ArabicNewSplits5_FineTuningAraBERT_run1_AugV5_k4_task5_organization

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: 0.7070
• Qwk: 0.7427
• Mse: 0.7070
• Rmse: 0.8408

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	0.0909	2	2.3002	0.0053	2.3002	1.5166
No log	0.1818	4	1.7960	0.0129	1.7960	1.3401
No log	0.2727	6	1.5699	0.0847	1.5699	1.2530
No log	0.3636	8	1.3694	0.1335	1.3694	1.1702
No log	0.4545	10	1.3771	0.1894	1.3771	1.1735
No log	0.5455	12	1.3646	0.1451	1.3646	1.1682
No log	0.6364	14	1.3371	0.1700	1.3371	1.1563
No log	0.7273	16	1.3432	0.1804	1.3432	1.1590
No log	0.8182	18	1.3344	0.2066	1.3344	1.1552
No log	0.9091	20	1.2542	0.2058	1.2542	1.1199
No log	1.0	22	1.2294	0.2412	1.2294	1.1088
No log	1.0909	24	1.2249	0.2561	1.2249	1.1068
No log	1.1818	26	1.2426	0.2800	1.2426	1.1147
No log	1.2727	28	1.1835	0.2908	1.1835	1.0879
No log	1.3636	30	1.1645	0.3558	1.1645	1.0791
No log	1.4545	32	1.1548	0.3524	1.1548	1.0746
No log	1.5455	34	1.0714	0.3968	1.0714	1.0351
No log	1.6364	36	1.0664	0.4709	1.0664	1.0327
No log	1.7273	38	1.1887	0.4167	1.1887	1.0903
No log	1.8182	40	1.2642	0.4052	1.2642	1.1243
No log	1.9091	42	1.1993	0.4393	1.1993	1.0951
No log	2.0	44	1.1735	0.4393	1.1735	1.0833
No log	2.0909	46	1.1728	0.4485	1.1728	1.0830
No log	2.1818	48	1.0419	0.4680	1.0419	1.0207
No log	2.2727	50	0.9564	0.5203	0.9564	0.9780
No log	2.3636	52	0.9706	0.5345	0.9706	0.9852
No log	2.4545	54	0.9865	0.5704	0.9865	0.9932
No log	2.5455	56	0.8902	0.6030	0.8902	0.9435
No log	2.6364	58	0.7935	0.6328	0.7935	0.8908
No log	2.7273	60	0.7789	0.6653	0.7789	0.8825
No log	2.8182	62	0.7488	0.7074	0.7488	0.8653
No log	2.9091	64	0.7456	0.6839	0.7456	0.8635
No log	3.0	66	0.8389	0.6667	0.8389	0.9159
No log	3.0909	68	1.0144	0.5859	1.0144	1.0072
No log	3.1818	70	0.8496	0.6475	0.8496	0.9218
No log	3.2727	72	0.6805	0.7498	0.6805	0.8249
No log	3.3636	74	0.6795	0.7455	0.6795	0.8243
No log	3.4545	76	0.6873	0.7393	0.6873	0.8290
No log	3.5455	78	0.6917	0.7254	0.6917	0.8317
No log	3.6364	80	0.7888	0.7084	0.7888	0.8882
No log	3.7273	82	0.7873	0.7087	0.7873	0.8873
No log	3.8182	84	0.7568	0.7045	0.7568	0.8699
No log	3.9091	86	0.6449	0.7308	0.6449	0.8031
No log	4.0	88	0.6303	0.7561	0.6303	0.7939
No log	4.0909	90	0.6499	0.7430	0.6499	0.8062
No log	4.1818	92	0.7156	0.7080	0.7156	0.8459
No log	4.2727	94	0.7550	0.6824	0.7550	0.8689
No log	4.3636	96	0.6813	0.7256	0.6813	0.8254
No log	4.4545	98	0.6441	0.7347	0.6441	0.8026
No log	4.5455	100	0.6183	0.7420	0.6183	0.7863
No log	4.6364	102	0.6718	0.7445	0.6718	0.8196
No log	4.7273	104	0.7536	0.7004	0.7536	0.8681
No log	4.8182	106	0.7231	0.6940	0.7231	0.8503
No log	4.9091	108	0.6383	0.7507	0.6383	0.7989
No log	5.0	110	0.6088	0.7530	0.6088	0.7802
No log	5.0909	112	0.6224	0.7096	0.6224	0.7889
No log	5.1818	114	0.6193	0.7591	0.6193	0.7869
No log	5.2727	116	0.6635	0.7447	0.6635	0.8145
No log	5.3636	118	0.7432	0.6964	0.7432	0.8621
No log	5.4545	120	0.8885	0.6736	0.8885	0.9426
No log	5.5455	122	0.8964	0.6736	0.8964	0.9468
No log	5.6364	124	0.8088	0.7055	0.8088	0.8993
No log	5.7273	126	0.7482	0.7112	0.7482	0.8650
No log	5.8182	128	0.7106	0.7270	0.7106	0.8430
No log	5.9091	130	0.6985	0.7442	0.6985	0.8358
No log	6.0	132	0.7573	0.7188	0.7573	0.8702
No log	6.0909	134	0.9545	0.6282	0.9545	0.9770
No log	6.1818	136	1.0673	0.6205	1.0673	1.0331
No log	6.2727	138	0.9725	0.6313	0.9725	0.9862
No log	6.3636	140	0.8174	0.7117	0.8174	0.9041
No log	6.4545	142	0.7071	0.7369	0.7071	0.8409
No log	6.5455	144	0.6574	0.7166	0.6574	0.8108
No log	6.6364	146	0.6593	0.7188	0.6593	0.8120
No log	6.7273	148	0.6617	0.7187	0.6617	0.8134
No log	6.8182	150	0.6590	0.7233	0.6590	0.8118
No log	6.9091	152	0.6848	0.7441	0.6848	0.8275
No log	7.0	154	0.7230	0.7353	0.7230	0.8503
No log	7.0909	156	0.7503	0.7206	0.7503	0.8662
No log	7.1818	158	0.8117	0.6876	0.8117	0.9009
No log	7.2727	160	0.9137	0.6551	0.9137	0.9559
No log	7.3636	162	1.0150	0.6362	1.0150	1.0075
No log	7.4545	164	1.0150	0.6282	1.0150	1.0075
No log	7.5455	166	0.9261	0.6401	0.9261	0.9623
No log	7.6364	168	0.8214	0.7055	0.8214	0.9063
No log	7.7273	170	0.7496	0.7164	0.7496	0.8658
No log	7.8182	172	0.7314	0.7423	0.7314	0.8552
No log	7.9091	174	0.7321	0.7293	0.7321	0.8556
No log	8.0	176	0.7387	0.7248	0.7387	0.8594
No log	8.0909	178	0.7496	0.7308	0.7496	0.8658
No log	8.1818	180	0.7687	0.7345	0.7687	0.8768
No log	8.2727	182	0.7631	0.7344	0.7631	0.8736
No log	8.3636	184	0.7500	0.7345	0.7500	0.8660
No log	8.4545	186	0.7530	0.7344	0.7530	0.8678
No log	8.5455	188	0.7401	0.7429	0.7401	0.8603
No log	8.6364	190	0.7402	0.7429	0.7402	0.8603
No log	8.7273	192	0.7419	0.7466	0.7419	0.8614
No log	8.8182	194	0.7385	0.7466	0.7385	0.8593
No log	8.9091	196	0.7303	0.7463	0.7303	0.8546
No log	9.0	198	0.7050	0.7429	0.7050	0.8396
No log	9.0909	200	0.6804	0.7545	0.6804	0.8248
No log	9.1818	202	0.6710	0.7619	0.6710	0.8191
No log	9.2727	204	0.6703	0.7614	0.6703	0.8187
No log	9.3636	206	0.6749	0.7561	0.6749	0.8215
No log	9.4545	208	0.6857	0.7574	0.6857	0.8281
No log	9.5455	210	0.6937	0.7429	0.6937	0.8329
No log	9.6364	212	0.7010	0.7429	0.7010	0.8372
No log	9.7273	214	0.7033	0.7429	0.7033	0.8386
No log	9.8182	216	0.7052	0.7427	0.7052	0.8398
No log	9.9091	218	0.7067	0.7427	0.7067	0.8406
No log	10.0	220	0.7070	0.7427	0.7070	0.8408

Framework versions

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