ArabicNewSplits6_FineTuningAraBERT_run1_AugV5_k2_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: 0.7292
• Qwk: 0.5572
• Mse: 0.7292
• Rmse: 0.8539

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	0.1538	2	3.8876	0.0094	3.8876	1.9717
No log	0.3077	4	1.9183	0.0869	1.9183	1.3850
No log	0.4615	6	1.3150	0.0528	1.3150	1.1467
No log	0.6154	8	1.0224	0.0182	1.0224	1.0111
No log	0.7692	10	0.9053	-0.0523	0.9053	0.9515
No log	0.9231	12	0.8608	0.0561	0.8608	0.9278
No log	1.0769	14	0.7632	0.1279	0.7632	0.8736
No log	1.2308	16	0.9135	0.1382	0.9135	0.9558
No log	1.3846	18	1.0538	0.1563	1.0538	1.0265
No log	1.5385	20	0.9753	0.1629	0.9753	0.9876
No log	1.6923	22	0.8401	0.2358	0.8401	0.9166
No log	1.8462	24	0.6651	0.3098	0.6651	0.8155
No log	2.0	26	0.5816	0.4697	0.5816	0.7626
No log	2.1538	28	0.5557	0.4250	0.5557	0.7455
No log	2.3077	30	0.6130	0.4013	0.6130	0.7830
No log	2.4615	32	0.9967	0.2550	0.9967	0.9983
No log	2.6154	34	1.3651	0.2543	1.3651	1.1684
No log	2.7692	36	1.3834	0.2472	1.3834	1.1762
No log	2.9231	38	0.9884	0.3095	0.9884	0.9942
No log	3.0769	40	0.6751	0.4481	0.6751	0.8216
No log	3.2308	42	0.6475	0.4907	0.6475	0.8047
No log	3.3846	44	0.8600	0.4280	0.8600	0.9274
No log	3.5385	46	1.2417	0.2925	1.2417	1.1143
No log	3.6923	48	1.2107	0.3047	1.2107	1.1003
No log	3.8462	50	0.9631	0.4425	0.9631	0.9814
No log	4.0	52	0.7222	0.4776	0.7222	0.8498
No log	4.1538	54	0.6645	0.5369	0.6645	0.8151
No log	4.3077	56	0.6726	0.5290	0.6726	0.8201
No log	4.4615	58	0.7373	0.5406	0.7373	0.8587
No log	4.6154	60	0.9866	0.4513	0.9866	0.9933
No log	4.7692	62	1.1324	0.3238	1.1324	1.0642
No log	4.9231	64	0.9961	0.4485	0.9961	0.9980
No log	5.0769	66	0.7467	0.5344	0.7467	0.8641
No log	5.2308	68	0.6462	0.5256	0.6462	0.8039
No log	5.3846	70	0.6706	0.5713	0.6706	0.8189
No log	5.5385	72	0.6811	0.5713	0.6811	0.8253
No log	5.6923	74	0.6881	0.5872	0.6881	0.8295
No log	5.8462	76	0.7306	0.5572	0.7306	0.8547
No log	6.0	78	0.7807	0.5489	0.7807	0.8836
No log	6.1538	80	0.7911	0.5420	0.7911	0.8895
No log	6.3077	82	0.7597	0.5528	0.7597	0.8716
No log	6.4615	84	0.7591	0.5636	0.7591	0.8712
No log	6.6154	86	0.7695	0.5825	0.7695	0.8772
No log	6.7692	88	0.7818	0.5688	0.7818	0.8842
No log	6.9231	90	0.7666	0.5688	0.7666	0.8756
No log	7.0769	92	0.7424	0.5527	0.7424	0.8616
No log	7.2308	94	0.7654	0.5459	0.7654	0.8748
No log	7.3846	96	0.7976	0.5562	0.7976	0.8931
No log	7.5385	98	0.7742	0.5621	0.7742	0.8799
No log	7.6923	100	0.7435	0.5499	0.7435	0.8622
No log	7.8462	102	0.7241	0.5395	0.7241	0.8509
No log	8.0	104	0.7087	0.5685	0.7087	0.8418
No log	8.1538	106	0.7238	0.5470	0.7238	0.8508
No log	8.3077	108	0.7448	0.5495	0.7448	0.8630
No log	8.4615	110	0.7529	0.5266	0.7529	0.8677
No log	8.6154	112	0.7400	0.5620	0.7400	0.8602
No log	8.7692	114	0.7252	0.5511	0.7252	0.8516
No log	8.9231	116	0.7175	0.5637	0.7175	0.8470
No log	9.0769	118	0.7259	0.5686	0.7259	0.8520
No log	9.2308	120	0.7390	0.5587	0.7390	0.8596
No log	9.3846	122	0.7425	0.5587	0.7425	0.8617
No log	9.5385	124	0.7389	0.5587	0.7389	0.8596
No log	9.6923	126	0.7339	0.5587	0.7339	0.8567
No log	9.8462	128	0.7311	0.5587	0.7311	0.8551
No log	10.0	130	0.7292	0.5572	0.7292	0.8539

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1