arabert_cross_organization_task4_fold4

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: 0.5141
• Qwk: 0.7810
• Mse: 0.5141

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	1.8597	0.1122	1.8597
No log	0.2222	4	1.3020	0.0409	1.3020
No log	0.3333	6	1.0459	0.3503	1.0459
No log	0.4444	8	0.7138	0.4906	0.7138
No log	0.5556	10	0.5900	0.6224	0.5900
No log	0.6667	12	0.5708	0.6214	0.5708
No log	0.7778	14	0.4881	0.6661	0.4881
No log	0.8889	16	0.4843	0.6880	0.4843
No log	1.0	18	0.5772	0.7185	0.5772
No log	1.1111	20	0.4344	0.7515	0.4344
No log	1.2222	22	0.4173	0.6853	0.4173
No log	1.3333	24	0.4326	0.7470	0.4326
No log	1.4444	26	0.6327	0.7329	0.6327
No log	1.5556	28	0.6486	0.7640	0.6486
No log	1.6667	30	0.4510	0.7633	0.4510
No log	1.7778	32	0.3885	0.7569	0.3885
No log	1.8889	34	0.4049	0.7722	0.4049
No log	2.0	36	0.5466	0.7900	0.5466
No log	2.1111	38	0.5445	0.7886	0.5445
No log	2.2222	40	0.4445	0.7553	0.4445
No log	2.3333	42	0.4182	0.7437	0.4182
No log	2.4444	44	0.4202	0.7536	0.4202
No log	2.5556	46	0.5364	0.7929	0.5364
No log	2.6667	48	0.6070	0.7880	0.6070
No log	2.7778	50	0.4960	0.7859	0.4960
No log	2.8889	52	0.4044	0.7719	0.4044
No log	3.0	54	0.3938	0.7606	0.3938
No log	3.1111	56	0.4669	0.7947	0.4669
No log	3.2222	58	0.5343	0.7820	0.5343
No log	3.3333	60	0.4763	0.7853	0.4763
No log	3.4444	62	0.4091	0.7835	0.4091
No log	3.5556	64	0.4119	0.7882	0.4119
No log	3.6667	66	0.4525	0.7813	0.4525
No log	3.7778	68	0.4761	0.7828	0.4761
No log	3.8889	70	0.4893	0.7931	0.4893
No log	4.0	72	0.4435	0.7862	0.4435
No log	4.1111	74	0.4754	0.7918	0.4754
No log	4.2222	76	0.5004	0.7931	0.5004
No log	4.3333	78	0.5554	0.8090	0.5554
No log	4.4444	80	0.5319	0.7947	0.5319
No log	4.5556	82	0.4459	0.7781	0.4459
No log	4.6667	84	0.4355	0.7725	0.4355
No log	4.7778	86	0.4699	0.7823	0.4699
No log	4.8889	88	0.4860	0.7900	0.4860
No log	5.0	90	0.4400	0.7892	0.4400
No log	5.1111	92	0.4221	0.7376	0.4221
No log	5.2222	94	0.4264	0.7879	0.4264
No log	5.3333	96	0.4728	0.8106	0.4728
No log	5.4444	98	0.5254	0.8043	0.5254
No log	5.5556	100	0.4876	0.8048	0.4876
No log	5.6667	102	0.4400	0.7767	0.4400
No log	5.7778	104	0.4191	0.7487	0.4191
No log	5.8889	106	0.4281	0.7643	0.4281
No log	6.0	108	0.4819	0.7888	0.4819
No log	6.1111	110	0.5487	0.8063	0.5487
No log	6.2222	112	0.6060	0.7903	0.6060
No log	6.3333	114	0.5618	0.7847	0.5618
No log	6.4444	116	0.5080	0.7689	0.5080
No log	6.5556	118	0.4883	0.7543	0.4883
No log	6.6667	120	0.4979	0.7597	0.4979
No log	6.7778	122	0.5155	0.7757	0.5155
No log	6.8889	124	0.5239	0.7883	0.5239
No log	7.0	126	0.5025	0.7973	0.5025
No log	7.1111	128	0.4784	0.7894	0.4784
No log	7.2222	130	0.4608	0.7714	0.4608
No log	7.3333	132	0.4592	0.7608	0.4592
No log	7.4444	134	0.4736	0.7898	0.4736
No log	7.5556	136	0.5099	0.7905	0.5099
No log	7.6667	138	0.5575	0.8010	0.5575
No log	7.7778	140	0.5556	0.8167	0.5556
No log	7.8889	142	0.5181	0.7957	0.5181
No log	8.0	144	0.4691	0.7885	0.4691
No log	8.1111	146	0.4424	0.7890	0.4424
No log	8.2222	148	0.4411	0.7803	0.4411
No log	8.3333	150	0.4646	0.7859	0.4646
No log	8.4444	152	0.4939	0.8070	0.4939
No log	8.5556	154	0.5186	0.8172	0.5186
No log	8.6667	156	0.5307	0.8195	0.5307
No log	8.7778	158	0.5184	0.8146	0.5184
No log	8.8889	160	0.4898	0.8099	0.4898
No log	9.0	162	0.4741	0.7957	0.4741
No log	9.1111	164	0.4724	0.7948	0.4724
No log	9.2222	166	0.4828	0.7937	0.4828
No log	9.3333	168	0.4861	0.7937	0.4861
No log	9.4444	170	0.4940	0.7848	0.4940
No log	9.5556	172	0.5042	0.7810	0.5042
No log	9.6667	174	0.5129	0.7810	0.5129
No log	9.7778	176	0.5142	0.7810	0.5142
No log	9.8889	178	0.5135	0.7810	0.5135
No log	10.0	180	0.5141	0.7810	0.5141

Framework versions

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