ArabicNewSplits5_FineTuningAraBERT_run1_AugV5_k2_task3_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: 0.7863
• Qwk: 0.2653
• Mse: 0.7863
• Rmse: 0.8867

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	3.3264	-0.0101	3.3264	1.8238
No log	0.2667	4	1.8470	-0.0667	1.8470	1.3591
No log	0.4	6	1.1826	0.0530	1.1826	1.0875
No log	0.5333	8	1.3627	0.0565	1.3627	1.1674
No log	0.6667	10	1.0252	0.1525	1.0252	1.0125
No log	0.8	12	1.0939	0.0551	1.0939	1.0459
No log	0.9333	14	2.0592	0.0536	2.0592	1.4350
No log	1.0667	16	2.7516	0.0817	2.7516	1.6588
No log	1.2	18	1.5536	0.0588	1.5536	1.2464
No log	1.3333	20	0.6419	0.2787	0.6419	0.8012
No log	1.4667	22	0.6000	0.3580	0.6000	0.7746
No log	1.6	24	0.6967	0.2877	0.6967	0.8347
No log	1.7333	26	0.8717	0.1059	0.8717	0.9337
No log	1.8667	28	0.8785	0.0698	0.8785	0.9373
No log	2.0	30	0.6959	0.2593	0.6959	0.8342
No log	2.1333	32	0.5691	0.2308	0.5691	0.7544
No log	2.2667	34	0.5681	0.2201	0.5681	0.7537
No log	2.4	36	0.5382	0.0725	0.5382	0.7336
No log	2.5333	38	0.6467	0.1529	0.6467	0.8042
No log	2.6667	40	1.2222	0.1126	1.2222	1.1055
No log	2.8	42	1.1674	0.1254	1.1674	1.0804
No log	2.9333	44	0.6365	0.2821	0.6365	0.7978
No log	3.0667	46	0.5641	0.3191	0.5641	0.7510
No log	3.2	48	0.4995	0.3735	0.4995	0.7068
No log	3.3333	50	0.5245	0.1565	0.5245	0.7243
No log	3.4667	52	0.5147	0.2000	0.5147	0.7174
No log	3.6	54	0.5089	0.2883	0.5089	0.7133
No log	3.7333	56	0.6969	0.2233	0.6969	0.8348
No log	3.8667	58	0.6016	0.2865	0.6016	0.7756
No log	4.0	60	0.5477	0.2683	0.5477	0.7401
No log	4.1333	62	0.5654	0.3684	0.5654	0.7519
No log	4.2667	64	0.6175	0.3661	0.6175	0.7858
No log	4.4	66	0.5953	0.3575	0.5953	0.7716
No log	4.5333	68	0.6256	0.3585	0.6256	0.7910
No log	4.6667	70	0.6290	0.3645	0.6290	0.7931
No log	4.8	72	0.7003	0.2294	0.7003	0.8369
No log	4.9333	74	0.7415	0.2263	0.7415	0.8611
No log	5.0667	76	0.8925	0.2353	0.8925	0.9447
No log	5.2	78	0.7941	0.2615	0.7941	0.8911
No log	5.3333	80	0.6291	0.3939	0.6291	0.7932
No log	5.4667	82	0.7123	0.3277	0.7123	0.8440
No log	5.6	84	0.6079	0.4537	0.6079	0.7797
No log	5.7333	86	0.7662	0.2868	0.7662	0.8753
No log	5.8667	88	1.0245	0.2464	1.0245	1.0122
No log	6.0	90	1.2498	0.1688	1.2498	1.1179
No log	6.1333	92	1.2267	0.1693	1.2267	1.1076
No log	6.2667	94	0.8569	0.2360	0.8569	0.9257
No log	6.4	96	0.5820	0.4502	0.5820	0.7629
No log	6.5333	98	0.6694	0.3091	0.6694	0.8181
No log	6.6667	100	0.6090	0.3803	0.6090	0.7804
No log	6.8	102	0.6119	0.4171	0.6119	0.7823
No log	6.9333	104	0.7937	0.2066	0.7937	0.8909
No log	7.0667	106	1.0088	0.2121	1.0088	1.0044
No log	7.2	108	0.9280	0.2366	0.9280	0.9633
No log	7.3333	110	0.6861	0.3363	0.6861	0.8283
No log	7.4667	112	0.6122	0.4123	0.6122	0.7824
No log	7.6	114	0.6407	0.3917	0.6407	0.8004
No log	7.7333	116	0.7565	0.2432	0.7565	0.8698
No log	7.8667	118	0.9915	0.2360	0.9915	0.9957
No log	8.0	120	1.1045	0.2353	1.1045	1.0509
No log	8.1333	122	1.0223	0.2360	1.0223	1.0111
No log	8.2667	124	0.8355	0.2389	0.8355	0.9141
No log	8.4	126	0.6540	0.3874	0.6540	0.8087
No log	8.5333	128	0.6095	0.4341	0.6095	0.7807
No log	8.6667	130	0.6175	0.4286	0.6175	0.7858
No log	8.8	132	0.6665	0.3537	0.6665	0.8164
No log	8.9333	134	0.7562	0.2903	0.7562	0.8696
No log	9.0667	136	0.8837	0.2061	0.8837	0.9400
No log	9.2	138	0.9436	0.2360	0.9436	0.9714
No log	9.3333	140	0.9508	0.2360	0.9508	0.9751
No log	9.4667	142	0.9037	0.2061	0.9037	0.9506
No log	9.6	144	0.8631	0.2062	0.8631	0.9290
No log	9.7333	146	0.8206	0.1746	0.8206	0.9059
No log	9.8667	148	0.7945	0.2066	0.7945	0.8914
No log	10.0	150	0.7863	0.2653	0.7863	0.8867

Framework versions

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