ArabicNewSplits4_FineTuningAraBERT_run2_AugV5_k4_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: 0.9401
• Qwk: 0.6521
• Mse: 0.9401
• Rmse: 0.9696

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	0.1333	2	2.3311	0.0579	2.3311	1.5268
No log	0.2667	4	1.5019	0.2059	1.5019	1.2255
No log	0.4	6	1.3696	0.1826	1.3696	1.1703
No log	0.5333	8	1.5564	0.1714	1.5564	1.2476
No log	0.6667	10	1.5417	0.2096	1.5417	1.2416
No log	0.8	12	1.4792	0.2604	1.4792	1.2162
No log	0.9333	14	1.4136	0.2081	1.4136	1.1890
No log	1.0667	16	1.3462	0.1423	1.3462	1.1603
No log	1.2	18	1.3331	0.1903	1.3331	1.1546
No log	1.3333	20	1.3817	0.2275	1.3817	1.1754
No log	1.4667	22	1.4655	0.3615	1.4655	1.2106
No log	1.6	24	1.5189	0.3872	1.5189	1.2324
No log	1.7333	26	1.5853	0.4058	1.5853	1.2591
No log	1.8667	28	1.5917	0.3974	1.5917	1.2616
No log	2.0	30	1.5476	0.4417	1.5476	1.2440
No log	2.1333	32	1.4383	0.4161	1.4383	1.1993
No log	2.2667	34	1.3609	0.4428	1.3609	1.1666
No log	2.4	36	1.2334	0.4485	1.2334	1.1106
No log	2.5333	38	1.2355	0.4853	1.2355	1.1115
No log	2.6667	40	1.3823	0.4190	1.3823	1.1757
No log	2.8	42	1.4818	0.3736	1.4818	1.2173
No log	2.9333	44	1.4005	0.4513	1.4005	1.1834
No log	3.0667	46	1.1998	0.5360	1.1998	1.0954
No log	3.2	48	1.1322	0.5410	1.1322	1.0641
No log	3.3333	50	1.1524	0.5592	1.1524	1.0735
No log	3.4667	52	1.1606	0.5122	1.1606	1.0773
No log	3.6	54	1.1610	0.5108	1.1610	1.0775
No log	3.7333	56	1.0510	0.5511	1.0510	1.0252
No log	3.8667	58	0.8471	0.6300	0.8471	0.9204
No log	4.0	60	0.7363	0.6350	0.7363	0.8581
No log	4.1333	62	0.7690	0.6118	0.7690	0.8769
No log	4.2667	64	0.8828	0.6321	0.8828	0.9396
No log	4.4	66	1.2311	0.5195	1.2311	1.1096
No log	4.5333	68	1.4559	0.4910	1.4559	1.2066
No log	4.6667	70	1.4507	0.4735	1.4507	1.2045
No log	4.8	72	1.3648	0.5001	1.3648	1.1682
No log	4.9333	74	1.1689	0.5402	1.1689	1.0811
No log	5.0667	76	1.0435	0.5907	1.0435	1.0215
No log	5.2	78	1.0005	0.5873	1.0005	1.0003
No log	5.3333	80	1.0177	0.5771	1.0177	1.0088
No log	5.4667	82	1.0492	0.5873	1.0492	1.0243
No log	5.6	84	1.1480	0.5588	1.1480	1.0715
No log	5.7333	86	1.1695	0.5376	1.1695	1.0814
No log	5.8667	88	1.1411	0.5682	1.1411	1.0682
No log	6.0	90	1.0997	0.5947	1.0997	1.0487
No log	6.1333	92	1.0337	0.5924	1.0337	1.0167
No log	6.2667	94	0.9754	0.6077	0.9754	0.9876
No log	6.4	96	0.9714	0.6145	0.9714	0.9856
No log	6.5333	98	0.9822	0.6318	0.9822	0.9911
No log	6.6667	100	0.9447	0.6487	0.9447	0.9720
No log	6.8	102	0.9478	0.6510	0.9478	0.9736
No log	6.9333	104	0.9651	0.6353	0.9651	0.9824
No log	7.0667	106	0.9520	0.6444	0.9520	0.9757
No log	7.2	108	0.9445	0.6444	0.9445	0.9719
No log	7.3333	110	0.9327	0.6536	0.9327	0.9657
No log	7.4667	112	0.9237	0.6613	0.9237	0.9611
No log	7.6	114	0.9268	0.6690	0.9268	0.9627
No log	7.7333	116	0.9361	0.6510	0.9361	0.9675
No log	7.8667	118	0.9536	0.6188	0.9536	0.9765
No log	8.0	120	0.9724	0.6073	0.9724	0.9861
No log	8.1333	122	0.9930	0.6011	0.9930	0.9965
No log	8.2667	124	0.9732	0.6011	0.9732	0.9865
No log	8.4	126	0.9619	0.6282	0.9619	0.9807
No log	8.5333	128	0.9575	0.6282	0.9575	0.9785
No log	8.6667	130	0.9402	0.6536	0.9402	0.9696
No log	8.8	132	0.9256	0.6579	0.9256	0.9621
No log	8.9333	134	0.9291	0.6639	0.9291	0.9639
No log	9.0667	136	0.9314	0.6521	0.9314	0.9651
No log	9.2	138	0.9339	0.6521	0.9339	0.9664
No log	9.3333	140	0.9333	0.6521	0.9333	0.9661
No log	9.4667	142	0.9295	0.6521	0.9295	0.9641
No log	9.6	144	0.9336	0.6521	0.9336	0.9662
No log	9.7333	146	0.9347	0.6521	0.9347	0.9668
No log	9.8667	148	0.9374	0.6521	0.9374	0.9682
No log	10.0	150	0.9401	0.6521	0.9401	0.9696

Framework versions

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