ArabicNewSplits6_FineTuningAraBERT_run2_AugV5_k5_task3_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: 0.7210
• Qwk: 0.2511
• Mse: 0.7210
• Rmse: 0.8491

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	0.08	2	3.3894	-0.0227	3.3894	1.8410
No log	0.16	4	1.8704	-0.0390	1.8704	1.3676
No log	0.24	6	1.4657	0.0255	1.4657	1.2106
No log	0.32	8	1.0716	-0.0159	1.0716	1.0352
No log	0.4	10	0.6668	0.0189	0.6668	0.8166
No log	0.48	12	0.6269	-0.0794	0.6269	0.7918
No log	0.56	14	0.6955	-0.1473	0.6955	0.8340
No log	0.64	16	0.7573	-0.1200	0.7573	0.8702
No log	0.72	18	0.6683	-0.1429	0.6683	0.8175
No log	0.8	20	0.6238	0.0303	0.6238	0.7898
No log	0.88	22	0.6730	0.1913	0.6730	0.8204
No log	0.96	24	0.6090	0.1304	0.6090	0.7804
No log	1.04	26	0.6373	-0.0732	0.6373	0.7983
No log	1.12	28	0.5971	-0.0159	0.5971	0.7727
No log	1.2	30	0.5935	0.1206	0.5935	0.7704
No log	1.28	32	0.5835	0.0	0.5835	0.7639
No log	1.3600	34	0.6278	0.0	0.6278	0.7923
No log	1.44	36	0.5659	0.0303	0.5659	0.7523
No log	1.52	38	0.6058	0.3103	0.6058	0.7783
No log	1.6	40	0.5919	0.3216	0.5919	0.7693
No log	1.6800	42	0.5912	0.0569	0.5912	0.7689
No log	1.76	44	0.6083	0.0	0.6083	0.7799
No log	1.8400	46	0.5827	0.2704	0.5827	0.7633
No log	1.92	48	0.7381	0.1852	0.7381	0.8591
No log	2.0	50	0.8052	0.0717	0.8052	0.8973
No log	2.08	52	0.7245	0.1765	0.7245	0.8512
No log	2.16	54	0.5726	0.0222	0.5726	0.7567
No log	2.24	56	0.5804	0.0	0.5804	0.7618
No log	2.32	58	0.6148	0.0	0.6148	0.7841
No log	2.4	60	0.6267	0.0222	0.6267	0.7917
No log	2.48	62	0.6476	0.0725	0.6476	0.8048
No log	2.56	64	0.6730	0.0725	0.6730	0.8204
No log	2.64	66	0.6959	0.0685	0.6959	0.8342
No log	2.7200	68	0.7927	0.0739	0.7927	0.8903
No log	2.8	70	0.7214	0.1000	0.7214	0.8493
No log	2.88	72	0.9360	0.0215	0.9360	0.9675
No log	2.96	74	0.9844	-0.0196	0.9844	0.9922
No log	3.04	76	0.8266	0.1917	0.8266	0.9092
No log	3.12	78	0.5706	0.1020	0.5706	0.7553
No log	3.2	80	0.9628	0.1475	0.9628	0.9812
No log	3.2800	82	1.0474	0.1461	1.0474	1.0234
No log	3.36	84	0.6982	0.2239	0.6982	0.8356
No log	3.44	86	0.5754	0.0388	0.5754	0.7585
No log	3.52	88	0.6268	0.0476	0.6268	0.7917
No log	3.6	90	0.5992	0.0303	0.5992	0.7741
No log	3.68	92	0.5856	0.1801	0.5856	0.7652
No log	3.76	94	0.7050	0.1759	0.7050	0.8397
No log	3.84	96	0.6689	0.1739	0.6689	0.8179
No log	3.92	98	0.6302	0.2000	0.6302	0.7939
No log	4.0	100	0.7531	0.1398	0.7531	0.8678
No log	4.08	102	0.7097	0.2609	0.7097	0.8424
No log	4.16	104	0.6015	0.1698	0.6015	0.7755
No log	4.24	106	0.6761	0.2000	0.6761	0.8222
No log	4.32	108	0.5976	0.1807	0.5976	0.7731
No log	4.4	110	0.6131	0.3043	0.6131	0.7830
No log	4.48	112	0.6802	0.2577	0.6802	0.8248
No log	4.5600	114	0.6181	0.3369	0.6181	0.7862
No log	4.64	116	0.5673	0.3043	0.5673	0.7532
No log	4.72	118	0.6183	0.3301	0.6183	0.7863
No log	4.8	120	0.5713	0.4105	0.5713	0.7559
No log	4.88	122	0.7513	0.2140	0.7513	0.8668
No log	4.96	124	0.7455	0.1718	0.7455	0.8634
No log	5.04	126	0.6295	0.3704	0.6295	0.7934
No log	5.12	128	0.7186	0.3719	0.7186	0.8477
No log	5.2	130	0.6685	0.4286	0.6685	0.8176
No log	5.28	132	0.6682	0.2762	0.6682	0.8174
No log	5.36	134	0.9598	0.1128	0.9598	0.9797
No log	5.44	136	1.1666	0.0234	1.1666	1.0801
No log	5.52	138	0.9618	0.0534	0.9618	0.9807
No log	5.6	140	0.6657	0.2762	0.6657	0.8159
No log	5.68	142	0.6149	0.3814	0.6149	0.7842
No log	5.76	144	0.6074	0.3706	0.6074	0.7793
No log	5.84	146	0.7049	0.3365	0.7049	0.8396
No log	5.92	148	0.8169	0.2340	0.8169	0.9039
No log	6.0	150	0.8127	0.2605	0.8127	0.9015
No log	6.08	152	0.6505	0.3242	0.6505	0.8065
No log	6.16	154	0.6451	0.3744	0.6451	0.8032
No log	6.24	156	0.6600	0.3665	0.6600	0.8124
No log	6.32	158	0.9227	0.2653	0.9227	0.9606
No log	6.4	160	1.1635	0.1367	1.1635	1.0786
No log	6.48	162	1.0618	0.1506	1.0618	1.0304
No log	6.5600	164	0.7300	0.2554	0.7300	0.8544
No log	6.64	166	0.5729	0.4833	0.5729	0.7569
No log	6.72	168	0.6192	0.3917	0.6192	0.7869
No log	6.8	170	0.5654	0.4400	0.5654	0.7519
No log	6.88	172	0.5263	0.3730	0.5263	0.7254
No log	6.96	174	0.6237	0.3131	0.6237	0.7897
No log	7.04	176	0.8629	0.2199	0.8629	0.9289
No log	7.12	178	1.0023	0.1496	1.0023	1.0011
No log	7.2	180	0.9019	0.1799	0.9019	0.9497
No log	7.28	182	0.7158	0.2920	0.7158	0.8460
No log	7.36	184	0.6065	0.4343	0.6065	0.7788
No log	7.44	186	0.6240	0.3744	0.6240	0.7900
No log	7.52	188	0.6135	0.4573	0.6135	0.7833
No log	7.6	190	0.6079	0.3623	0.6079	0.7797
No log	7.68	192	0.6711	0.3433	0.6711	0.8192
No log	7.76	194	0.7080	0.2536	0.7080	0.8415
No log	7.84	196	0.6997	0.2563	0.6997	0.8365
No log	7.92	198	0.6158	0.2917	0.6158	0.7847
No log	8.0	200	0.5602	0.3966	0.5602	0.7484
No log	8.08	202	0.5513	0.4023	0.5513	0.7425
No log	8.16	204	0.5645	0.4348	0.5645	0.7513
No log	8.24	206	0.5873	0.4033	0.5873	0.7663
No log	8.32	208	0.6432	0.3231	0.6432	0.8020
No log	8.4	210	0.7141	0.2744	0.7141	0.8451
No log	8.48	212	0.7859	0.2838	0.7859	0.8865
No log	8.56	214	0.7959	0.2479	0.7959	0.8921
No log	8.64	216	0.7265	0.2727	0.7265	0.8524
No log	8.72	218	0.6449	0.3433	0.6449	0.8031
No log	8.8	220	0.6118	0.4059	0.6118	0.7822
No log	8.88	222	0.6099	0.4	0.6099	0.7810
No log	8.96	224	0.6194	0.4059	0.6194	0.7870
No log	9.04	226	0.6420	0.3585	0.6420	0.8012
No log	9.12	228	0.6750	0.3103	0.6750	0.8216
No log	9.2	230	0.6878	0.3103	0.6878	0.8293
No log	9.28	232	0.6769	0.3103	0.6769	0.8227
No log	9.36	234	0.6866	0.2390	0.6866	0.8286
No log	9.44	236	0.6817	0.2390	0.6817	0.8257
No log	9.52	238	0.6893	0.2381	0.6893	0.8303
No log	9.6	240	0.7171	0.2442	0.7171	0.8468
No log	9.68	242	0.7301	0.25	0.7301	0.8545
No log	9.76	244	0.7333	0.25	0.7333	0.8563
No log	9.84	246	0.7287	0.25	0.7287	0.8537
No log	9.92	248	0.7250	0.2511	0.7250	0.8515
No log	10.0	250	0.7210	0.2511	0.7210	0.8491

Framework versions

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