arabert_cross_organization_task3_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: 0.4384
• Qwk: 0.7923
• Mse: 0.4384

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	0.1176	2	2.0659	0.1309	2.0659
No log	0.2353	4	1.1312	0.2513	1.1312
No log	0.3529	6	1.2762	0.3076	1.2762
No log	0.4706	8	0.9742	0.3913	0.9742
No log	0.5882	10	0.8094	0.3950	0.8094
No log	0.7059	12	0.6869	0.4798	0.6869
No log	0.8235	14	0.5774	0.5954	0.5774
No log	0.9412	16	0.6003	0.7210	0.6003
No log	1.0588	18	0.5323	0.7387	0.5323
No log	1.1765	20	0.4319	0.7027	0.4319
No log	1.2941	22	0.4200	0.7041	0.4200
No log	1.4118	24	0.4896	0.7702	0.4896
No log	1.5294	26	0.4854	0.7637	0.4854
No log	1.6471	28	0.4346	0.7311	0.4346
No log	1.7647	30	0.4283	0.7225	0.4283
No log	1.8824	32	0.4469	0.6962	0.4469
No log	2.0	34	0.4554	0.7521	0.4554
No log	2.1176	36	0.6141	0.7808	0.6141
No log	2.2353	38	0.5954	0.7714	0.5954
No log	2.3529	40	0.4972	0.7836	0.4972
No log	2.4706	42	0.3941	0.7529	0.3941
No log	2.5882	44	0.3651	0.7446	0.3651
No log	2.7059	46	0.3529	0.7605	0.3529
No log	2.8235	48	0.3699	0.7870	0.3699
No log	2.9412	50	0.4844	0.8194	0.4844
No log	3.0588	52	0.5294	0.8187	0.5294
No log	3.1765	54	0.4291	0.8009	0.4291
No log	3.2941	56	0.3600	0.7917	0.3600
No log	3.4118	58	0.3588	0.7841	0.3588
No log	3.5294	60	0.4067	0.7876	0.4067
No log	3.6471	62	0.5014	0.8211	0.5014
No log	3.7647	64	0.4957	0.8087	0.4957
No log	3.8824	66	0.3997	0.7705	0.3997
No log	4.0	68	0.3670	0.7411	0.3670
No log	4.1176	70	0.3666	0.7485	0.3666
No log	4.2353	72	0.4185	0.7811	0.4185
No log	4.3529	74	0.5672	0.8114	0.5672
No log	4.4706	76	0.6146	0.7985	0.6146
No log	4.5882	78	0.5087	0.7803	0.5087
No log	4.7059	80	0.4307	0.7841	0.4307
No log	4.8235	82	0.4210	0.7338	0.4210
No log	4.9412	84	0.4161	0.7329	0.4161
No log	5.0588	86	0.4113	0.7754	0.4113
No log	5.1765	88	0.4423	0.8003	0.4423
No log	5.2941	90	0.4908	0.8117	0.4908
No log	5.4118	92	0.4559	0.8009	0.4559
No log	5.5294	94	0.4337	0.7982	0.4337
No log	5.6471	96	0.4283	0.8001	0.4283
No log	5.7647	98	0.4085	0.7829	0.4085
No log	5.8824	100	0.4060	0.7781	0.4060
No log	6.0	102	0.4389	0.7865	0.4389
No log	6.1176	104	0.4913	0.7924	0.4913
No log	6.2353	106	0.4747	0.7811	0.4747
No log	6.3529	108	0.4248	0.7908	0.4248
No log	6.4706	110	0.3799	0.7679	0.3799
No log	6.5882	112	0.3690	0.7547	0.3690
No log	6.7059	114	0.3767	0.7741	0.3767
No log	6.8235	116	0.4030	0.7842	0.4030
No log	6.9412	118	0.4559	0.8101	0.4559
No log	7.0588	120	0.4943	0.8129	0.4943
No log	7.1765	122	0.4753	0.8154	0.4753
No log	7.2941	124	0.4148	0.8032	0.4148
No log	7.4118	126	0.3867	0.7623	0.3867
No log	7.5294	128	0.3894	0.7606	0.3894
No log	7.6471	130	0.4128	0.7745	0.4128
No log	7.7647	132	0.4714	0.7941	0.4714
No log	7.8824	134	0.5145	0.8135	0.5145
No log	8.0	136	0.5184	0.8135	0.5184
No log	8.1176	138	0.4920	0.8112	0.4920
No log	8.2353	140	0.4462	0.8017	0.4462
No log	8.3529	142	0.4059	0.7727	0.4059
No log	8.4706	144	0.3943	0.7707	0.3943
No log	8.5882	146	0.3955	0.7715	0.3955
No log	8.7059	148	0.4024	0.7702	0.4024
No log	8.8235	150	0.4170	0.7875	0.4170
No log	8.9412	152	0.4406	0.7995	0.4406
No log	9.0588	154	0.4635	0.7980	0.4635
No log	9.1765	156	0.4777	0.8082	0.4777
No log	9.2941	158	0.4780	0.8082	0.4780
No log	9.4118	160	0.4663	0.7989	0.4663
No log	9.5294	162	0.4565	0.7957	0.4565
No log	9.6471	164	0.4496	0.7945	0.4496
No log	9.7647	166	0.4453	0.7965	0.4453
No log	9.8824	168	0.4404	0.7994	0.4404
No log	10.0	170	0.4384	0.7923	0.4384

Framework versions

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