arabert_cross_organization_task5_fold5

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: 0.5296
  • Qwk: 0.7489
  • Mse: 0.5305

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: 64
  • eval_batch_size: 64
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Qwk Mse
No log 0.1333 2 1.4169 0.1640 1.4166
No log 0.2667 4 0.9230 0.3339 0.9238
No log 0.4 6 1.4115 0.4671 1.4126
No log 0.5333 8 1.1236 0.5621 1.1252
No log 0.6667 10 0.7828 0.6007 0.7841
No log 0.8 12 0.6989 0.6557 0.7002
No log 0.9333 14 0.8321 0.7305 0.8335
No log 1.0667 16 0.7821 0.7297 0.7834
No log 1.2 18 0.5976 0.7026 0.5987
No log 1.3333 20 0.6186 0.7347 0.6197
No log 1.4667 22 0.6820 0.7525 0.6833
No log 1.6 24 0.6199 0.7565 0.6211
No log 1.7333 26 0.5771 0.7084 0.5781
No log 1.8667 28 0.6023 0.7337 0.6032
No log 2.0 30 0.7108 0.7551 0.7119
No log 2.1333 32 0.7073 0.7504 0.7084
No log 2.2667 34 0.5846 0.7098 0.5855
No log 2.4 36 0.5420 0.7162 0.5428
No log 2.5333 38 0.5540 0.7297 0.5549
No log 2.6667 40 0.6749 0.7673 0.6762
No log 2.8 42 0.6898 0.7776 0.6910
No log 2.9333 44 0.5753 0.7569 0.5763
No log 3.0667 46 0.5244 0.7115 0.5252
No log 3.2 48 0.5081 0.6940 0.5088
No log 3.3333 50 0.5254 0.7274 0.5263
No log 3.4667 52 0.5548 0.7368 0.5556
No log 3.6 54 0.6214 0.7749 0.6224
No log 3.7333 56 0.5928 0.7642 0.5937
No log 3.8667 58 0.5335 0.7287 0.5343
No log 4.0 60 0.5042 0.6752 0.5047
No log 4.1333 62 0.5119 0.6453 0.5124
No log 4.2667 64 0.5174 0.7081 0.5180
No log 4.4 66 0.6384 0.7697 0.6394
No log 4.5333 68 0.7315 0.7872 0.7326
No log 4.6667 70 0.6496 0.7748 0.6505
No log 4.8 72 0.5301 0.7292 0.5308
No log 4.9333 74 0.4995 0.6857 0.5001
No log 5.0667 76 0.4920 0.6723 0.4926
No log 5.2 78 0.5054 0.7441 0.5061
No log 5.3333 80 0.5271 0.7576 0.5279
No log 5.4667 82 0.5542 0.7741 0.5551
No log 5.6 84 0.5519 0.7756 0.5529
No log 5.7333 86 0.5269 0.7591 0.5279
No log 5.8667 88 0.4992 0.7190 0.5001
No log 6.0 90 0.4899 0.6987 0.4907
No log 6.1333 92 0.5001 0.7250 0.5010
No log 6.2667 94 0.5278 0.7506 0.5288
No log 6.4 96 0.5719 0.7785 0.5729
No log 6.5333 98 0.5579 0.7602 0.5590
No log 6.6667 100 0.5083 0.7449 0.5092
No log 6.8 102 0.4983 0.7348 0.4992
No log 6.9333 104 0.5107 0.7421 0.5116
No log 7.0667 106 0.5437 0.7504 0.5447
No log 7.2 108 0.5707 0.7563 0.5718
No log 7.3333 110 0.5453 0.7564 0.5463
No log 7.4667 112 0.5168 0.7345 0.5177
No log 7.6 114 0.5072 0.6851 0.5080
No log 7.7333 116 0.5096 0.6839 0.5104
No log 7.8667 118 0.5172 0.7127 0.5180
No log 8.0 120 0.5438 0.7557 0.5448
No log 8.1333 122 0.5579 0.7462 0.5589
No log 8.2667 124 0.5500 0.7510 0.5510
No log 8.4 126 0.5347 0.7477 0.5356
No log 8.5333 128 0.5325 0.7477 0.5334
No log 8.6667 130 0.5374 0.7495 0.5383
No log 8.8 132 0.5355 0.7495 0.5365
No log 8.9333 134 0.5332 0.7477 0.5341
No log 9.0667 136 0.5306 0.7422 0.5315
No log 9.2 138 0.5286 0.7452 0.5295
No log 9.3333 140 0.5283 0.7489 0.5292
No log 9.4667 142 0.5296 0.7489 0.5305
No log 9.6 144 0.5299 0.7489 0.5308
No log 9.7333 146 0.5289 0.7360 0.5299
No log 9.8667 148 0.5295 0.7489 0.5304
No log 10.0 150 0.5296 0.7489 0.5305

Framework versions

  • Transformers 4.44.0
  • Pytorch 2.4.0
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
2
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 HF Inference API was unable to determine this model's library.

Model tree for salbatarni/arabert_cross_organization_task5_fold5

Finetuned
(4222)
this model