ArabicNewSplits5_FineTuningAraBERT_run3_AugV5_k3_task1_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: 0.7450
• Qwk: 0.7077
• Mse: 0.7450
• Rmse: 0.8632

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	5.4652	-0.0207	5.4652	2.3378
No log	0.1818	4	3.3809	0.0611	3.3809	1.8387
No log	0.2727	6	1.9927	0.0840	1.9927	1.4116
No log	0.3636	8	1.5362	0.0187	1.5362	1.2394
No log	0.4545	10	1.3395	0.1263	1.3395	1.1574
No log	0.5455	12	1.2577	0.2686	1.2577	1.1215
No log	0.6364	14	1.2555	0.2541	1.2555	1.1205
No log	0.7273	16	1.3868	0.1104	1.3868	1.1776
No log	0.8182	18	1.5574	0.1270	1.5574	1.2480
No log	0.9091	20	1.4184	0.1834	1.4184	1.1910
No log	1.0	22	1.2148	0.2866	1.2148	1.1022
No log	1.0909	24	1.2918	0.2892	1.2918	1.1366
No log	1.1818	26	1.1910	0.3524	1.1910	1.0913
No log	1.2727	28	1.0363	0.5231	1.0363	1.0180
No log	1.3636	30	1.0043	0.5238	1.0043	1.0021
No log	1.4545	32	0.9816	0.5244	0.9816	0.9908
No log	1.5455	34	1.2631	0.4983	1.2631	1.1239
No log	1.6364	36	1.4017	0.4388	1.4017	1.1839
No log	1.7273	38	1.0116	0.5560	1.0116	1.0058
No log	1.8182	40	0.8942	0.5624	0.8942	0.9456
No log	1.9091	42	0.9205	0.5587	0.9205	0.9594
No log	2.0	44	0.9202	0.5671	0.9202	0.9593
No log	2.0909	46	0.8065	0.5973	0.8065	0.8981
No log	2.1818	48	0.8410	0.5766	0.8410	0.9171
No log	2.2727	50	0.8365	0.5764	0.8365	0.9146
No log	2.3636	52	0.8466	0.5711	0.8466	0.9201
No log	2.4545	54	0.8609	0.5951	0.8609	0.9278
No log	2.5455	56	0.7560	0.6394	0.7560	0.8695
No log	2.6364	58	0.6699	0.6566	0.6699	0.8184
No log	2.7273	60	0.6846	0.6070	0.6846	0.8274
No log	2.8182	62	0.6367	0.6861	0.6367	0.7980
No log	2.9091	64	0.6384	0.6861	0.6384	0.7990
No log	3.0	66	0.8295	0.6169	0.8295	0.9108
No log	3.0909	68	0.8862	0.6026	0.8862	0.9414
No log	3.1818	70	0.7573	0.6408	0.7573	0.8702
No log	3.2727	72	0.6174	0.6932	0.6174	0.7857
No log	3.3636	74	0.7613	0.6102	0.7613	0.8725
No log	3.4545	76	0.8870	0.5083	0.8870	0.9418
No log	3.5455	78	0.7687	0.6105	0.7687	0.8768
No log	3.6364	80	0.6294	0.7255	0.6294	0.7933
No log	3.7273	82	0.7687	0.6987	0.7687	0.8767
No log	3.8182	84	0.8426	0.6412	0.8426	0.9179
No log	3.9091	86	0.7471	0.6967	0.7471	0.8643
No log	4.0	88	0.6445	0.7367	0.6445	0.8028
No log	4.0909	90	0.7149	0.7045	0.7149	0.8455
No log	4.1818	92	0.8078	0.6342	0.8078	0.8988
No log	4.2727	94	0.7739	0.6719	0.7739	0.8797
No log	4.3636	96	0.6948	0.7430	0.6948	0.8335
No log	4.4545	98	0.6851	0.7240	0.6851	0.8277
No log	4.5455	100	0.6741	0.7416	0.6741	0.8210
No log	4.6364	102	0.6582	0.7407	0.6582	0.8113
No log	4.7273	104	0.6647	0.7205	0.6647	0.8153
No log	4.8182	106	0.6671	0.6988	0.6671	0.8167
No log	4.9091	108	0.6605	0.7001	0.6605	0.8127
No log	5.0	110	0.6552	0.7013	0.6552	0.8094
No log	5.0909	112	0.6655	0.6977	0.6655	0.8158
No log	5.1818	114	0.6892	0.6603	0.6892	0.8302
No log	5.2727	116	0.7353	0.6719	0.7353	0.8575
No log	5.3636	118	0.7838	0.6623	0.7838	0.8854
No log	5.4545	120	0.8208	0.6661	0.8208	0.9060
No log	5.5455	122	0.7975	0.6609	0.7975	0.8930
No log	5.6364	124	0.7567	0.7126	0.7567	0.8699
No log	5.7273	126	0.7276	0.7239	0.7276	0.8530
No log	5.8182	128	0.7156	0.7173	0.7156	0.8459
No log	5.9091	130	0.7048	0.7236	0.7048	0.8395
No log	6.0	132	0.6983	0.7116	0.6983	0.8356
No log	6.0909	134	0.7042	0.7117	0.7042	0.8392
No log	6.1818	136	0.7342	0.6825	0.7342	0.8568
No log	6.2727	138	0.7190	0.6993	0.7190	0.8479
No log	6.3636	140	0.7000	0.7108	0.7000	0.8367
No log	6.4545	142	0.7061	0.7144	0.7061	0.8403
No log	6.5455	144	0.7106	0.7179	0.7106	0.8430
No log	6.6364	146	0.7249	0.7159	0.7249	0.8514
No log	6.7273	148	0.7462	0.7140	0.7462	0.8638
No log	6.8182	150	0.7533	0.7169	0.7533	0.8679
No log	6.9091	152	0.7582	0.7266	0.7582	0.8707
No log	7.0	154	0.7649	0.7113	0.7649	0.8746
No log	7.0909	156	0.7746	0.6998	0.7746	0.8801
No log	7.1818	158	0.7774	0.6830	0.7774	0.8817
No log	7.2727	160	0.7809	0.6865	0.7809	0.8837
No log	7.3636	162	0.7808	0.7012	0.7808	0.8836
No log	7.4545	164	0.7811	0.6928	0.7811	0.8838
No log	7.5455	166	0.7881	0.6877	0.7881	0.8877
No log	7.6364	168	0.7932	0.6906	0.7932	0.8906
No log	7.7273	170	0.7866	0.6906	0.7866	0.8869
No log	7.8182	172	0.7859	0.7082	0.7859	0.8865
No log	7.9091	174	0.7787	0.7052	0.7787	0.8824
No log	8.0	176	0.7686	0.6933	0.7686	0.8767
No log	8.0909	178	0.7657	0.6909	0.7657	0.8751
No log	8.1818	180	0.7693	0.6872	0.7693	0.8771
No log	8.2727	182	0.7769	0.6872	0.7769	0.8814
No log	8.3636	184	0.7849	0.6836	0.7849	0.8860
No log	8.4545	186	0.7939	0.6926	0.7939	0.8910
No log	8.5455	188	0.7947	0.7016	0.7947	0.8915
No log	8.6364	190	0.7915	0.7016	0.7915	0.8897
No log	8.7273	192	0.7827	0.6973	0.7827	0.8847
No log	8.8182	194	0.7708	0.7108	0.7708	0.8779
No log	8.9091	196	0.7620	0.7048	0.7620	0.8729
No log	9.0	198	0.7530	0.7031	0.7530	0.8677
No log	9.0909	200	0.7490	0.6970	0.7490	0.8655
No log	9.1818	202	0.7457	0.6970	0.7457	0.8635
No log	9.2727	204	0.7441	0.6970	0.7441	0.8626
No log	9.3636	206	0.7453	0.6982	0.7453	0.8633
No log	9.4545	208	0.7445	0.6994	0.7445	0.8628
No log	9.5455	210	0.7434	0.7138	0.7434	0.8622
No log	9.6364	212	0.7429	0.7138	0.7429	0.8619
No log	9.7273	214	0.7431	0.7077	0.7431	0.8620
No log	9.8182	216	0.7437	0.7077	0.7437	0.8624
No log	9.9091	218	0.7446	0.7077	0.7446	0.8629
No log	10.0	220	0.7450	0.7077	0.7450	0.8632

Framework versions

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