arabert_cross_organization_task4_fold0

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: 0.7387
• Qwk: 0.5925
• Mse: 0.7387

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	3.7003	0.0594	3.6971
No log	0.2353	4	2.4103	0.0127	2.4085
No log	0.3529	6	1.7582	0.2177	1.7571
No log	0.4706	8	1.5666	0.3222	1.5658
No log	0.5882	10	0.9860	0.4576	0.9866
No log	0.7059	12	0.9585	0.5399	0.9591
No log	0.8235	14	1.3953	0.4078	1.3949
No log	0.9412	16	1.6278	0.3501	1.6269
No log	1.0588	18	1.2485	0.4062	1.2480
No log	1.1765	20	0.9325	0.4653	0.9323
No log	1.2941	22	0.8750	0.5046	0.8747
No log	1.4118	24	0.9684	0.4851	0.9679
No log	1.5294	26	1.3655	0.3843	1.3644
No log	1.6471	28	1.0699	0.4636	1.0693
No log	1.7647	30	0.7323	0.6264	0.7322
No log	1.8824	32	0.7409	0.6340	0.7408
No log	2.0	34	0.7342	0.6187	0.7341
No log	2.1176	36	0.7935	0.5765	0.7932
No log	2.2353	38	1.0980	0.4535	1.0972
No log	2.3529	40	1.0511	0.4714	1.0504
No log	2.4706	42	0.8556	0.5268	0.8552
No log	2.5882	44	0.7684	0.5983	0.7681
No log	2.7059	46	0.8207	0.5721	0.8206
No log	2.8235	48	0.9052	0.5593	0.9050
No log	2.9412	50	0.8417	0.5743	0.8416
No log	3.0588	52	0.7671	0.6038	0.7671
No log	3.1765	54	0.7859	0.5949	0.7859
No log	3.2941	56	0.8844	0.5560	0.8842
No log	3.4118	58	0.9141	0.5516	0.9137
No log	3.5294	60	0.7828	0.5943	0.7825
No log	3.6471	62	0.7194	0.6082	0.7191
No log	3.7647	64	0.7235	0.6070	0.7233
No log	3.8824	66	0.7606	0.5926	0.7604
No log	4.0	68	0.9394	0.5299	0.9388
No log	4.1176	70	0.9642	0.5092	0.9636
No log	4.2353	72	0.7479	0.5973	0.7477
No log	4.3529	74	0.6642	0.6133	0.6642
No log	4.4706	76	0.6706	0.6129	0.6705
No log	4.5882	78	0.7184	0.6006	0.7183
No log	4.7059	80	0.7525	0.5921	0.7523
No log	4.8235	82	0.7429	0.5934	0.7428
No log	4.9412	84	0.7555	0.5857	0.7554
No log	5.0588	86	0.7166	0.5978	0.7165
No log	5.1765	88	0.7034	0.5949	0.7033
No log	5.2941	90	0.7179	0.5869	0.7178
No log	5.4118	92	0.8419	0.5574	0.8416
No log	5.5294	94	0.9040	0.5229	0.9037
No log	5.6471	96	0.8841	0.5357	0.8839
No log	5.7647	98	0.7383	0.5893	0.7383
No log	5.8824	100	0.6787	0.6146	0.6789
No log	6.0	102	0.6933	0.6105	0.6933
No log	6.1176	104	0.7751	0.5812	0.7749
No log	6.2353	106	0.7938	0.5700	0.7936
No log	6.3529	108	0.7205	0.6067	0.7204
No log	6.4706	110	0.6695	0.6121	0.6695
No log	6.5882	112	0.6759	0.6171	0.6758
No log	6.7059	114	0.6919	0.6099	0.6918
No log	6.8235	116	0.7467	0.5990	0.7465
No log	6.9412	118	0.7453	0.5978	0.7452
No log	7.0588	120	0.7515	0.5967	0.7514
No log	7.1765	122	0.7221	0.6021	0.7221
No log	7.2941	124	0.7307	0.6043	0.7307
No log	7.4118	126	0.7413	0.6043	0.7414
No log	7.5294	128	0.7247	0.6004	0.7249
No log	7.6471	130	0.7088	0.6058	0.7091
No log	7.7647	132	0.7232	0.6037	0.7235
No log	7.8824	134	0.7747	0.5935	0.7748
No log	8.0	136	0.8677	0.5568	0.8676
No log	8.1176	138	0.8769	0.5521	0.8768
No log	8.2353	140	0.8117	0.5845	0.8116
No log	8.3529	142	0.7379	0.5954	0.7379
No log	8.4706	144	0.7203	0.6024	0.7203
No log	8.5882	146	0.7255	0.6036	0.7255
No log	8.7059	148	0.7177	0.6024	0.7177
No log	8.8235	150	0.7099	0.6036	0.7099
No log	8.9412	152	0.7149	0.6036	0.7150
No log	9.0588	154	0.7323	0.5976	0.7323
No log	9.1765	156	0.7453	0.5919	0.7453
No log	9.2941	158	0.7533	0.5901	0.7533
No log	9.4118	160	0.7538	0.5902	0.7538
No log	9.5294	162	0.7490	0.5903	0.7490
No log	9.6471	164	0.7411	0.5952	0.7411
No log	9.7647	166	0.7407	0.5952	0.7407
No log	9.8824	168	0.7383	0.5925	0.7383
No log	10.0	170	0.7387	0.5925	0.7387

Framework versions

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