ArabicNewSplits7_FineTuningAraBERT_run3_AugV5_k1_task5_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: 0.6502
  • Qwk: 0.5578
  • Mse: 0.6502
  • Rmse: 0.8064

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 0.4 2 4.0923 0.0 4.0923 2.0229
No log 0.8 4 2.9759 -0.0242 2.9759 1.7251
No log 1.2 6 1.4468 0.0 1.4468 1.2028
No log 1.6 8 1.0334 0.2857 1.0334 1.0166
No log 2.0 10 0.9606 0.2544 0.9606 0.9801
No log 2.4 12 1.0353 0.1962 1.0353 1.0175
No log 2.8 14 1.1293 0.1361 1.1293 1.0627
No log 3.2 16 1.0317 0.1997 1.0317 1.0157
No log 3.6 18 0.8784 0.3162 0.8784 0.9373
No log 4.0 20 0.7874 0.3821 0.7874 0.8874
No log 4.4 22 0.7384 0.4660 0.7384 0.8593
No log 4.8 24 0.9148 0.3944 0.9148 0.9565
No log 5.2 26 0.9277 0.4560 0.9277 0.9632
No log 5.6 28 0.6802 0.5725 0.6802 0.8247
No log 6.0 30 0.6621 0.6197 0.6621 0.8137
No log 6.4 32 0.6916 0.6235 0.6916 0.8316
No log 6.8 34 0.7129 0.6120 0.7129 0.8444
No log 7.2 36 0.7368 0.6241 0.7368 0.8583
No log 7.6 38 0.8262 0.6089 0.8262 0.9089
No log 8.0 40 0.9690 0.4894 0.9690 0.9844
No log 8.4 42 0.8814 0.5727 0.8814 0.9388
No log 8.8 44 0.8681 0.5026 0.8681 0.9317
No log 9.2 46 0.7804 0.6019 0.7804 0.8834
No log 9.6 48 0.7676 0.6242 0.7676 0.8761
No log 10.0 50 0.6799 0.5898 0.6799 0.8245
No log 10.4 52 0.6863 0.6094 0.6863 0.8284
No log 10.8 54 0.7178 0.5815 0.7178 0.8472
No log 11.2 56 0.8726 0.5965 0.8726 0.9341
No log 11.6 58 0.8296 0.6225 0.8296 0.9108
No log 12.0 60 0.9171 0.5896 0.9171 0.9576
No log 12.4 62 0.7419 0.6239 0.7419 0.8614
No log 12.8 64 0.6620 0.6197 0.6620 0.8136
No log 13.2 66 0.6618 0.6160 0.6618 0.8135
No log 13.6 68 0.8943 0.5948 0.8943 0.9457
No log 14.0 70 0.8615 0.5896 0.8615 0.9282
No log 14.4 72 0.6340 0.6042 0.6340 0.7962
No log 14.8 74 0.6028 0.6695 0.6028 0.7764
No log 15.2 76 0.6113 0.6695 0.6113 0.7818
No log 15.6 78 0.6103 0.6022 0.6103 0.7812
No log 16.0 80 0.7449 0.5988 0.7449 0.8631
No log 16.4 82 0.7361 0.5988 0.7361 0.8579
No log 16.8 84 0.6563 0.5945 0.6563 0.8101
No log 17.2 86 0.6119 0.6160 0.6119 0.7822
No log 17.6 88 0.5946 0.6244 0.5946 0.7711
No log 18.0 90 0.6269 0.6598 0.6269 0.7917
No log 18.4 92 0.6063 0.6482 0.6063 0.7787
No log 18.8 94 0.5969 0.6383 0.5969 0.7726
No log 19.2 96 0.5764 0.6244 0.5764 0.7592
No log 19.6 98 0.6075 0.6005 0.6075 0.7794
No log 20.0 100 0.7463 0.6136 0.7463 0.8639
No log 20.4 102 0.7146 0.6126 0.7146 0.8453
No log 20.8 104 0.5959 0.6022 0.5959 0.7719
No log 21.2 106 0.6028 0.6872 0.6028 0.7764
No log 21.6 108 0.6380 0.7155 0.6380 0.7987
No log 22.0 110 0.6157 0.6215 0.6157 0.7847
No log 22.4 112 0.7477 0.6067 0.7477 0.8647
No log 22.8 114 0.9173 0.6157 0.9173 0.9578
No log 23.2 116 0.8441 0.6002 0.8441 0.9187
No log 23.6 118 0.6946 0.6051 0.6946 0.8334
No log 24.0 120 0.6811 0.6471 0.6811 0.8253
No log 24.4 122 0.6469 0.6570 0.6469 0.8043
No log 24.8 124 0.6075 0.6051 0.6075 0.7794
No log 25.2 126 0.6577 0.5536 0.6577 0.8110
No log 25.6 128 0.6863 0.6461 0.6863 0.8284
No log 26.0 130 0.6679 0.5578 0.6679 0.8172
No log 26.4 132 0.6489 0.6129 0.6489 0.8056
No log 26.8 134 0.6686 0.6129 0.6686 0.8177
No log 27.2 136 0.6978 0.5657 0.6978 0.8353
No log 27.6 138 0.8516 0.6038 0.8516 0.9228
No log 28.0 140 0.9436 0.5712 0.9436 0.9714
No log 28.4 142 0.8213 0.6022 0.8213 0.9063
No log 28.8 144 0.6700 0.5470 0.6700 0.8186
No log 29.2 146 0.6646 0.6479 0.6646 0.8152
No log 29.6 148 0.6617 0.6900 0.6617 0.8134
No log 30.0 150 0.6549 0.6112 0.6549 0.8092
No log 30.4 152 0.7518 0.6026 0.7518 0.8671
No log 30.8 154 0.7626 0.6193 0.7626 0.8733
No log 31.2 156 0.6756 0.5466 0.6756 0.8219
No log 31.6 158 0.6638 0.5859 0.6638 0.8147
No log 32.0 160 0.7096 0.6005 0.7096 0.8424
No log 32.4 162 0.7150 0.5462 0.7150 0.8455
No log 32.8 164 0.6981 0.5470 0.6981 0.8355
No log 33.2 166 0.6615 0.5955 0.6615 0.8133
No log 33.6 168 0.6568 0.5955 0.6568 0.8105
No log 34.0 170 0.6557 0.5567 0.6557 0.8098
No log 34.4 172 0.6679 0.5470 0.6679 0.8173
No log 34.8 174 0.6898 0.5275 0.6898 0.8306
No log 35.2 176 0.6970 0.5470 0.6970 0.8349
No log 35.6 178 0.7395 0.5593 0.7395 0.8600
No log 36.0 180 0.7886 0.5862 0.7886 0.8880
No log 36.4 182 0.7544 0.5698 0.7544 0.8685
No log 36.8 184 0.6773 0.5859 0.6773 0.8230
No log 37.2 186 0.6379 0.5955 0.6379 0.7987
No log 37.6 188 0.6222 0.5895 0.6222 0.7888
No log 38.0 190 0.6045 0.6051 0.6045 0.7775
No log 38.4 192 0.5981 0.6041 0.5981 0.7734
No log 38.8 194 0.6066 0.6138 0.6066 0.7789
No log 39.2 196 0.6333 0.5955 0.6333 0.7958
No log 39.6 198 0.6592 0.5859 0.6592 0.8119
No log 40.0 200 0.7012 0.5835 0.7012 0.8374
No log 40.4 202 0.6785 0.5686 0.6785 0.8237
No log 40.8 204 0.6495 0.5859 0.6495 0.8059
No log 41.2 206 0.6380 0.6341 0.6380 0.7987
No log 41.6 208 0.6315 0.6041 0.6315 0.7946
No log 42.0 210 0.6384 0.5859 0.6384 0.7990
No log 42.4 212 0.6760 0.5678 0.6760 0.8222
No log 42.8 214 0.7193 0.6575 0.7193 0.8481
No log 43.2 216 0.7171 0.6041 0.7171 0.8468
No log 43.6 218 0.6860 0.6303 0.6860 0.8282
No log 44.0 220 0.6667 0.5927 0.6667 0.8165
No log 44.4 222 0.6281 0.5955 0.6281 0.7925
No log 44.8 224 0.6088 0.5955 0.6088 0.7802
No log 45.2 226 0.6130 0.6186 0.6130 0.7829
No log 45.6 228 0.6483 0.6154 0.6483 0.8052
No log 46.0 230 0.6507 0.6154 0.6507 0.8066
No log 46.4 232 0.6339 0.5784 0.6339 0.7962
No log 46.8 234 0.6286 0.5955 0.6286 0.7928
No log 47.2 236 0.6360 0.6051 0.6360 0.7975
No log 47.6 238 0.6469 0.5567 0.6469 0.8043
No log 48.0 240 0.6574 0.5915 0.6574 0.8108
No log 48.4 242 0.6990 0.5676 0.6990 0.8361
No log 48.8 244 0.7242 0.5688 0.7242 0.8510
No log 49.2 246 0.7100 0.6014 0.7100 0.8426
No log 49.6 248 0.6500 0.5805 0.6500 0.8062
No log 50.0 250 0.6156 0.5945 0.6156 0.7846
No log 50.4 252 0.6187 0.6041 0.6187 0.7866
No log 50.8 254 0.6140 0.6041 0.6140 0.7836
No log 51.2 256 0.6229 0.6051 0.6229 0.7892
No log 51.6 258 0.6646 0.5773 0.6646 0.8153
No log 52.0 260 0.6960 0.6045 0.6960 0.8343
No log 52.4 262 0.6643 0.6244 0.6643 0.8150
No log 52.8 264 0.6313 0.5470 0.6313 0.7946
No log 53.2 266 0.6270 0.5567 0.6270 0.7918
No log 53.6 268 0.6288 0.5955 0.6288 0.7930
No log 54.0 270 0.6265 0.5955 0.6265 0.7915
No log 54.4 272 0.6214 0.5955 0.6214 0.7883
No log 54.8 274 0.6212 0.5955 0.6212 0.7882
No log 55.2 276 0.6176 0.5955 0.6176 0.7859
No log 55.6 278 0.6036 0.5955 0.6036 0.7769
No log 56.0 280 0.5958 0.5955 0.5958 0.7719
No log 56.4 282 0.5999 0.5849 0.5999 0.7745
No log 56.8 284 0.6202 0.5921 0.6202 0.7876
No log 57.2 286 0.6545 0.5795 0.6545 0.8090
No log 57.6 288 0.6757 0.5990 0.6757 0.8220
No log 58.0 290 0.6856 0.6014 0.6856 0.8280
No log 58.4 292 0.6776 0.6014 0.6776 0.8232
No log 58.8 294 0.6615 0.6014 0.6615 0.8134
No log 59.2 296 0.6566 0.5783 0.6566 0.8103
No log 59.6 298 0.6331 0.5794 0.6331 0.7957
No log 60.0 300 0.6322 0.5794 0.6322 0.7951
No log 60.4 302 0.6311 0.5466 0.6311 0.7944
No log 60.8 304 0.6307 0.5941 0.6307 0.7942
No log 61.2 306 0.6321 0.5941 0.6321 0.7950
No log 61.6 308 0.6371 0.5941 0.6371 0.7982
No log 62.0 310 0.6299 0.5955 0.6299 0.7937
No log 62.4 312 0.6271 0.5955 0.6271 0.7919
No log 62.8 314 0.6391 0.5667 0.6391 0.7994
No log 63.2 316 0.6622 0.5688 0.6622 0.8137
No log 63.6 318 0.6763 0.5876 0.6763 0.8224
No log 64.0 320 0.6768 0.5688 0.6768 0.8227
No log 64.4 322 0.6620 0.5169 0.6620 0.8137
No log 64.8 324 0.6527 0.5955 0.6527 0.8079
No log 65.2 326 0.6449 0.5955 0.6449 0.8031
No log 65.6 328 0.6350 0.5955 0.6350 0.7969
No log 66.0 330 0.6252 0.5955 0.6252 0.7907
No log 66.4 332 0.6166 0.5567 0.6166 0.7852
No log 66.8 334 0.6195 0.5817 0.6195 0.7871
No log 67.2 336 0.6215 0.5567 0.6215 0.7883
No log 67.6 338 0.6262 0.5567 0.6262 0.7913
No log 68.0 340 0.6387 0.5567 0.6387 0.7992
No log 68.4 342 0.6553 0.5167 0.6553 0.8095
No log 68.8 344 0.6683 0.5169 0.6683 0.8175
No log 69.2 346 0.6895 0.5576 0.6895 0.8304
No log 69.6 348 0.6891 0.5576 0.6891 0.8301
No log 70.0 350 0.6710 0.5369 0.6710 0.8191
No log 70.4 352 0.6577 0.5369 0.6577 0.8110
No log 70.8 354 0.6405 0.5470 0.6405 0.8003
No log 71.2 356 0.6283 0.5859 0.6283 0.7927
No log 71.6 358 0.6273 0.5470 0.6273 0.7920
No log 72.0 360 0.6270 0.5470 0.6270 0.7918
No log 72.4 362 0.6296 0.5470 0.6296 0.7935
No log 72.8 364 0.6329 0.5470 0.6329 0.7955
No log 73.2 366 0.6257 0.5470 0.6257 0.7910
No log 73.6 368 0.6236 0.5470 0.6236 0.7897
No log 74.0 370 0.6201 0.5567 0.6201 0.7875
No log 74.4 372 0.6185 0.5567 0.6185 0.7865
No log 74.8 374 0.6214 0.5567 0.6214 0.7883
No log 75.2 376 0.6261 0.5470 0.6261 0.7913
No log 75.6 378 0.6321 0.5470 0.6321 0.7951
No log 76.0 380 0.6408 0.5470 0.6408 0.8005
No log 76.4 382 0.6575 0.5589 0.6575 0.8109
No log 76.8 384 0.6855 0.5936 0.6855 0.8280
No log 77.2 386 0.7131 0.6126 0.7131 0.8444
No log 77.6 388 0.7144 0.5864 0.7144 0.8452
No log 78.0 390 0.6959 0.5773 0.6959 0.8342
No log 78.4 392 0.6804 0.6063 0.6804 0.8249
No log 78.8 394 0.6631 0.5572 0.6631 0.8143
No log 79.2 396 0.6454 0.5572 0.6454 0.8034
No log 79.6 398 0.6362 0.5572 0.6362 0.7976
No log 80.0 400 0.6387 0.5706 0.6387 0.7992
No log 80.4 402 0.6427 0.5940 0.6427 0.8017
No log 80.8 404 0.6406 0.5706 0.6406 0.8004
No log 81.2 406 0.6366 0.5706 0.6366 0.7979
No log 81.6 408 0.6284 0.5578 0.6284 0.7927
No log 82.0 410 0.6241 0.5470 0.6241 0.7900
No log 82.4 412 0.6218 0.5955 0.6218 0.7885
No log 82.8 414 0.6239 0.5955 0.6239 0.7899
No log 83.2 416 0.6260 0.5955 0.6260 0.7912
No log 83.6 418 0.6285 0.5859 0.6285 0.7928
No log 84.0 420 0.6328 0.5859 0.6328 0.7955
No log 84.4 422 0.6395 0.5578 0.6395 0.7997
No log 84.8 424 0.6455 0.5686 0.6455 0.8035
No log 85.2 426 0.6475 0.5686 0.6475 0.8047
No log 85.6 428 0.6521 0.5686 0.6521 0.8075
No log 86.0 430 0.6542 0.5686 0.6542 0.8088
No log 86.4 432 0.6563 0.5686 0.6563 0.8101
No log 86.8 434 0.6593 0.5686 0.6593 0.8120
No log 87.2 436 0.6627 0.5678 0.6627 0.8141
No log 87.6 438 0.6638 0.5678 0.6638 0.8148
No log 88.0 440 0.6626 0.5578 0.6626 0.8140
No log 88.4 442 0.6575 0.5578 0.6575 0.8109
No log 88.8 444 0.6544 0.5578 0.6544 0.8090
No log 89.2 446 0.6569 0.5578 0.6569 0.8105
No log 89.6 448 0.6616 0.5578 0.6616 0.8134
No log 90.0 450 0.6625 0.5572 0.6625 0.8140
No log 90.4 452 0.6678 0.5678 0.6678 0.8172
No log 90.8 454 0.6739 0.5382 0.6739 0.8209
No log 91.2 456 0.6761 0.5382 0.6761 0.8223
No log 91.6 458 0.6750 0.5382 0.6750 0.8216
No log 92.0 460 0.6705 0.5382 0.6705 0.8188
No log 92.4 462 0.6634 0.5686 0.6634 0.8145
No log 92.8 464 0.6605 0.5686 0.6605 0.8127
No log 93.2 466 0.6603 0.5686 0.6603 0.8126
No log 93.6 468 0.6593 0.5686 0.6593 0.8119
No log 94.0 470 0.6565 0.5578 0.6565 0.8103
No log 94.4 472 0.6554 0.5578 0.6554 0.8096
No log 94.8 474 0.6536 0.5578 0.6536 0.8084
No log 95.2 476 0.6512 0.5578 0.6512 0.8069
No log 95.6 478 0.6486 0.5578 0.6486 0.8054
No log 96.0 480 0.6457 0.5578 0.6457 0.8036
No log 96.4 482 0.6437 0.5578 0.6437 0.8023
No log 96.8 484 0.6429 0.5470 0.6429 0.8018
No log 97.2 486 0.6438 0.5578 0.6438 0.8024
No log 97.6 488 0.6452 0.5578 0.6452 0.8032
No log 98.0 490 0.6467 0.5578 0.6467 0.8042
No log 98.4 492 0.6475 0.5578 0.6475 0.8047
No log 98.8 494 0.6485 0.5578 0.6485 0.8053
No log 99.2 496 0.6493 0.5578 0.6493 0.8058
No log 99.6 498 0.6499 0.5578 0.6499 0.8062
0.1647 100.0 500 0.6502 0.5578 0.6502 0.8064

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_run3_AugV5_k1_task5_organization

Finetuned
(4223)
this model