ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run1_AugV5_k5_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: 0.7598
• Qwk: 0.6971
• Mse: 0.7598
• Rmse: 0.8717

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	0.0667	2	5.1336	0.0056	5.1336	2.2657
No log	0.1333	4	3.4338	0.0724	3.4338	1.8531
No log	0.2	6	1.9484	0.1341	1.9484	1.3959
No log	0.2667	8	1.2691	0.2462	1.2691	1.1265
No log	0.3333	10	1.1206	0.2222	1.1206	1.0586
No log	0.4	12	1.1712	0.1896	1.1712	1.0822
No log	0.4667	14	1.1020	0.3582	1.1020	1.0498
No log	0.5333	16	1.0843	0.2242	1.0843	1.0413
No log	0.6	18	1.0970	0.2406	1.0970	1.0474
No log	0.6667	20	1.0150	0.2858	1.0150	1.0075
No log	0.7333	22	1.0285	0.2879	1.0285	1.0141
No log	0.8	24	1.0063	0.2604	1.0063	1.0032
No log	0.8667	26	0.9520	0.3434	0.9520	0.9757
No log	0.9333	28	0.9063	0.3928	0.9063	0.9520
No log	1.0	30	0.8739	0.4154	0.8739	0.9348
No log	1.0667	32	0.8440	0.5015	0.8440	0.9187
No log	1.1333	34	0.8192	0.5226	0.8192	0.9051
No log	1.2	36	0.7797	0.5547	0.7797	0.8830
No log	1.2667	38	0.7867	0.5497	0.7867	0.8870
No log	1.3333	40	0.7574	0.5910	0.7574	0.8703
No log	1.4	42	0.7191	0.6583	0.7191	0.8480
No log	1.4667	44	0.6939	0.6570	0.6939	0.8330
No log	1.5333	46	0.8509	0.6129	0.8509	0.9224
No log	1.6	48	0.8310	0.6339	0.8310	0.9116
No log	1.6667	50	0.6780	0.6659	0.6780	0.8234
No log	1.7333	52	0.6540	0.6357	0.6540	0.8087
No log	1.8	54	0.6924	0.5936	0.6924	0.8321
No log	1.8667	56	0.6230	0.6463	0.6230	0.7893
No log	1.9333	58	0.7225	0.6747	0.7225	0.8500
No log	2.0	60	1.0352	0.4963	1.0352	1.0174
No log	2.0667	62	1.0943	0.4425	1.0943	1.0461
No log	2.1333	64	0.8059	0.6084	0.8059	0.8977
No log	2.2	66	0.7112	0.6492	0.7112	0.8433
No log	2.2667	68	0.6742	0.6388	0.6742	0.8211
No log	2.3333	70	0.6633	0.6770	0.6633	0.8144
No log	2.4	72	0.6605	0.6674	0.6605	0.8127
No log	2.4667	74	0.6606	0.6498	0.6606	0.8127
No log	2.5333	76	0.7160	0.6106	0.7160	0.8462
No log	2.6	78	0.7167	0.6100	0.7167	0.8466
No log	2.6667	80	0.6774	0.6242	0.6774	0.8231
No log	2.7333	82	0.7566	0.6651	0.7566	0.8698
No log	2.8	84	0.9281	0.6042	0.9281	0.9634
No log	2.8667	86	1.0114	0.6464	1.0114	1.0057
No log	2.9333	88	0.8950	0.5833	0.8950	0.9461
No log	3.0	90	0.8995	0.6369	0.8995	0.9484
No log	3.0667	92	1.1863	0.4858	1.1863	1.0892
No log	3.1333	94	1.1861	0.4798	1.1861	1.0891
No log	3.2	96	0.9166	0.6298	0.9166	0.9574
No log	3.2667	98	0.8953	0.7023	0.8953	0.9462
No log	3.3333	100	1.0282	0.6432	1.0282	1.0140
No log	3.4	102	1.0183	0.6497	1.0183	1.0091
No log	3.4667	104	0.9233	0.6719	0.9233	0.9609
No log	3.5333	106	0.8602	0.6585	0.8602	0.9275
No log	3.6	108	0.7687	0.6638	0.7687	0.8768
No log	3.6667	110	0.7504	0.6593	0.7504	0.8662
No log	3.7333	112	0.7540	0.6632	0.7540	0.8683
No log	3.8	114	0.7813	0.6828	0.7813	0.8839
No log	3.8667	116	0.7749	0.6676	0.7749	0.8803
No log	3.9333	118	0.7682	0.6604	0.7682	0.8765
No log	4.0	120	0.7625	0.6774	0.7625	0.8732
No log	4.0667	122	0.7648	0.6710	0.7648	0.8745
No log	4.1333	124	0.7921	0.6800	0.7921	0.8900
No log	4.2	126	0.8575	0.6657	0.8575	0.9260
No log	4.2667	128	0.8397	0.6958	0.8397	0.9164
No log	4.3333	130	0.7525	0.7018	0.7525	0.8675
No log	4.4	132	0.6825	0.7074	0.6825	0.8261
No log	4.4667	134	0.6760	0.6859	0.6760	0.8222
No log	4.5333	136	0.7025	0.6752	0.7025	0.8382
No log	4.6	138	0.6992	0.6759	0.6992	0.8362
No log	4.6667	140	0.6758	0.7009	0.6758	0.8221
No log	4.7333	142	0.6843	0.7040	0.6843	0.8272
No log	4.8	144	0.7270	0.7221	0.7270	0.8526
No log	4.8667	146	0.8029	0.7054	0.8029	0.8960
No log	4.9333	148	0.8173	0.7294	0.8173	0.9041
No log	5.0	150	0.8203	0.7145	0.8203	0.9057
No log	5.0667	152	0.7505	0.7106	0.7505	0.8663
No log	5.1333	154	0.7354	0.7187	0.7354	0.8575
No log	5.2	156	0.7047	0.7199	0.7047	0.8395
No log	5.2667	158	0.6608	0.7336	0.6608	0.8129
No log	5.3333	160	0.6206	0.7261	0.6206	0.7878
No log	5.4	162	0.6195	0.7245	0.6195	0.7871
No log	5.4667	164	0.6214	0.7200	0.6214	0.7883
No log	5.5333	166	0.6388	0.7203	0.6388	0.7993
No log	5.6	168	0.7030	0.7380	0.7030	0.8384
No log	5.6667	170	0.7324	0.7199	0.7324	0.8558
No log	5.7333	172	0.7476	0.7115	0.7476	0.8646
No log	5.8	174	0.7529	0.6953	0.7529	0.8677
No log	5.8667	176	0.7686	0.6793	0.7686	0.8767
No log	5.9333	178	0.7954	0.7055	0.7954	0.8918
No log	6.0	180	0.8382	0.7027	0.8382	0.9155
No log	6.0667	182	0.8640	0.7033	0.8640	0.9295
No log	6.1333	184	0.8707	0.6481	0.8707	0.9331
No log	6.2	186	0.7912	0.6859	0.7912	0.8895
No log	6.2667	188	0.7487	0.6833	0.7487	0.8653
No log	6.3333	190	0.7754	0.6789	0.7754	0.8805
No log	6.4	192	0.7906	0.6783	0.7906	0.8892
No log	6.4667	194	0.8238	0.6597	0.8238	0.9076
No log	6.5333	196	0.8581	0.6637	0.8581	0.9264
No log	6.6	198	0.8252	0.6927	0.8252	0.9084
No log	6.6667	200	0.7941	0.6984	0.7941	0.8911
No log	6.7333	202	0.7987	0.6902	0.7987	0.8937
No log	6.8	204	0.8160	0.7034	0.8160	0.9033
No log	6.8667	206	0.8187	0.7029	0.8187	0.9048
No log	6.9333	208	0.8182	0.7110	0.8182	0.9045
No log	7.0	210	0.8063	0.7092	0.8063	0.8979
No log	7.0667	212	0.7686	0.7372	0.7686	0.8767
No log	7.1333	214	0.7494	0.7372	0.7494	0.8657
No log	7.2	216	0.7397	0.7094	0.7397	0.8601
No log	7.2667	218	0.7414	0.7094	0.7414	0.8610
No log	7.3333	220	0.7585	0.7168	0.7585	0.8709
No log	7.4	222	0.7791	0.7172	0.7791	0.8827
No log	7.4667	224	0.8082	0.7161	0.8082	0.8990
No log	7.5333	226	0.8058	0.7102	0.8058	0.8977
No log	7.6	228	0.7907	0.7099	0.7907	0.8892
No log	7.6667	230	0.7989	0.6954	0.7989	0.8938
No log	7.7333	232	0.7981	0.6982	0.7981	0.8934
No log	7.8	234	0.7931	0.7033	0.7931	0.8905
No log	7.8667	236	0.7832	0.7017	0.7832	0.8850
No log	7.9333	238	0.7677	0.7021	0.7677	0.8762
No log	8.0	240	0.7644	0.7064	0.7644	0.8743
No log	8.0667	242	0.7807	0.7137	0.7807	0.8836
No log	8.1333	244	0.8096	0.7129	0.8096	0.8998
No log	8.2	246	0.8502	0.7123	0.8502	0.9221
No log	8.2667	248	0.8713	0.6974	0.8713	0.9334
No log	8.3333	250	0.8440	0.7123	0.8440	0.9187
No log	8.4	252	0.8047	0.7167	0.8047	0.8971
No log	8.4667	254	0.7688	0.7122	0.7688	0.8768
No log	8.5333	256	0.7493	0.7041	0.7493	0.8656
No log	8.6	258	0.7394	0.7066	0.7394	0.8599
No log	8.6667	260	0.7376	0.7041	0.7376	0.8589
No log	8.7333	262	0.7511	0.6989	0.7511	0.8667
No log	8.8	264	0.7551	0.7077	0.7551	0.8690
No log	8.8667	266	0.7510	0.6928	0.7510	0.8666
No log	8.9333	268	0.7508	0.6896	0.7508	0.8665
No log	9.0	270	0.7444	0.6637	0.7444	0.8628
No log	9.0667	272	0.7331	0.6903	0.7331	0.8562
No log	9.1333	274	0.7230	0.6848	0.7230	0.8503
No log	9.2	276	0.7233	0.6880	0.7233	0.8504
No log	9.2667	278	0.7313	0.6941	0.7313	0.8552
No log	9.3333	280	0.7379	0.6941	0.7379	0.8590
No log	9.4	282	0.7417	0.6941	0.7417	0.8612
No log	9.4667	284	0.7459	0.6941	0.7459	0.8637
No log	9.5333	286	0.7531	0.6959	0.7531	0.8678
No log	9.6	288	0.7620	0.6850	0.7620	0.8729
No log	9.6667	290	0.7665	0.6843	0.7665	0.8755
No log	9.7333	292	0.7650	0.6886	0.7650	0.8747
No log	9.8	294	0.7630	0.6886	0.7630	0.8735
No log	9.8667	296	0.7606	0.6905	0.7606	0.8721
No log	9.9333	298	0.7595	0.6971	0.7595	0.8715
No log	10.0	300	0.7598	0.6971	0.7598	0.8717

Framework versions

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