arabert_cross_organization_task5_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: 1.1250
• Qwk: 0.0300
• Mse: 1.1225

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	4.8650	-0.0023	4.8621
No log	0.25	4	2.3813	0.0030	2.3764
No log	0.375	6	0.8965	0.0095	0.8941
No log	0.5	8	0.8278	0.0983	0.8251
No log	0.625	10	0.8457	0.0276	0.8417
No log	0.75	12	0.9196	0.0618	0.9166
No log	0.875	14	1.0317	-0.0899	1.0303
No log	1.0	16	1.1713	0.0	1.1691
No log	1.125	18	1.2251	0.0	1.2227
No log	1.25	20	1.3511	0.0	1.3490
No log	1.375	22	1.2633	-0.0232	1.2615
No log	1.5	24	1.3009	0.0	1.2985
No log	1.625	26	1.1982	0.0182	1.1952
No log	1.75	28	1.0708	0.0333	1.0677
No log	1.875	30	1.0776	0.0360	1.0747
No log	2.0	32	1.0691	0.0333	1.0663
No log	2.125	34	1.0114	0.0253	1.0084
No log	2.25	36	1.0232	0.0710	1.0205
No log	2.375	38	1.0445	0.0155	1.0419
No log	2.5	40	1.0455	0.0680	1.0427
No log	2.625	42	1.1297	0.0680	1.1268
No log	2.75	44	1.2353	0.0333	1.2324
No log	2.875	46	1.0593	0.0176	1.0562
No log	3.0	48	1.0440	0.0424	1.0409
No log	3.125	50	1.1623	0.0360	1.1594
No log	3.25	52	1.2841	0.0	1.2814
No log	3.375	54	1.1945	0.0	1.1918
No log	3.5	56	0.9910	0.0680	0.9881
No log	3.625	58	1.0024	0.0397	0.9994
No log	3.75	60	1.2429	0.0360	1.2400
No log	3.875	62	1.4419	0.0226	1.4390
No log	4.0	64	1.4478	0.0750	1.4448
No log	4.125	66	1.2244	0.0740	1.2215
No log	4.25	68	1.1031	0.0416	1.1002
No log	4.375	70	1.1904	0.0710	1.1876
No log	4.5	72	1.3453	0.0182	1.3428
No log	4.625	74	1.3082	0.0	1.3057
No log	4.75	76	1.1660	0.0565	1.1633
No log	4.875	78	1.0042	0.0424	1.0013
No log	5.0	80	0.9419	0.1245	0.9389
No log	5.125	82	1.0260	0.0680	1.0232
No log	5.25	84	1.1655	0.0388	1.1629
No log	5.375	86	1.2867	0.0	1.2843
No log	5.5	88	1.3234	0.0025	1.3209
No log	5.625	90	1.2955	0.0182	1.2928
No log	5.75	92	1.1588	0.0536	1.1559
No log	5.875	94	1.1432	0.0536	1.1403
No log	6.0	96	1.1569	0.0536	1.1542
No log	6.125	98	1.0621	0.0536	1.0594
No log	6.25	100	1.0340	0.1214	1.0312
No log	6.375	102	1.0071	0.0789	1.0043
No log	6.5	104	1.0650	0.0849	1.0622
No log	6.625	106	1.1044	0.0536	1.1017
No log	6.75	108	1.0963	0.0536	1.0936
No log	6.875	110	1.0911	0.0279	1.0883
No log	7.0	112	1.0569	0.0789	1.0542
No log	7.125	114	1.0545	0.0789	1.0517
No log	7.25	116	1.0799	0.1081	1.0772
No log	7.375	118	1.1261	0.0536	1.1234
No log	7.5	120	1.1714	0.0536	1.1688
No log	7.625	122	1.1479	0.0508	1.1451
No log	7.75	124	1.0979	0.0424	1.0951
No log	7.875	126	1.0280	0.0789	1.0252
No log	8.0	128	1.0373	0.0622	1.0347
No log	8.125	130	1.0287	0.0650	1.0261
No log	8.25	132	1.0350	0.0650	1.0325
No log	8.375	134	1.0478	0.0650	1.0452
No log	8.5	136	1.0550	0.0650	1.0524
No log	8.625	138	1.0875	0.0880	1.0848
No log	8.75	140	1.1279	0.0279	1.1252
No log	8.875	142	1.1587	0.0326	1.1561
No log	9.0	144	1.1762	0.0326	1.1736
No log	9.125	146	1.1639	0.0326	1.1613
No log	9.25	148	1.1345	0.0300	1.1319
No log	9.375	150	1.1260	0.0300	1.1233
No log	9.5	152	1.1247	0.0300	1.1221
No log	9.625	154	1.1251	0.0300	1.1225
No log	9.75	156	1.1248	0.0300	1.1223
No log	9.875	158	1.1243	0.0300	1.1218
No log	10.0	160	1.1250	0.0300	1.1225

Framework versions

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