arabert_cross_organization_task7_fold5

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: 0.5800
• Qwk: 0.7690
• Mse: 0.5800

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	0.125	2	3.1878	0.0032	3.1878
No log	0.25	4	1.6075	0.1098	1.6075
No log	0.375	6	1.1035	0.2670	1.1035
No log	0.5	8	1.4112	0.3086	1.4112
No log	0.625	10	1.3436	0.3880	1.3436
No log	0.75	12	1.1181	0.3654	1.1181
No log	0.875	14	0.8279	0.4164	0.8279
No log	1.0	16	0.6795	0.4918	0.6795
No log	1.125	18	0.6997	0.5370	0.6997
No log	1.25	20	0.8208	0.6120	0.8208
No log	1.375	22	0.7785	0.7348	0.7785
No log	1.5	24	0.6388	0.7210	0.6388
No log	1.625	26	0.6688	0.7365	0.6688
No log	1.75	28	0.6507	0.7403	0.6507
No log	1.875	30	0.5320	0.7356	0.5320
No log	2.0	32	0.5254	0.7454	0.5254
No log	2.125	34	0.5348	0.7325	0.5348
No log	2.25	36	0.6139	0.7376	0.6139
No log	2.375	38	0.6648	0.7474	0.6648
No log	2.5	40	0.5894	0.7707	0.5894
No log	2.625	42	0.5580	0.7530	0.5580
No log	2.75	44	0.5824	0.7698	0.5824
No log	2.875	46	0.6444	0.7641	0.6444
No log	3.0	48	0.5327	0.7206	0.5327
No log	3.125	50	0.5871	0.7517	0.5871
No log	3.25	52	0.5331	0.7366	0.5331
No log	3.375	54	0.6130	0.7665	0.6130
No log	3.5	56	0.5889	0.7650	0.5889
No log	3.625	58	0.5848	0.7758	0.5848
No log	3.75	60	0.7089	0.7737	0.7089
No log	3.875	62	0.7846	0.7865	0.7846
No log	4.0	64	0.6552	0.7793	0.6552
No log	4.125	66	0.5020	0.7284	0.5020
No log	4.25	68	0.5170	0.7322	0.5170
No log	4.375	70	0.5877	0.7481	0.5877
No log	4.5	72	0.5700	0.7494	0.5700
No log	4.625	74	0.5147	0.7380	0.5147
No log	4.75	76	0.5942	0.7664	0.5942
No log	4.875	78	0.6564	0.7710	0.6564
No log	5.0	80	0.6565	0.7710	0.6565
No log	5.125	82	0.6572	0.7802	0.6572
No log	5.25	84	0.6860	0.7836	0.6860
No log	5.375	86	0.6265	0.7687	0.6265
No log	5.5	88	0.5116	0.7530	0.5116
No log	5.625	90	0.5026	0.7603	0.5026
No log	5.75	92	0.5588	0.7542	0.5588
No log	5.875	94	0.6752	0.7902	0.6752
No log	6.0	96	0.7891	0.7984	0.7891
No log	6.125	98	0.7038	0.7947	0.7038
No log	6.25	100	0.5797	0.7519	0.5797
No log	6.375	102	0.5895	0.7634	0.5895
No log	6.5	104	0.6498	0.7782	0.6498
No log	6.625	106	0.5864	0.7623	0.5864
No log	6.75	108	0.5259	0.7227	0.5259
No log	6.875	110	0.5133	0.7040	0.5133
No log	7.0	112	0.5219	0.7120	0.5219
No log	7.125	114	0.5822	0.7464	0.5822
No log	7.25	116	0.6526	0.7676	0.6526
No log	7.375	118	0.6628	0.7818	0.6628
No log	7.5	120	0.6080	0.7726	0.6080
No log	7.625	122	0.5645	0.7416	0.5645
No log	7.75	124	0.5592	0.7409	0.5592
No log	7.875	126	0.5637	0.7527	0.5637
No log	8.0	128	0.5640	0.7522	0.5640
No log	8.125	130	0.5743	0.7522	0.5743
No log	8.25	132	0.6128	0.7551	0.6128
No log	8.375	134	0.6083	0.7551	0.6083
No log	8.5	136	0.5761	0.7661	0.5761
No log	8.625	138	0.5522	0.7596	0.5522
No log	8.75	140	0.5418	0.7580	0.5418
No log	8.875	142	0.5541	0.7626	0.5541
No log	9.0	144	0.5890	0.7591	0.5890
No log	9.125	146	0.6347	0.7704	0.6347
No log	9.25	148	0.6524	0.7643	0.6524
No log	9.375	150	0.6441	0.7643	0.6441
No log	9.5	152	0.6196	0.7597	0.6196
No log	9.625	154	0.5988	0.7575	0.5988
No log	9.75	156	0.5835	0.7591	0.5835
No log	9.875	158	0.5803	0.7672	0.5803
No log	10.0	160	0.5800	0.7690	0.5800

Framework versions

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