arabert_cross_organization_task1_fold3

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: 0.5415
• Qwk: 0.7472
• Mse: 0.5415

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	0.1333	2	1.8315	0.1530	1.8315
No log	0.2667	4	1.4232	0.1303	1.4232
No log	0.4	6	1.3057	0.4221	1.3057
No log	0.5333	8	0.9837	0.5056	0.9837
No log	0.6667	10	0.9513	0.7091	0.9513
No log	0.8	12	0.8371	0.7415	0.8371
No log	0.9333	14	0.7233	0.6918	0.7233
No log	1.0667	16	0.6922	0.7196	0.6922
No log	1.2	18	0.6535	0.7225	0.6535
No log	1.3333	20	0.6214	0.6382	0.6214
No log	1.4667	22	0.6214	0.6481	0.6214
No log	1.6	24	0.6577	0.6929	0.6577
No log	1.7333	26	0.6144	0.6652	0.6144
No log	1.8667	28	0.7474	0.5722	0.7474
No log	2.0	30	0.6810	0.6294	0.6810
No log	2.1333	32	0.6202	0.7358	0.6202
No log	2.2667	34	0.6151	0.6751	0.6151
No log	2.4	36	0.6002	0.6738	0.6002
No log	2.5333	38	0.5749	0.7000	0.5749
No log	2.6667	40	0.5666	0.7675	0.5666
No log	2.8	42	0.6051	0.7937	0.6051
No log	2.9333	44	0.5435	0.7839	0.5435
No log	3.0667	46	0.5462	0.6799	0.5462
No log	3.2	48	0.5199	0.7618	0.5199
No log	3.3333	50	0.5686	0.7827	0.5686
No log	3.4667	52	0.5681	0.7750	0.5681
No log	3.6	54	0.5202	0.7756	0.5202
No log	3.7333	56	0.5295	0.7412	0.5295
No log	3.8667	58	0.5478	0.7632	0.5478
No log	4.0	60	0.5733	0.7573	0.5733
No log	4.1333	62	0.6257	0.7566	0.6257
No log	4.2667	64	0.5977	0.7451	0.5977
No log	4.4	66	0.6316	0.6778	0.6316
No log	4.5333	68	0.5693	0.7216	0.5693
No log	4.6667	70	0.5614	0.7730	0.5614
No log	4.8	72	0.5561	0.7771	0.5561
No log	4.9333	74	0.5439	0.7692	0.5439
No log	5.0667	76	0.5722	0.7788	0.5722
No log	5.2	78	0.5816	0.7777	0.5816
No log	5.3333	80	0.5828	0.7436	0.5828
No log	5.4667	82	0.5916	0.6944	0.5916
No log	5.6	84	0.5655	0.7296	0.5655
No log	5.7333	86	0.5767	0.7728	0.5767
No log	5.8667	88	0.5590	0.7795	0.5590
No log	6.0	90	0.5207	0.7385	0.5207
No log	6.1333	92	0.5177	0.7374	0.5177
No log	6.2667	94	0.5230	0.7756	0.5230
No log	6.4	96	0.5599	0.7771	0.5599
No log	6.5333	98	0.5687	0.7802	0.5687
No log	6.6667	100	0.5450	0.7589	0.5450
No log	6.8	102	0.5470	0.7329	0.5470
No log	6.9333	104	0.5539	0.7073	0.5539
No log	7.0667	106	0.5586	0.7035	0.5586
No log	7.2	108	0.5647	0.7449	0.5647
No log	7.3333	110	0.5807	0.7533	0.5807
No log	7.4667	112	0.5667	0.7369	0.5667
No log	7.6	114	0.5602	0.7153	0.5602
No log	7.7333	116	0.5570	0.7251	0.5570
No log	7.8667	118	0.5524	0.7335	0.5524
No log	8.0	120	0.5563	0.7569	0.5563
No log	8.1333	122	0.5503	0.7474	0.5503
No log	8.2667	124	0.5434	0.7544	0.5434
No log	8.4	126	0.5579	0.7695	0.5579
No log	8.5333	128	0.5668	0.7794	0.5668
No log	8.6667	130	0.5537	0.7567	0.5537
No log	8.8	132	0.5391	0.7511	0.5391
No log	8.9333	134	0.5365	0.7457	0.5365
No log	9.0667	136	0.5367	0.7452	0.5367
No log	9.2	138	0.5391	0.7439	0.5391
No log	9.3333	140	0.5431	0.7438	0.5431
No log	9.4667	142	0.5435	0.7438	0.5435
No log	9.6	144	0.5455	0.7477	0.5455
No log	9.7333	146	0.5440	0.7438	0.5440
No log	9.8667	148	0.5422	0.7427	0.5422
No log	10.0	150	0.5415	0.7472	0.5415

Framework versions

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