arabert_cross_organization_task7_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: 1.2954
• Qwk: 0.0375
• Mse: 1.2921

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	5.8813	0.0002	5.8787
No log	0.25	4	2.9326	0.0187	2.9290
No log	0.375	6	1.3335	0.0483	1.3299
No log	0.5	8	0.8293	0.1248	0.8268
No log	0.625	10	0.8358	0.0116	0.8324
No log	0.75	12	0.9072	0.0353	0.9036
No log	0.875	14	0.8593	-0.0221	0.8566
No log	1.0	16	0.9323	0.0032	0.9303
No log	1.125	18	0.9667	-0.0502	0.9646
No log	1.25	20	1.0888	0.0	1.0865
No log	1.375	22	1.3618	0.0	1.3592
No log	1.5	24	1.5735	0.0025	1.5707
No log	1.625	26	1.5108	0.0226	1.5078
No log	1.75	28	1.4546	0.0226	1.4515
No log	1.875	30	1.4052	0.0182	1.4022
No log	2.0	32	1.2679	0.0182	1.2650
No log	2.125	34	1.3052	0.0182	1.3025
No log	2.25	36	1.3852	0.0	1.3827
No log	2.375	38	1.3541	0.0182	1.3513
No log	2.5	40	1.1151	0.0182	1.1121
No log	2.625	42	1.0936	0.0182	1.0906
No log	2.75	44	1.2306	0.0182	1.2277
No log	2.875	46	1.3712	0.0182	1.3683
No log	3.0	48	1.3673	0.0182	1.3643
No log	3.125	50	1.1531	-0.0018	1.1496
No log	3.25	52	1.1023	0.0034	1.0987
No log	3.375	54	1.2866	0.0421	1.2831
No log	3.5	56	1.5177	0.0289	1.5145
No log	3.625	58	1.5134	0.0204	1.5104
No log	3.75	60	1.4435	0.0182	1.4406
No log	3.875	62	1.4043	0.0204	1.4012
No log	4.0	64	1.3959	0.0226	1.3928
No log	4.125	66	1.2847	0.0204	1.2816
No log	4.25	68	1.1653	0.0182	1.1623
No log	4.375	70	1.1255	0.0155	1.1225
No log	4.5	72	1.1599	0.0182	1.1570
No log	4.625	74	1.1826	0.0155	1.1796
No log	4.75	76	1.2119	0.0182	1.2088
No log	4.875	78	1.2305	0.0182	1.2274
No log	5.0	80	1.3039	0.0247	1.3006
No log	5.125	82	1.2190	0.0196	1.2157
No log	5.25	84	1.2372	0.0238	1.2338
No log	5.375	86	1.3596	0.0289	1.3563
No log	5.5	88	1.3896	0.0268	1.3865
No log	5.625	90	1.4002	0.0268	1.3971
No log	5.75	92	1.3231	0.0204	1.3202
No log	5.875	94	1.2852	0.0182	1.2824
No log	6.0	96	1.3081	0.0182	1.3054
No log	6.125	98	1.2707	0.0182	1.2681
No log	6.25	100	1.2533	0.0182	1.2505
No log	6.375	102	1.2295	0.0182	1.2266
No log	6.5	104	1.2384	0.0633	1.2354
No log	6.625	106	1.3078	0.0483	1.3048
No log	6.75	108	1.3657	0.0501	1.3628
No log	6.875	110	1.4466	0.0551	1.4437
No log	7.0	112	1.4426	0.0567	1.4396
No log	7.125	114	1.3478	0.0557	1.3447
No log	7.25	116	1.2982	0.1010	1.2950
No log	7.375	118	1.2917	0.0632	1.2886
No log	7.5	120	1.3501	0.0525	1.3471
No log	7.625	122	1.4125	0.0567	1.4094
No log	7.75	124	1.4720	0.0777	1.4689
No log	7.875	126	1.4728	0.0764	1.4697
No log	8.0	128	1.4105	0.0535	1.4075
No log	8.125	130	1.3530	0.0289	1.3500
No log	8.25	132	1.3150	0.0289	1.3120
No log	8.375	134	1.3406	0.0268	1.3376
No log	8.5	136	1.3885	0.0309	1.3855
No log	8.625	138	1.4249	0.0551	1.4218
No log	8.75	140	1.4578	0.0598	1.4546
No log	8.875	142	1.4364	0.0583	1.4332
No log	9.0	144	1.3917	0.0341	1.3885
No log	9.125	146	1.3599	0.0133	1.3567
No log	9.25	148	1.3391	0.0754	1.3359
No log	9.375	150	1.3212	0.0754	1.3179
No log	9.5	152	1.2992	0.0375	1.2960
No log	9.625	154	1.2901	0.0351	1.2868
No log	9.75	156	1.2864	0.0351	1.2831
No log	9.875	158	1.2913	0.0375	1.2881
No log	10.0	160	1.2954	0.0375	1.2921

Framework versions

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