ArabicNewSplits4_WithDuplicationsForScore5_FineTuningAraBERT_run1_AugV5_k7_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: 0.6017
• Qwk: 0.3706
• Mse: 0.6017
• Rmse: 0.7757

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	0.0625	2	3.4620	-0.0066	3.4620	1.8606
No log	0.125	4	1.8841	-0.0101	1.8841	1.3726
No log	0.1875	6	1.4848	0.0255	1.4848	1.2185
No log	0.25	8	1.2536	0.0632	1.2536	1.1197
No log	0.3125	10	0.6960	0.0058	0.6960	0.8343
No log	0.375	12	0.7420	-0.0595	0.7420	0.8614
No log	0.4375	14	0.6481	0.0	0.6481	0.8050
No log	0.5	16	0.7361	-0.1429	0.7361	0.8580
No log	0.5625	18	0.8658	0.0609	0.8658	0.9305
No log	0.625	20	0.6212	0.0145	0.6212	0.7882
No log	0.6875	22	0.6185	0.0569	0.6185	0.7865
No log	0.75	24	0.7106	0.0	0.7106	0.8430
No log	0.8125	26	0.7430	0.0	0.7430	0.8620
No log	0.875	28	0.5899	0.0388	0.5899	0.7681
No log	0.9375	30	0.8958	0.1111	0.8958	0.9464
No log	1.0	32	1.1842	0.0977	1.1842	1.0882
No log	1.0625	34	0.8141	0.0476	0.8141	0.9023
No log	1.125	36	0.6261	0.0222	0.6261	0.7913
No log	1.1875	38	0.5969	0.0303	0.5969	0.7726
No log	1.25	40	0.5678	0.0569	0.5678	0.7535
No log	1.3125	42	0.5813	0.0476	0.5813	0.7624
No log	1.375	44	0.5959	0.0365	0.5959	0.7720
No log	1.4375	46	0.6171	0.0199	0.6171	0.7856
No log	1.5	48	0.7946	0.1456	0.7946	0.8914
No log	1.5625	50	0.7107	0.1917	0.7107	0.8430
No log	1.625	52	0.8208	0.1443	0.8208	0.9060
No log	1.6875	54	1.0163	0.1318	1.0163	1.0081
No log	1.75	56	0.6687	0.1910	0.6687	0.8178
No log	1.8125	58	1.7561	0.0435	1.7561	1.3252
No log	1.875	60	2.4596	0.0606	2.4596	1.5683
No log	1.9375	62	1.7325	0.0404	1.7325	1.3162
No log	2.0	64	0.7525	0.1619	0.7525	0.8675
No log	2.0625	66	0.5613	0.2644	0.5613	0.7492
No log	2.125	68	0.6354	0.1364	0.6354	0.7971
No log	2.1875	70	0.5651	0.3548	0.5651	0.7517
No log	2.25	72	0.5880	0.2707	0.5880	0.7668
No log	2.3125	74	0.7195	0.1304	0.7195	0.8482
No log	2.375	76	0.7730	0.2072	0.7730	0.8792
No log	2.4375	78	0.7711	0.2793	0.7711	0.8781
No log	2.5	80	1.2529	0.1399	1.2529	1.1193
No log	2.5625	82	1.0087	0.2062	1.0087	1.0043
No log	2.625	84	0.8052	0.2542	0.8052	0.8973
No log	2.6875	86	0.7443	0.1610	0.7443	0.8627
No log	2.75	88	0.6711	0.2513	0.6711	0.8192
No log	2.8125	90	0.6815	0.3237	0.6815	0.8255
No log	2.875	92	1.0605	0.1777	1.0605	1.0298
No log	2.9375	94	0.8256	0.2327	0.8256	0.9086
No log	3.0	96	0.6481	0.4404	0.6481	0.8050
No log	3.0625	98	0.6098	0.3161	0.6098	0.7809
No log	3.125	100	0.6272	0.3258	0.6272	0.7920
No log	3.1875	102	0.6736	0.3398	0.6736	0.8207
No log	3.25	104	0.6921	0.3365	0.6921	0.8319
No log	3.3125	106	0.8167	0.2126	0.8167	0.9037
No log	3.375	108	0.7755	0.3016	0.7755	0.8806
No log	3.4375	110	0.8457	0.2558	0.8457	0.9196
No log	3.5	112	0.9226	0.3114	0.9226	0.9605
No log	3.5625	114	0.7673	0.3722	0.7673	0.8760
No log	3.625	116	0.8138	0.2681	0.8138	0.9021
No log	3.6875	118	0.8883	0.2199	0.8883	0.9425
No log	3.75	120	0.7123	0.3818	0.7123	0.8440
No log	3.8125	122	0.6586	0.3892	0.6586	0.8116
No log	3.875	124	0.6793	0.2965	0.6793	0.8242
No log	3.9375	126	0.6590	0.3498	0.6590	0.8118
No log	4.0	128	0.7029	0.2621	0.7029	0.8384
No log	4.0625	130	0.6505	0.3267	0.6505	0.8066
No log	4.125	132	0.6288	0.2653	0.6288	0.7930
No log	4.1875	134	0.6368	0.3398	0.6368	0.7980
No log	4.25	136	0.9549	0.2060	0.9549	0.9772
No log	4.3125	138	0.7973	0.3422	0.7973	0.8929
No log	4.375	140	0.5904	0.4	0.5904	0.7684
No log	4.4375	142	0.6039	0.3061	0.6039	0.7771
No log	4.5	144	0.7558	0.3422	0.7558	0.8694
No log	4.5625	146	0.7170	0.3722	0.7170	0.8468
No log	4.625	148	0.5846	0.3769	0.5846	0.7646
No log	4.6875	150	0.6078	0.2990	0.6078	0.7796
No log	4.75	152	0.5609	0.4400	0.5609	0.7490
No log	4.8125	154	0.8169	0.3362	0.8169	0.9038
No log	4.875	156	0.8182	0.3633	0.8182	0.9045
No log	4.9375	158	0.5741	0.4346	0.5741	0.7577
No log	5.0	160	0.6008	0.2965	0.6008	0.7751
No log	5.0625	162	0.7115	0.3091	0.7115	0.8435
No log	5.125	164	0.5820	0.3267	0.5820	0.7629
No log	5.1875	166	0.5582	0.3862	0.5582	0.7471
No log	5.25	168	0.6242	0.3645	0.6242	0.7901
No log	5.3125	170	0.6837	0.3704	0.6837	0.8269
No log	5.375	172	0.6743	0.3365	0.6743	0.8212
No log	5.4375	174	0.6822	0.3365	0.6822	0.8260
No log	5.5	176	0.8696	0.2741	0.8696	0.9325
No log	5.5625	178	1.1389	0.1742	1.1389	1.0672
No log	5.625	180	0.9605	0.2768	0.9605	0.9801
No log	5.6875	182	0.7101	0.3301	0.7101	0.8427
No log	5.75	184	0.6750	0.3208	0.6750	0.8216
No log	5.8125	186	0.7052	0.3398	0.7052	0.8398
No log	5.875	188	0.7135	0.3303	0.7135	0.8447
No log	5.9375	190	0.6863	0.3427	0.6863	0.8284
No log	6.0	192	0.5857	0.4167	0.5857	0.7653
No log	6.0625	194	0.5784	0.3617	0.5784	0.7605
No log	6.125	196	0.5843	0.3862	0.5843	0.7644
No log	6.1875	198	0.6893	0.3524	0.6893	0.8303
No log	6.25	200	0.6460	0.3462	0.6460	0.8037
No log	6.3125	202	0.6022	0.3878	0.6022	0.7760
No log	6.375	204	0.7082	0.3274	0.7082	0.8415
No log	6.4375	206	0.7276	0.3274	0.7276	0.8530
No log	6.5	208	0.6197	0.3469	0.6197	0.7872
No log	6.5625	210	0.6077	0.3365	0.6077	0.7796
No log	6.625	212	0.6069	0.3704	0.6069	0.7790
No log	6.6875	214	0.6433	0.3684	0.6433	0.8021
No log	6.75	216	0.7720	0.3030	0.7720	0.8786
No log	6.8125	218	0.7748	0.3030	0.7748	0.8802
No log	6.875	220	0.6859	0.3365	0.6859	0.8282
No log	6.9375	222	0.6457	0.3524	0.6457	0.8036
No log	7.0	224	0.6615	0.3271	0.6615	0.8133
No log	7.0625	226	0.7183	0.2920	0.7183	0.8475
No log	7.125	228	0.6623	0.3365	0.6623	0.8138
No log	7.1875	230	0.6037	0.3402	0.6037	0.7770
No log	7.25	232	0.5806	0.3878	0.5806	0.7620
No log	7.3125	234	0.5812	0.3878	0.5812	0.7624
No log	7.375	236	0.6508	0.3398	0.6508	0.8067
No log	7.4375	238	0.7187	0.2961	0.7187	0.8478
No log	7.5	240	0.7749	0.2960	0.7749	0.8803
No log	7.5625	242	0.7720	0.2653	0.7720	0.8787
No log	7.625	244	0.6672	0.3744	0.6672	0.8168
No log	7.6875	246	0.5979	0.3706	0.5979	0.7732
No log	7.75	248	0.5829	0.3978	0.5829	0.7634
No log	7.8125	250	0.5839	0.3978	0.5839	0.7641
No log	7.875	252	0.5815	0.3978	0.5815	0.7626
No log	7.9375	254	0.5797	0.36	0.5797	0.7614
No log	8.0	256	0.5901	0.3508	0.5901	0.7682
No log	8.0625	258	0.6324	0.3469	0.6324	0.7952
No log	8.125	260	0.6538	0.3365	0.6538	0.8086
No log	8.1875	262	0.7008	0.4128	0.7008	0.8371
No log	8.25	264	0.6940	0.3665	0.6940	0.8331
No log	8.3125	266	0.6555	0.3365	0.6555	0.8096
No log	8.375	268	0.6317	0.3725	0.6317	0.7948
No log	8.4375	270	0.6115	0.36	0.6115	0.7820
No log	8.5	272	0.6130	0.36	0.6130	0.7830
No log	8.5625	274	0.6270	0.3769	0.6270	0.7918
No log	8.625	276	0.6589	0.3645	0.6589	0.8117
No log	8.6875	278	0.6501	0.3684	0.6501	0.8063
No log	8.75	280	0.6222	0.3561	0.6222	0.7888
No log	8.8125	282	0.6145	0.36	0.6145	0.7839
No log	8.875	284	0.6122	0.3398	0.6122	0.7825
No log	8.9375	286	0.6153	0.3498	0.6153	0.7844
No log	9.0	288	0.6308	0.3663	0.6308	0.7942
No log	9.0625	290	0.6635	0.3645	0.6635	0.8145
No log	9.125	292	0.6740	0.3704	0.6740	0.8209
No log	9.1875	294	0.6713	0.3704	0.6713	0.8193
No log	9.25	296	0.6494	0.3684	0.6494	0.8058
No log	9.3125	298	0.6277	0.3769	0.6277	0.7923
No log	9.375	300	0.6064	0.3706	0.6064	0.7787
No log	9.4375	302	0.5969	0.36	0.5969	0.7726
No log	9.5	304	0.5963	0.36	0.5963	0.7722
No log	9.5625	306	0.5951	0.36	0.5951	0.7714
No log	9.625	308	0.5971	0.36	0.5971	0.7728
No log	9.6875	310	0.5989	0.3706	0.5989	0.7739
No log	9.75	312	0.5995	0.3706	0.5995	0.7742
No log	9.8125	314	0.5998	0.3706	0.5998	0.7745
No log	9.875	316	0.6007	0.3706	0.6007	0.7750
No log	9.9375	318	0.6013	0.3706	0.6013	0.7755
No log	10.0	320	0.6017	0.3706	0.6017	0.7757

Framework versions

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