ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run1_AugV5_k2_task1_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: 0.6117
• Qwk: 0.7456
• Mse: 0.6117
• Rmse: 0.7821

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	5.4527	-0.0139	5.4527	2.3351
No log	0.2667	4	3.2715	0.0501	3.2715	1.8087
No log	0.4	6	1.8490	0.1298	1.8490	1.3598
No log	0.5333	8	1.2821	0.2568	1.2821	1.1323
No log	0.6667	10	1.2051	0.2012	1.2051	1.0978
No log	0.8	12	1.1555	0.1975	1.1555	1.0749
No log	0.9333	14	0.9413	0.3173	0.9413	0.9702
No log	1.0667	16	1.5950	0.1925	1.5950	1.2629
No log	1.2	18	2.2138	0.1657	2.2138	1.4879
No log	1.3333	20	1.6966	0.2259	1.6966	1.3025
No log	1.4667	22	1.0577	0.3418	1.0577	1.0284
No log	1.6	24	0.9208	0.3673	0.9208	0.9596
No log	1.7333	26	1.4168	0.1032	1.4168	1.1903
No log	1.8667	28	1.7921	-0.2887	1.7921	1.3387
No log	2.0	30	1.8481	-0.1721	1.8481	1.3595
No log	2.1333	32	1.5824	-0.2097	1.5824	1.2579
No log	2.2667	34	1.3191	0.1688	1.3191	1.1485
No log	2.4	36	1.1859	0.2394	1.1859	1.0890
No log	2.5333	38	1.1241	0.2184	1.1241	1.0602
No log	2.6667	40	1.0754	0.2768	1.0754	1.0370
No log	2.8	42	1.0082	0.3197	1.0082	1.0041
No log	2.9333	44	0.9232	0.4152	0.9232	0.9608
No log	3.0667	46	0.8529	0.5306	0.8529	0.9235
No log	3.2	48	0.7879	0.5387	0.7879	0.8876
No log	3.3333	50	0.8242	0.5068	0.8242	0.9079
No log	3.4667	52	1.1057	0.4080	1.1057	1.0515
No log	3.6	54	1.2133	0.4081	1.2133	1.1015
No log	3.7333	56	1.1090	0.4306	1.1090	1.0531
No log	3.8667	58	0.7512	0.6279	0.7512	0.8667
No log	4.0	60	0.6585	0.6620	0.6585	0.8115
No log	4.1333	62	0.6849	0.6268	0.6849	0.8276
No log	4.2667	64	0.6080	0.6957	0.6080	0.7797
No log	4.4	66	0.6475	0.6885	0.6475	0.8047
No log	4.5333	68	0.8333	0.6393	0.8333	0.9129
No log	4.6667	70	0.8627	0.6441	0.8627	0.9288
No log	4.8	72	0.8206	0.6385	0.8206	0.9059
No log	4.9333	74	0.7169	0.6793	0.7169	0.8467
No log	5.0667	76	0.6083	0.7066	0.6083	0.7799
No log	5.2	78	0.5742	0.6991	0.5742	0.7577
No log	5.3333	80	0.5829	0.7007	0.5829	0.7635
No log	5.4667	82	0.6245	0.7205	0.6245	0.7903
No log	5.6	84	0.6558	0.6985	0.6558	0.8098
No log	5.7333	86	0.6811	0.6818	0.6811	0.8253
No log	5.8667	88	0.7301	0.6774	0.7301	0.8544
No log	6.0	90	0.8089	0.6589	0.8089	0.8994
No log	6.1333	92	0.9372	0.6263	0.9372	0.9681
No log	6.2667	94	0.9184	0.6386	0.9184	0.9583
No log	6.4	96	0.7856	0.6752	0.7856	0.8864
No log	6.5333	98	0.7259	0.6855	0.7259	0.8520
No log	6.6667	100	0.6704	0.7280	0.6704	0.8188
No log	6.8	102	0.6351	0.7285	0.6351	0.7969
No log	6.9333	104	0.6206	0.7373	0.6206	0.7878
No log	7.0667	106	0.6260	0.7458	0.6260	0.7912
No log	7.2	108	0.6522	0.7135	0.6522	0.8076
No log	7.3333	110	0.6539	0.7135	0.6539	0.8086
No log	7.4667	112	0.6824	0.7101	0.6824	0.8261
No log	7.6	114	0.7159	0.6963	0.7159	0.8461
No log	7.7333	116	0.7115	0.6963	0.7115	0.8435
No log	7.8667	118	0.6866	0.6918	0.6866	0.8286
No log	8.0	120	0.6556	0.7145	0.6556	0.8097
No log	8.1333	122	0.6334	0.7348	0.6334	0.7959
No log	8.2667	124	0.6155	0.7426	0.6155	0.7845
No log	8.4	126	0.6157	0.7427	0.6157	0.7847
No log	8.5333	128	0.6156	0.7441	0.6156	0.7846
No log	8.6667	130	0.6172	0.7477	0.6172	0.7856
No log	8.8	132	0.6114	0.7485	0.6114	0.7819
No log	8.9333	134	0.6069	0.7456	0.6069	0.7790
No log	9.0667	136	0.6066	0.7456	0.6066	0.7789
No log	9.2	138	0.6093	0.7413	0.6093	0.7806
No log	9.3333	140	0.6111	0.7471	0.6111	0.7817
No log	9.4667	142	0.6116	0.7413	0.6116	0.7820
No log	9.6	144	0.6116	0.7413	0.6116	0.7820
No log	9.7333	146	0.6120	0.7413	0.6120	0.7823
No log	9.8667	148	0.6119	0.7413	0.6119	0.7822
No log	10.0	150	0.6117	0.7456	0.6117	0.7821

Framework versions

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