arabert_cross_organization_task5_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: 0.4286
• Qwk: 0.7401
• Mse: 0.4286

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	1.9126	0.0900	1.9126
No log	0.25	4	1.3553	0.0025	1.3553
No log	0.375	6	1.0461	0.3696	1.0461
No log	0.5	8	0.7260	0.5218	0.7260
No log	0.625	10	0.6063	0.5755	0.6063
No log	0.75	12	0.6127	0.7262	0.6127
No log	0.875	14	0.6736	0.7318	0.6736
No log	1.0	16	0.5956	0.6135	0.5956
No log	1.125	18	0.5532	0.5825	0.5532
No log	1.25	20	0.5116	0.5953	0.5116
No log	1.375	22	0.5776	0.7239	0.5776
No log	1.5	24	0.5445	0.7456	0.5445
No log	1.625	26	0.5385	0.7516	0.5385
No log	1.75	28	0.4556	0.7391	0.4556
No log	1.875	30	0.4530	0.7546	0.4530
No log	2.0	32	0.4987	0.7575	0.4987
No log	2.125	34	0.4840	0.7553	0.4840
No log	2.25	36	0.4749	0.7522	0.4749
No log	2.375	38	0.3991	0.7577	0.3991
No log	2.5	40	0.3891	0.7567	0.3891
No log	2.625	42	0.5607	0.7796	0.5607
No log	2.75	44	0.6672	0.7717	0.6672
No log	2.875	46	0.4754	0.7796	0.4754
No log	3.0	48	0.4533	0.6964	0.4533
No log	3.125	50	0.4390	0.7211	0.4390
No log	3.25	52	0.4645	0.7813	0.4645
No log	3.375	54	0.5369	0.7841	0.5369
No log	3.5	56	0.4539	0.7751	0.4539
No log	3.625	58	0.3966	0.7289	0.3966
No log	3.75	60	0.3996	0.7120	0.3996
No log	3.875	62	0.4091	0.7331	0.4091
No log	4.0	64	0.4476	0.7566	0.4476
No log	4.125	66	0.5300	0.7780	0.5300
No log	4.25	68	0.5160	0.7858	0.5160
No log	4.375	70	0.4266	0.7744	0.4266
No log	4.5	72	0.3914	0.7537	0.3914
No log	4.625	74	0.4220	0.7523	0.4220
No log	4.75	76	0.4624	0.7629	0.4624
No log	4.875	78	0.4224	0.7701	0.4224
No log	5.0	80	0.3791	0.7493	0.3791
No log	5.125	82	0.3895	0.7222	0.3895
No log	5.25	84	0.3859	0.7449	0.3859
No log	5.375	86	0.4706	0.7869	0.4706
No log	5.5	88	0.5223	0.7814	0.5223
No log	5.625	90	0.4664	0.7791	0.4664
No log	5.75	92	0.4027	0.7311	0.4027
No log	5.875	94	0.4225	0.6937	0.4225
No log	6.0	96	0.4142	0.6982	0.4142
No log	6.125	98	0.4140	0.7389	0.4140
No log	6.25	100	0.4989	0.7642	0.4989
No log	6.375	102	0.6033	0.7735	0.6033
No log	6.5	104	0.5857	0.7770	0.5857
No log	6.625	106	0.4835	0.7720	0.4835
No log	6.75	108	0.4292	0.7392	0.4292
No log	6.875	110	0.4236	0.7113	0.4236
No log	7.0	112	0.4315	0.7304	0.4315
No log	7.125	114	0.4781	0.7663	0.4781
No log	7.25	116	0.5131	0.7782	0.5131
No log	7.375	118	0.4884	0.7743	0.4884
No log	7.5	120	0.4351	0.7513	0.4351
No log	7.625	122	0.4135	0.7412	0.4135
No log	7.75	124	0.4144	0.7379	0.4144
No log	7.875	126	0.4260	0.7560	0.4260
No log	8.0	128	0.4698	0.7681	0.4698
No log	8.125	130	0.5081	0.8027	0.5081
No log	8.25	132	0.5012	0.8033	0.5012
No log	8.375	134	0.4563	0.7630	0.4563
No log	8.5	136	0.4233	0.7367	0.4233
No log	8.625	138	0.4146	0.7168	0.4146
No log	8.75	140	0.4102	0.7219	0.4102
No log	8.875	142	0.4038	0.7250	0.4038
No log	9.0	144	0.4045	0.7339	0.4045
No log	9.125	146	0.4174	0.7425	0.4174
No log	9.25	148	0.4352	0.7626	0.4352
No log	9.375	150	0.4412	0.7624	0.4412
No log	9.5	152	0.4385	0.7588	0.4385
No log	9.625	154	0.4340	0.7458	0.4340
No log	9.75	156	0.4311	0.7401	0.4311
No log	9.875	158	0.4286	0.7401	0.4286
No log	10.0	160	0.4286	0.7401	0.4286

Framework versions

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