arabert_cross_organization_task1_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: 1.1732
• Qwk: 0.1337
• Mse: 1.1732

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	4.6922	-0.0262	4.6922
No log	0.2353	4	1.9697	-0.0193	1.9697
No log	0.3529	6	1.2474	-0.0024	1.2474
No log	0.4706	8	1.3039	-0.1115	1.3039
No log	0.5882	10	1.2027	-0.0222	1.2027
No log	0.7059	12	1.1028	0.0290	1.1028
No log	0.8235	14	1.1603	0.0257	1.1603
No log	0.9412	16	1.1285	-0.0227	1.1285
No log	1.0588	18	1.1018	0.0835	1.1018
No log	1.1765	20	1.1034	-0.0341	1.1034
No log	1.2941	22	1.0966	0.0201	1.0966
No log	1.4118	24	1.0911	0.0407	1.0911
No log	1.5294	26	1.0941	0.0854	1.0941
No log	1.6471	28	1.1027	0.0379	1.1027
No log	1.7647	30	1.1067	0.0462	1.1067
No log	1.8824	32	1.1142	-0.0019	1.1142
No log	2.0	34	1.1210	0.0205	1.1210
No log	2.1176	36	1.1209	0.0124	1.1209
No log	2.2353	38	1.1138	0.0345	1.1138
No log	2.3529	40	1.1537	0.0350	1.1537
No log	2.4706	42	1.1387	0.0164	1.1387
No log	2.5882	44	1.1467	0.0448	1.1467
No log	2.7059	46	1.1779	0.0372	1.1779
No log	2.8235	48	1.1734	0.0435	1.1734
No log	2.9412	50	1.0971	0.0562	1.0971
No log	3.0588	52	1.0868	0.0993	1.0868
No log	3.1765	54	1.1937	0.0208	1.1937
No log	3.2941	56	1.1870	-0.0026	1.1870
No log	3.4118	58	1.1154	0.0605	1.1154
No log	3.5294	60	1.1425	0.0586	1.1425
No log	3.6471	62	1.1710	0.0755	1.1710
No log	3.7647	64	1.2418	-0.0251	1.2418
No log	3.8824	66	1.2678	-0.0157	1.2678
No log	4.0	68	1.3259	-0.0370	1.3259
No log	4.1176	70	1.4421	-0.0243	1.4421
No log	4.2353	72	1.3217	-0.0265	1.3217
No log	4.3529	74	1.1731	0.1242	1.1731
No log	4.4706	76	1.1656	0.1082	1.1656
No log	4.5882	78	1.2346	-0.0387	1.2346
No log	4.7059	80	1.3196	0.0306	1.3196
No log	4.8235	82	1.2508	-0.0093	1.2508
No log	4.9412	84	1.1905	-0.0118	1.1905
No log	5.0588	86	1.1863	0.0002	1.1863
No log	5.1765	88	1.2464	0.0063	1.2464
No log	5.2941	90	1.2248	0.0177	1.2248
No log	5.4118	92	1.2140	0.0391	1.2140
No log	5.5294	94	1.2402	0.0177	1.2402
No log	5.6471	96	1.1852	0.1192	1.1852
No log	5.7647	98	1.1973	0.0602	1.1973
No log	5.8824	100	1.2330	0.0886	1.2330
No log	6.0	102	1.1570	0.0504	1.1570
No log	6.1176	104	1.1259	0.1211	1.1259
No log	6.2353	106	1.1231	0.1211	1.1231
No log	6.3529	108	1.1216	0.0835	1.1216
No log	6.4706	110	1.1530	0.0907	1.1530
No log	6.5882	112	1.1410	0.1176	1.1410
No log	6.7059	114	1.1564	0.1120	1.1564
No log	6.8235	116	1.2102	0.0988	1.2102
No log	6.9412	118	1.2596	0.0681	1.2596
No log	7.0588	120	1.2375	0.0722	1.2375
No log	7.1765	122	1.1821	0.1425	1.1821
No log	7.2941	124	1.1856	0.1180	1.1856
No log	7.4118	126	1.1952	0.1542	1.1952
No log	7.5294	128	1.2010	0.1643	1.2010
No log	7.6471	130	1.1966	0.1542	1.1966
No log	7.7647	132	1.2024	0.1589	1.2024
No log	7.8824	134	1.2473	0.0820	1.2473
No log	8.0	136	1.2492	0.0847	1.2492
No log	8.1176	138	1.2147	0.1538	1.2147
No log	8.2353	140	1.1859	0.0946	1.1859
No log	8.3529	142	1.1767	0.1220	1.1767
No log	8.4706	144	1.1738	0.1091	1.1738
No log	8.5882	146	1.1774	0.1326	1.1774
No log	8.7059	148	1.1779	0.1337	1.1779
No log	8.8235	150	1.1724	0.1326	1.1724
No log	8.9412	152	1.1684	0.1303	1.1684
No log	9.0588	154	1.1673	0.1468	1.1673
No log	9.1765	156	1.1674	0.1468	1.1674
No log	9.2941	158	1.1677	0.1326	1.1677
No log	9.4118	160	1.1752	0.1369	1.1752
No log	9.5294	162	1.1826	0.1538	1.1826
No log	9.6471	164	1.1796	0.1229	1.1796
No log	9.7647	166	1.1767	0.1369	1.1767
No log	9.8824	168	1.1735	0.1337	1.1735
No log	10.0	170	1.1732	0.1337	1.1732

Framework versions

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