metadata

base_model: aubmindlab/bert-base-arabertv02
tags:
  - generated_from_trainer
model-index:
  - name: arabert_cross_relevance_task7_fold0
    results: []

arabert_cross_relevance_task7_fold0

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: 0.2152
Qwk: 0.1252
Mse: 0.2152

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: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse
No log	0.0351	2	1.7444	0.0064	1.7443
No log	0.0702	4	0.5072	0.0044	0.5075
No log	0.1053	6	0.3202	0.1573	0.3202
No log	0.1404	8	0.2932	0.1238	0.2931
No log	0.1754	10	0.2459	0.0491	0.2459
No log	0.2105	12	0.2721	0.0611	0.2721
No log	0.2456	14	0.3258	0.1020	0.3257
No log	0.2807	16	0.3460	0.1206	0.3459
No log	0.3158	18	0.2997	0.1063	0.2996
No log	0.3509	20	0.2574	0.1124	0.2574
No log	0.3860	22	0.2457	0.1501	0.2457
No log	0.4211	24	0.2347	0.1830	0.2348
No log	0.4561	26	0.2255	0.1739	0.2255
No log	0.4912	28	0.2268	0.1123	0.2267
No log	0.5263	30	0.2372	0.1054	0.2370
No log	0.5614	32	0.2505	0.1020	0.2503
No log	0.5965	34	0.2609	0.1020	0.2606
No log	0.6316	36	0.2711	0.0959	0.2709
No log	0.6667	38	0.2674	0.0679	0.2672
No log	0.7018	40	0.2605	0.0756	0.2602
No log	0.7368	42	0.2455	0.0756	0.2453
No log	0.7719	44	0.2290	0.0756	0.2289
No log	0.8070	46	0.2191	0.0791	0.2190
No log	0.8421	48	0.2160	0.0898	0.2159
No log	0.8772	50	0.2154	0.1042	0.2153
No log	0.9123	52	0.2153	0.1215	0.2152
No log	0.9474	54	0.2154	0.1252	0.2153
No log	0.9825	56	0.2152	0.1252	0.2152

Framework versions

Transformers 4.44.0
Pytorch 2.4.0
Datasets 2.21.0
Tokenizers 0.19.1