arabert_augWithOrig_disEqu_k1_organization_task3_fold1

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: 0.4299
• Qwk: 0.2924
• Mse: 0.4299
• Rmse: 0.6557

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: 8
• eval_batch_size: 8
• 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	Rmse
No log	0.1538	2	3.4754	-0.0417	3.4754	1.8642
No log	0.3077	4	1.9363	0.1323	1.9363	1.3915
No log	0.4615	6	0.7436	0.1413	0.7436	0.8623
No log	0.6154	8	0.6419	0.0763	0.6419	0.8012
No log	0.7692	10	0.6199	0.0120	0.6199	0.7873
No log	0.9231	12	0.7422	0.0	0.7422	0.8615
No log	1.0769	14	0.8566	0.0	0.8566	0.9255
No log	1.2308	16	0.7918	0.0	0.7918	0.8898
No log	1.3846	18	0.6896	0.0	0.6896	0.8304
No log	1.5385	20	0.6602	0.0	0.6602	0.8125
No log	1.6923	22	0.6098	0.0	0.6098	0.7809
No log	1.8462	24	0.6193	0.0	0.6193	0.7870
No log	2.0	26	0.5895	0.0	0.5895	0.7678
No log	2.1538	28	0.5361	0.0	0.5361	0.7322
No log	2.3077	30	0.5055	0.0120	0.5055	0.7110
No log	2.4615	32	0.5259	0.3186	0.5259	0.7252
No log	2.6154	34	0.5209	0.2443	0.5209	0.7217
No log	2.7692	36	0.6275	0.0	0.6275	0.7922
No log	2.9231	38	0.8433	0.0	0.8433	0.9183
No log	3.0769	40	1.1791	0.0756	1.1791	1.0858
No log	3.2308	42	1.0044	0.0	1.0044	1.0022
No log	3.3846	44	0.6815	0.0	0.6815	0.8256
No log	3.5385	46	0.5598	0.0	0.5598	0.7482
No log	3.6923	48	0.5428	0.1437	0.5428	0.7367
No log	3.8462	50	0.5920	0.0	0.5920	0.7694
No log	4.0	52	0.6133	0.0	0.6133	0.7831
No log	4.1538	54	0.6115	0.0	0.6115	0.7820
No log	4.3077	56	0.5877	0.0	0.5877	0.7666
No log	4.4615	58	0.6311	0.0	0.6311	0.7944
No log	4.6154	60	0.6503	0.0	0.6503	0.8064
No log	4.7692	62	0.5752	0.0	0.5752	0.7584
No log	4.9231	64	0.5321	0.0	0.5321	0.7295
No log	5.0769	66	0.5299	0.0	0.5299	0.7280
No log	5.2308	68	0.5443	0.0	0.5443	0.7378
No log	5.3846	70	0.5221	0.1437	0.5221	0.7225
No log	5.5385	72	0.5033	0.1437	0.5033	0.7094
No log	5.6923	74	0.4764	0.1437	0.4764	0.6902
No log	5.8462	76	0.4689	0.1437	0.4689	0.6847
No log	6.0	78	0.4610	0.1437	0.4610	0.6789
No log	6.1538	80	0.4547	0.1437	0.4547	0.6743
No log	6.3077	82	0.5185	0.2222	0.5185	0.7201
No log	6.4615	84	0.5746	0.1834	0.5746	0.7580
No log	6.6154	86	0.7710	0.3231	0.7710	0.8781
No log	6.7692	88	0.7664	0.3231	0.7664	0.8754
No log	6.9231	90	0.6473	0.3231	0.6473	0.8046
No log	7.0769	92	0.6733	0.3231	0.6733	0.8206
No log	7.2308	94	0.5743	0.1834	0.5743	0.7578
No log	7.3846	96	0.5071	0.1834	0.5071	0.7121
No log	7.5385	98	0.5820	0.1834	0.5819	0.7629
No log	7.6923	100	0.6588	0.1834	0.6588	0.8117
No log	7.8462	102	0.6029	0.1834	0.6029	0.7764
No log	8.0	104	0.5449	0.1834	0.5449	0.7382
No log	8.1538	106	0.5040	0.2000	0.5040	0.7099
No log	8.3077	108	0.5017	0.2000	0.5017	0.7083
No log	8.4615	110	0.5231	0.1834	0.5231	0.7233
No log	8.6154	112	0.5106	0.2000	0.5106	0.7146
No log	8.7692	114	0.4726	0.2180	0.4726	0.6874
No log	8.9231	116	0.4493	0.1444	0.4493	0.6703
No log	9.0769	118	0.4457	0.1444	0.4457	0.6676
No log	9.2308	120	0.4301	0.2924	0.4301	0.6558
No log	9.3846	122	0.4282	0.2924	0.4282	0.6544
No log	9.5385	124	0.4313	0.2924	0.4313	0.6567
No log	9.6923	126	0.4357	0.2924	0.4357	0.6601
No log	9.8462	128	0.4314	0.2924	0.4314	0.6568
No log	10.0	130	0.4299	0.2924	0.4299	0.6557

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1