arabert_cross_development_task1_fold3

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: 0.5861
  • Qwk: 0.7634
  • Mse: 0.5861

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 1.7502 0.1271 1.7502
No log 0.2667 4 1.2292 0.1987 1.2292
No log 0.4 6 1.3080 0.3555 1.3080
No log 0.5333 8 0.9607 0.4976 0.9607
No log 0.6667 10 0.8578 0.6328 0.8578
No log 0.8 12 0.8243 0.6683 0.8243
No log 0.9333 14 0.6736 0.6082 0.6736
No log 1.0667 16 0.6841 0.5941 0.6841
No log 1.2 18 0.6510 0.6300 0.6510
No log 1.3333 20 0.7744 0.7345 0.7744
No log 1.4667 22 0.6786 0.7336 0.6786
No log 1.6 24 0.5769 0.6663 0.5769
No log 1.7333 26 0.5647 0.6751 0.5647
No log 1.8667 28 0.6191 0.7228 0.6191
No log 2.0 30 0.6480 0.7149 0.6480
No log 2.1333 32 0.5930 0.6377 0.5930
No log 2.2667 34 0.5792 0.6840 0.5792
No log 2.4 36 0.6399 0.7684 0.6399
No log 2.5333 38 0.6099 0.7730 0.6099
No log 2.6667 40 0.5513 0.7336 0.5513
No log 2.8 42 0.5787 0.7674 0.5787
No log 2.9333 44 0.6353 0.7926 0.6353
No log 3.0667 46 0.5670 0.7594 0.5670
No log 3.2 48 0.6004 0.7827 0.6004
No log 3.3333 50 0.6263 0.7869 0.6263
No log 3.4667 52 0.5762 0.7498 0.5762
No log 3.6 54 0.5570 0.7472 0.5570
No log 3.7333 56 0.6280 0.7790 0.6280
No log 3.8667 58 0.6056 0.7764 0.6056
No log 4.0 60 0.5239 0.7299 0.5239
No log 4.1333 62 0.5192 0.7305 0.5192
No log 4.2667 64 0.5703 0.7631 0.5703
No log 4.4 66 0.6518 0.7918 0.6518
No log 4.5333 68 0.7302 0.7883 0.7302
No log 4.6667 70 0.6654 0.7960 0.6654
No log 4.8 72 0.5714 0.7539 0.5714
No log 4.9333 74 0.5149 0.7046 0.5149
No log 5.0667 76 0.5129 0.6751 0.5129
No log 5.2 78 0.5200 0.7257 0.5200
No log 5.3333 80 0.5968 0.7701 0.5968
No log 5.4667 82 0.6356 0.7899 0.6356
No log 5.6 84 0.5976 0.7718 0.5976
No log 5.7333 86 0.5510 0.7528 0.5510
No log 5.8667 88 0.5499 0.7505 0.5499
No log 6.0 90 0.5581 0.7507 0.5581
No log 6.1333 92 0.5846 0.7624 0.5846
No log 6.2667 94 0.6247 0.7828 0.6247
No log 6.4 96 0.6363 0.7865 0.6363
No log 6.5333 98 0.6065 0.7792 0.6065
No log 6.6667 100 0.5753 0.7552 0.5753
No log 6.8 102 0.5617 0.7438 0.5617
No log 6.9333 104 0.5593 0.7415 0.5593
No log 7.0667 106 0.5501 0.7410 0.5501
No log 7.2 108 0.5736 0.7489 0.5736
No log 7.3333 110 0.6235 0.7773 0.6235
No log 7.4667 112 0.6392 0.7840 0.6392
No log 7.6 114 0.6211 0.7732 0.6211
No log 7.7333 116 0.5970 0.7733 0.5970
No log 7.8667 118 0.5611 0.7530 0.5611
No log 8.0 120 0.5439 0.7470 0.5439
No log 8.1333 122 0.5497 0.7484 0.5497
No log 8.2667 124 0.5836 0.7580 0.5836
No log 8.4 126 0.6389 0.7749 0.6389
No log 8.5333 128 0.6778 0.7817 0.6778
No log 8.6667 130 0.6794 0.7790 0.6794
No log 8.8 132 0.6575 0.7769 0.6575
No log 8.9333 134 0.6182 0.7797 0.6182
No log 9.0667 136 0.5921 0.7689 0.5921
No log 9.2 138 0.5771 0.7588 0.5771
No log 9.3333 140 0.5657 0.7530 0.5657
No log 9.4667 142 0.5672 0.7530 0.5672
No log 9.6 144 0.5759 0.7639 0.5759
No log 9.7333 146 0.5809 0.7611 0.5809
No log 9.8667 148 0.5847 0.7611 0.5847
No log 10.0 150 0.5861 0.7634 0.5861

Framework versions

  • Transformers 4.44.0
  • Pytorch 2.4.0
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
7
Safetensors
Model size
135M params
Tensor type
F32
·
Inference API
Unable to determine this model's library. Check the docs .

Model tree for salbatarni/arabert_cross_development_task1_fold3

Finetuned
(2321)
this model