arabert_cross_organization_task7_fold6

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: 0.6600
• Qwk: 0.5581
• Mse: 0.6587

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	2.0463	0.0838	2.0448
No log	0.25	4	1.0746	0.1939	1.0725
No log	0.375	6	1.0643	0.3449	1.0639
No log	0.5	8	0.8689	0.5214	0.8684
No log	0.625	10	0.7524	0.3263	0.7518
No log	0.75	12	0.6423	0.3754	0.6420
No log	0.875	14	0.5644	0.5844	0.5643
No log	1.0	16	0.5224	0.6557	0.5223
No log	1.125	18	0.4855	0.6292	0.4850
No log	1.25	20	0.5779	0.5418	0.5767
No log	1.375	22	0.5208	0.6043	0.5197
No log	1.5	24	0.5175	0.7174	0.5174
No log	1.625	26	0.4998	0.7107	0.4998
No log	1.75	28	0.4818	0.6457	0.4809
No log	1.875	30	0.4990	0.6364	0.4979
No log	2.0	32	0.5085	0.6403	0.5073
No log	2.125	34	0.4978	0.6611	0.4969
No log	2.25	36	0.4811	0.6848	0.4805
No log	2.375	38	0.4675	0.6672	0.4669
No log	2.5	40	0.4889	0.6232	0.4881
No log	2.625	42	0.5071	0.6102	0.5062
No log	2.75	44	0.5162	0.6263	0.5151
No log	2.875	46	0.5184	0.6317	0.5172
No log	3.0	48	0.5229	0.6543	0.5219
No log	3.125	50	0.5389	0.6233	0.5377
No log	3.25	52	0.5879	0.5675	0.5861
No log	3.375	54	0.6183	0.5488	0.6164
No log	3.5	56	0.5578	0.5898	0.5563
No log	3.625	58	0.5612	0.6909	0.5607
No log	3.75	60	0.5964	0.7100	0.5962
No log	3.875	62	0.5615	0.6815	0.5609
No log	4.0	64	0.5730	0.5963	0.5716
No log	4.125	66	0.6867	0.5243	0.6849
No log	4.25	68	0.6700	0.5276	0.6682
No log	4.375	70	0.5889	0.5659	0.5873
No log	4.5	72	0.5446	0.6149	0.5434
No log	4.625	74	0.5556	0.6355	0.5547
No log	4.75	76	0.5886	0.6034	0.5871
No log	4.875	78	0.6730	0.5568	0.6709
No log	5.0	80	0.6892	0.5344	0.6871
No log	5.125	82	0.6046	0.5665	0.6029
No log	5.25	84	0.5605	0.6134	0.5591
No log	5.375	86	0.5415	0.6417	0.5404
No log	5.5	88	0.5515	0.6247	0.5504
No log	5.625	90	0.5964	0.5762	0.5948
No log	5.75	92	0.6466	0.5489	0.6449
No log	5.875	94	0.6325	0.5648	0.6310
No log	6.0	96	0.6036	0.6097	0.6022
No log	6.125	98	0.5955	0.6483	0.5944
No log	6.25	100	0.6017	0.6168	0.6005
No log	6.375	102	0.6349	0.5846	0.6335
No log	6.5	104	0.6941	0.5277	0.6925
No log	6.625	106	0.6740	0.5262	0.6724
No log	6.75	108	0.6043	0.5829	0.6030
No log	6.875	110	0.5813	0.6039	0.5802
No log	7.0	112	0.5847	0.6056	0.5836
No log	7.125	114	0.6031	0.5987	0.6019
No log	7.25	116	0.6490	0.5645	0.6474
No log	7.375	118	0.6772	0.5326	0.6756
No log	7.5	120	0.6849	0.5311	0.6833
No log	7.625	122	0.6620	0.5393	0.6606
No log	7.75	124	0.6230	0.5696	0.6217
No log	7.875	126	0.5912	0.5983	0.5901
No log	8.0	128	0.5924	0.5983	0.5913
No log	8.125	130	0.6124	0.5864	0.6112
No log	8.25	132	0.6364	0.5615	0.6351
No log	8.375	134	0.6650	0.5476	0.6635
No log	8.5	136	0.6693	0.5397	0.6678
No log	8.625	138	0.6639	0.5516	0.6624
No log	8.75	140	0.6658	0.5467	0.6643
No log	8.875	142	0.6772	0.5437	0.6757
No log	9.0	144	0.6778	0.5489	0.6763
No log	9.125	146	0.6641	0.5504	0.6627
No log	9.25	148	0.6614	0.5557	0.6600
No log	9.375	150	0.6564	0.5609	0.6551
No log	9.5	152	0.6530	0.5618	0.6517
No log	9.625	154	0.6533	0.5618	0.6520
No log	9.75	156	0.6546	0.5581	0.6533
No log	9.875	158	0.6579	0.5581	0.6566
No log	10.0	160	0.6600	0.5581	0.6587

Framework versions

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