arabert_cross_relevance_task3_fold2

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: 0.3250
• Qwk: -0.0612
• Mse: 0.3250

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: 64
• eval_batch_size: 64
• 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
No log	0.1111	2	0.3649	0.0417	0.3649
No log	0.2222	4	0.3523	0.0574	0.3523
No log	0.3333	6	0.3294	-0.0145	0.3294
No log	0.4444	8	0.3150	0.0	0.3150
No log	0.5556	10	0.3425	-0.0766	0.3425
No log	0.6667	12	0.5181	-0.0897	0.5181
No log	0.7778	14	0.5767	-0.0417	0.5767
No log	0.8889	16	0.4852	0.0584	0.4852
No log	1.0	18	0.4593	0.0244	0.4593
No log	1.1111	20	0.4073	-0.0106	0.4073
No log	1.2222	22	0.3911	0.1832	0.3911
No log	1.3333	24	0.5107	0.0132	0.5107
No log	1.4444	26	0.5836	0.0068	0.5836
No log	1.5556	28	0.4981	0.0584	0.4981
No log	1.6667	30	0.4625	-0.0802	0.4625
No log	1.7778	32	0.4676	-0.0714	0.4676
No log	1.8889	34	0.4572	-0.1310	0.4572
No log	2.0	36	0.4709	-0.0802	0.4709
No log	2.1111	38	0.4247	-0.1559	0.4247
No log	2.2222	40	0.4232	-0.1413	0.4232
No log	2.3333	42	0.3999	-0.1667	0.3999
No log	2.4444	44	0.4332	-0.1813	0.4332
No log	2.5556	46	0.3931	-0.1574	0.3931
No log	2.6667	48	0.3256	-0.1331	0.3256
No log	2.7778	50	0.3008	-0.0235	0.3008
No log	2.8889	52	0.3005	-0.0473	0.3005
No log	3.0	54	0.3003	-0.0714	0.3003
No log	3.1111	56	0.3285	-0.0507	0.3285
No log	3.2222	58	0.3502	-0.1508	0.3502
No log	3.3333	60	0.3399	-0.0556	0.3399
No log	3.4444	62	0.3233	-0.0563	0.3233
No log	3.5556	64	0.3562	-0.0833	0.3562
No log	3.6667	66	0.3846	-0.0981	0.3846
No log	3.7778	68	0.3873	-0.2037	0.3873
No log	3.8889	70	0.3435	-0.0870	0.3435
No log	4.0	72	0.3453	-0.0971	0.3453
No log	4.1111	74	0.3477	-0.1594	0.3477
No log	4.2222	76	0.3299	-0.0563	0.3299
No log	4.3333	78	0.3344	-0.0507	0.3344
No log	4.4444	80	0.3511	-0.2097	0.3511
No log	4.5556	82	0.3423	-0.1154	0.3423
No log	4.6667	84	0.3303	-0.0971	0.3303
No log	4.7778	86	0.3029	-0.0473	0.3029
No log	4.8889	88	0.2956	-0.0235	0.2956
No log	5.0	90	0.2958	-0.0235	0.2958
No log	5.1111	92	0.2967	-0.0235	0.2967
No log	5.2222	94	0.3097	-0.0616	0.3097
No log	5.3333	96	0.3615	-0.0366	0.3615
No log	5.4444	98	0.4248	-0.1224	0.4248
No log	5.5556	100	0.4152	-0.1000	0.4152
No log	5.6667	102	0.3541	0.0200	0.3541
No log	5.7778	104	0.3081	-0.0764	0.3081
No log	5.8889	106	0.2999	-0.0473	0.2999
No log	6.0	108	0.3089	-0.0616	0.3089
No log	6.1111	110	0.3305	-0.1620	0.3305
No log	6.2222	112	0.3428	-0.1397	0.3428
No log	6.3333	114	0.3362	-0.1496	0.3362
No log	6.4444	116	0.3379	-0.1194	0.3379
No log	6.5556	118	0.3505	-0.0827	0.3505
No log	6.6667	120	0.3567	-0.0827	0.3567
No log	6.7778	122	0.3610	-0.0772	0.3610
No log	6.8889	124	0.3491	-0.0769	0.3491
No log	7.0	126	0.3359	-0.0448	0.3359
No log	7.1111	128	0.3213	-0.0714	0.3213
No log	7.2222	130	0.3241	-0.0714	0.3241
No log	7.3333	132	0.3400	-0.0985	0.3400
No log	7.4444	134	0.3581	-0.0537	0.3581
No log	7.5556	136	0.3622	-0.0833	0.3622
No log	7.6667	138	0.3448	-0.0659	0.3448
No log	7.7778	140	0.3219	-0.0764	0.3219
No log	7.8889	142	0.3076	-0.0616	0.3076
No log	8.0	144	0.3064	-0.0616	0.3064
No log	8.1111	146	0.3132	-0.0616	0.3132
No log	8.2222	148	0.3268	-0.0612	0.3268
No log	8.3333	150	0.3441	-0.1260	0.3441
No log	8.4444	152	0.3508	-0.1328	0.3508
No log	8.5556	154	0.3460	-0.1154	0.3460
No log	8.6667	156	0.3385	-0.0606	0.3385
No log	8.7778	158	0.3312	-0.0294	0.3312
No log	8.8889	160	0.3268	-0.0294	0.3268
No log	9.0	162	0.3251	-0.0507	0.3251
No log	9.1111	164	0.3282	-0.0294	0.3282
No log	9.2222	166	0.3331	-0.1090	0.3331
No log	9.3333	168	0.3347	-0.1090	0.3347
No log	9.4444	170	0.3357	-0.1090	0.3357
No log	9.5556	172	0.3343	-0.1090	0.3343
No log	9.6667	174	0.3318	-0.0556	0.3318
No log	9.7778	176	0.3286	-0.0662	0.3286
No log	9.8889	178	0.3261	-0.0766	0.3261
No log	10.0	180	0.3250	-0.0612	0.3250

Framework versions

• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1