 0.5176 0.8209 0.9060
No log 62.3333 374 0.8182 0.5528 0.8182 0.9046
No log 62.6667 376 0.8207 0.5528 0.8207 0.9059
No log 63.0 378 0.8155 0.5176 0.8155 0.9031
No log 63.3333 380 0.8133 0.5057 0.8133 0.9018
No log 63.6667 382 0.8086 0.5057 0.8086 0.8992
No log 64.0 384 0.8077 0.5275 0.8077 0.8987
No log 64.3333 386 0.7994 0.5391 0.7994 0.8941
No log 64.6667 388 0.7977 0.5621 0.7977 0.8932
No log 65.0 390 0.8109 0.5791 0.8109 0.9005
No log 65.3333 392 0.8205 0.5443 0.8205 0.9058
No log 65.6667 394 0.8125 0.5089 0.8125 0.9014
No log 66.0 396 0.8028 0.5458 0.8028 0.8960
No log 66.3333 398 0.8001 0.5408 0.8001 0.8945
No log 66.6667 400 0.8159 0.5057 0.8159 0.9033
No log 67.0 402 0.8381 0.4685 0.8381 0.9155
No log 67.3333 404 0.8651 0.4565 0.8651 0.9301
No log 67.6667 406 0.8703 0.4565 0.8703 0.9329
No log 68.0 408 0.8610 0.4685 0.8610 0.9279
No log 68.3333 410 0.8411 0.5057 0.8411 0.9171
No log 68.6667 412 0.8264 0.5057 0.8264 0.9091
No log 69.0 414 0.8202 0.5057 0.8202 0.9057
No log 69.3333 416 0.8154 0.5287 0.8154 0.9030
No log 69.6667 418 0.8125 0.5287 0.8125 0.9014
No log 70.0 420 0.8157 0.5287 0.8157 0.9032
No log 70.3333 422 0.8222 0.5057 0.8222 0.9067
No log 70.6667 424 0.8288 0.5057 0.8288 0.9104
No log 71.0 426 0.8338 0.5057 0.8338 0.9131
No log 71.3333 428 0.8369 0.5176 0.8369 0.9148
No log 71.6667 430 0.8449 0.5057 0.8449 0.9192
No log 72.0 432 0.8459 0.5057 0.8459 0.9197
No log 72.3333 434 0.8417 0.5176 0.8417 0.9175
No log 72.6667 436 0.8405 0.5176 0.8405 0.9168
No log 73.0 438 0.8469 0.5596 0.8469 0.9202
No log 73.3333 440 0.8507 0.5422 0.8507 0.9223
No log 73.6667 442 0.8466 0.5391 0.8466 0.9201
No log 74.0 444 0.8414 0.5176 0.8414 0.9173
No log 74.3333 446 0.8431 0.5176 0.8431 0.9182
No log 74.6667 448 0.8520 0.4937 0.8520 0.9230
No log 75.0 450 0.8550 0.4937 0.8550 0.9247
No log 75.3333 452 0.8525 0.5176 0.8525 0.9233
No log 75.6667 454 0.8508 0.5176 0.8508 0.9224
No log 76.0 456 0.8471 0.5176 0.8471 0.9204
No log 76.3333 458 0.8430 0.5176 0.8430 0.9182
No log 76.6667 460 0.8402 0.5176 0.8402 0.9166
No log 77.0 462 0.8343 0.5176 0.8343 0.9134
No log 77.3333 464 0.8298 0.5176 0.8298 0.9109
No log 77.6667 466 0.8257 0.5176 0.8257 0.9087
No log 78.0 468 0.8210 0.5176 0.8210 0.9061
No log 78.3333 470 0.8174 0.5408 0.8174 0.9041
No log 78.6667 472 0.8159 0.5408 0.8159 0.9033
No log 79.0 474 0.8172 0.5408 0.8172 0.9040
No log 79.3333 476 0.8164 0.5408 0.8164 0.9035
No log 79.6667 478 0.8160 0.5408 0.8160 0.9033
No log 80.0 480 0.8173 0.5287 0.8173 0.9041
No log 80.3333 482 0.8215 0.4816 0.8215 0.9064
No log 80.6667 484 0.8247 0.4816 0.8247 0.9081
No log 81.0 486 0.8250 0.4816 0.8250 0.9083
No log 81.3333 488 0.8281 0.4816 0.8281 0.9100
No log 81.6667 490 0.8296 0.5057 0.8296 0.9108
No log 82.0 492 0.8301 0.5057 0.8301 0.9111
No log 82.3333 494 0.8297 0.5176 0.8297 0.9109
No log 82.6667 496 0.8324 0.5176 0.8324 0.9123
No log 83.0 498 0.8357 0.5176 0.8357 0.9142
0.2041 83.3333 500 0.8380 0.5176 0.8380 0.9154
0.2041 83.6667 502 0.8407 0.5176 0.8407 0.9169
0.2041 84.0 504 0.8398 0.5176 0.8398 0.9164
0.2041 84.3333 506 0.8376 0.5176 0.8376 0.9152
0.2041 84.6667 508 0.8376 0.5176 0.8376 0.9152
0.2041 85.0 510 0.8404 0.5057 0.8404 0.9167

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0+cu118
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
1
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/ArabicNewSplits7_OSS_usingWellWrittenEssays_FineTuningAraBERT_run3_AugV5_k1_task2_organization

Finetuned
(4019)
this model