arabert_baseline_augmented_organization_task1_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: 1.0135
• Qwk: 0.6379
• Mse: 1.0135
• Rmse: 1.0067

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	3.8330	0.0810	3.8330	1.9578
No log	0.2667	4	2.2441	0.1096	2.2441	1.4980
No log	0.4	6	1.5116	0.1312	1.5116	1.2295
No log	0.5333	8	1.4466	0.0742	1.4466	1.2027
No log	0.6667	10	1.7505	-0.0269	1.7505	1.3231
No log	0.8	12	1.5458	-0.0269	1.5458	1.2433
No log	0.9333	14	1.4340	0.1004	1.4340	1.1975
No log	1.0667	16	1.4830	0.0	1.4830	1.2178
No log	1.2	18	1.4715	0.0	1.4715	1.2131
No log	1.3333	20	1.4461	0.1004	1.4461	1.2025
No log	1.4667	22	1.2989	0.2546	1.2989	1.1397
No log	1.6	24	1.2831	0.1352	1.2831	1.1327
No log	1.7333	26	1.3299	0.1803	1.3299	1.1532
No log	1.8667	28	1.2982	0.2849	1.2982	1.1394
No log	2.0	30	1.2938	0.3758	1.2938	1.1375
No log	2.1333	32	1.3870	0.4160	1.3870	1.1777
No log	2.2667	34	1.4143	0.3568	1.4143	1.1893
No log	2.4	36	1.4197	0.3625	1.4197	1.1915
No log	2.5333	38	1.4235	0.3696	1.4235	1.1931
No log	2.6667	40	1.3429	0.3261	1.3429	1.1588
No log	2.8	42	1.2068	0.4112	1.2068	1.0985
No log	2.9333	44	1.2104	0.4059	1.2104	1.1002
No log	3.0667	46	1.0886	0.4280	1.0886	1.0433
No log	3.2	48	1.0380	0.5086	1.0380	1.0188
No log	3.3333	50	1.1909	0.4362	1.1909	1.0913
No log	3.4667	52	1.1982	0.3893	1.1982	1.0946
No log	3.6	54	1.0160	0.4615	1.0160	1.0080
No log	3.7333	56	0.9465	0.5097	0.9465	0.9729
No log	3.8667	58	0.9380	0.5097	0.9380	0.9685
No log	4.0	60	0.9526	0.4421	0.9526	0.9760
No log	4.1333	62	1.1180	0.4134	1.1180	1.0573
No log	4.2667	64	1.2298	0.3893	1.2298	1.1090
No log	4.4	66	1.3905	0.3651	1.3905	1.1792
No log	4.5333	68	1.2865	0.4385	1.2865	1.1342
No log	4.6667	70	1.1647	0.4647	1.1647	1.0792
No log	4.8	72	1.0680	0.4509	1.0680	1.0334
No log	4.9333	74	1.0009	0.5084	1.0009	1.0005
No log	5.0667	76	1.0218	0.6023	1.0218	1.0109
No log	5.2	78	1.1705	0.5664	1.1705	1.0819
No log	5.3333	80	1.2873	0.5664	1.2873	1.1346
No log	5.4667	82	1.2282	0.5808	1.2282	1.1083
No log	5.6	84	1.2344	0.5808	1.2344	1.1110
No log	5.7333	86	1.1699	0.5874	1.1699	1.0816
No log	5.8667	88	1.0685	0.5844	1.0685	1.0337
No log	6.0	90	0.9766	0.6763	0.9766	0.9883
No log	6.1333	92	1.0255	0.5059	1.0255	1.0127
No log	6.2667	94	1.1403	0.5056	1.1403	1.0678
No log	6.4	96	1.1525	0.5056	1.1525	1.0735
No log	6.5333	98	1.0805	0.5056	1.0805	1.0395
No log	6.6667	100	0.9544	0.6100	0.9544	0.9769
No log	6.8	102	0.8322	0.5811	0.8322	0.9123
No log	6.9333	104	0.8310	0.5811	0.8310	0.9116
No log	7.0667	106	0.9255	0.5898	0.9255	0.9620
No log	7.2	108	1.1221	0.5495	1.1221	1.0593
No log	7.3333	110	1.1615	0.4847	1.1615	1.0777
No log	7.4667	112	1.0503	0.5495	1.0503	1.0249
No log	7.6	114	0.9256	0.5674	0.9256	0.9621
No log	7.7333	116	0.8957	0.5084	0.8957	0.9464
No log	7.8667	118	0.9132	0.5625	0.9132	0.9556
No log	8.0	120	0.9943	0.6503	0.9943	0.9971
No log	8.1333	122	1.0845	0.5276	1.0845	1.0414
No log	8.2667	124	1.1315	0.4834	1.1315	1.0637
No log	8.4	126	1.1630	0.5046	1.1630	1.0784
No log	8.5333	128	1.1916	0.5046	1.1916	1.0916
No log	8.6667	130	1.2045	0.5046	1.2045	1.0975
No log	8.8	132	1.1996	0.5046	1.1996	1.0953
No log	8.9333	134	1.1838	0.6084	1.1838	1.0880
No log	9.0667	136	1.1246	0.5254	1.1246	1.0605
No log	9.2	138	1.0645	0.6155	1.0645	1.0317
No log	9.3333	140	1.0407	0.6341	1.0407	1.0201
No log	9.4667	142	1.0151	0.6379	1.0151	1.0075
No log	9.6	144	1.0129	0.6379	1.0129	1.0064
No log	9.7333	146	1.0142	0.6379	1.0142	1.0071
No log	9.8667	148	1.0140	0.6379	1.0140	1.0070
No log	10.0	150	1.0135	0.6379	1.0135	1.0067

Framework versions

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