ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run2_AugV5_k4_task2_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: 1.0043
• Qwk: 0.4610
• Mse: 1.0043
• Rmse: 1.0021

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	4.2326	-0.0051	4.2326	2.0573
No log	0.16	4	2.1027	0.0578	2.1027	1.4501
No log	0.24	6	1.3092	0.1474	1.3092	1.1442
No log	0.32	8	0.8169	0.1395	0.8169	0.9038
No log	0.4	10	0.6840	0.1882	0.6840	0.8270
No log	0.48	12	0.7687	0.1028	0.7687	0.8768
No log	0.56	14	0.8230	0.0918	0.8230	0.9072
No log	0.64	16	0.7966	0.1447	0.7966	0.8925
No log	0.72	18	0.6775	0.1241	0.6775	0.8231
No log	0.8	20	0.7279	0.2901	0.7279	0.8532
No log	0.88	22	0.7059	0.3030	0.7059	0.8402
No log	0.96	24	0.6909	0.3342	0.6909	0.8312
No log	1.04	26	0.6091	0.375	0.6091	0.7804
No log	1.12	28	0.6099	0.3195	0.6099	0.7810
No log	1.2	30	0.5851	0.3394	0.5851	0.7649
No log	1.28	32	0.5610	0.4556	0.5610	0.7490
No log	1.3600	34	0.5654	0.4431	0.5654	0.7519
No log	1.44	36	0.6382	0.4827	0.6382	0.7989
No log	1.52	38	0.8306	0.3981	0.8306	0.9114
No log	1.6	40	0.8404	0.4247	0.8404	0.9167
No log	1.6800	42	0.6982	0.5190	0.6982	0.8356
No log	1.76	44	1.0030	0.3898	1.0030	1.0015
No log	1.8400	46	1.1895	0.2890	1.1895	1.0906
No log	1.92	48	0.9583	0.4380	0.9583	0.9789
No log	2.0	50	0.7778	0.4460	0.7778	0.8819
No log	2.08	52	0.7495	0.4987	0.7495	0.8657
No log	2.16	54	0.7300	0.5051	0.7300	0.8544
No log	2.24	56	0.7198	0.4906	0.7198	0.8484
No log	2.32	58	0.7395	0.4983	0.7395	0.8599
No log	2.4	60	0.7321	0.4954	0.7321	0.8556
No log	2.48	62	0.7379	0.5039	0.7379	0.8590
No log	2.56	64	0.7942	0.4992	0.7942	0.8912
No log	2.64	66	1.0320	0.4592	1.0320	1.0159
No log	2.7200	68	0.9575	0.4598	0.9575	0.9785
No log	2.8	70	0.8996	0.4935	0.8996	0.9485
No log	2.88	72	0.9453	0.4856	0.9453	0.9723
No log	2.96	74	0.9263	0.4805	0.9263	0.9625
No log	3.04	76	0.9090	0.5104	0.9090	0.9534
No log	3.12	78	0.9228	0.4685	0.9228	0.9606
No log	3.2	80	0.8833	0.5124	0.8833	0.9398
No log	3.2800	82	0.8541	0.4971	0.8541	0.9242
No log	3.36	84	0.8429	0.5147	0.8429	0.9181
No log	3.44	86	0.8452	0.5286	0.8452	0.9194
No log	3.52	88	0.7993	0.5357	0.7993	0.8940
No log	3.6	90	0.8277	0.4411	0.8277	0.9098
No log	3.68	92	0.8156	0.4350	0.8156	0.9031
No log	3.76	94	0.8137	0.4842	0.8137	0.9021
No log	3.84	96	0.8645	0.5090	0.8645	0.9298
No log	3.92	98	0.8947	0.4964	0.8947	0.9459
No log	4.0	100	0.9527	0.4453	0.9527	0.9760
No log	4.08	102	1.0212	0.4228	1.0212	1.0106
No log	4.16	104	1.0004	0.4614	1.0004	1.0002
No log	4.24	106	0.9908	0.4753	0.9908	0.9954
No log	4.32	108	0.9804	0.4753	0.9804	0.9902
No log	4.4	110	0.9759	0.4753	0.9759	0.9879
No log	4.48	112	0.9476	0.4478	0.9476	0.9734
No log	4.5600	114	0.9493	0.4887	0.9493	0.9743
No log	4.64	116	1.0895	0.4361	1.0895	1.0438
No log	4.72	118	1.1371	0.3963	1.1371	1.0663
No log	4.8	120	0.9869	0.4650	0.9869	0.9934
No log	4.88	122	0.8731	0.4927	0.8731	0.9344
No log	4.96	124	0.8623	0.4787	0.8623	0.9286
No log	5.04	126	0.8562	0.4860	0.8562	0.9253
No log	5.12	128	0.8523	0.5092	0.8523	0.9232
No log	5.2	130	0.8657	0.5353	0.8657	0.9304
No log	5.28	132	0.9492	0.4630	0.9492	0.9743
No log	5.36	134	1.0265	0.4313	1.0265	1.0132
No log	5.44	136	0.9505	0.4532	0.9505	0.9749
No log	5.52	138	0.8281	0.5224	0.8281	0.9100
No log	5.6	140	0.8200	0.4379	0.8200	0.9055
No log	5.68	142	0.8292	0.4343	0.8292	0.9106
No log	5.76	144	0.8289	0.4796	0.8289	0.9104
No log	5.84	146	0.9124	0.4748	0.9124	0.9552
No log	5.92	148	1.1171	0.4113	1.1171	1.0569
No log	6.0	150	1.1839	0.3846	1.1839	1.0880
No log	6.08	152	1.0839	0.4657	1.0839	1.0411
No log	6.16	154	1.0077	0.4567	1.0077	1.0038
No log	6.24	156	1.0301	0.4435	1.0301	1.0149
No log	6.32	158	1.0606	0.4422	1.0606	1.0298
No log	6.4	160	1.0564	0.4496	1.0564	1.0278
No log	6.48	162	1.0266	0.4554	1.0266	1.0132
No log	6.5600	164	1.0117	0.4599	1.0117	1.0059
No log	6.64	166	0.9940	0.4478	0.9940	0.9970
No log	6.72	168	0.9740	0.4416	0.9740	0.9869
No log	6.8	170	0.9512	0.4543	0.9512	0.9753
No log	6.88	172	0.9393	0.4699	0.9393	0.9692
No log	6.96	174	0.9273	0.4648	0.9273	0.9630
No log	7.04	176	0.9109	0.4620	0.9109	0.9544
No log	7.12	178	0.9425	0.4635	0.9425	0.9708
No log	7.2	180	0.9776	0.4239	0.9776	0.9887
No log	7.28	182	0.9689	0.4516	0.9689	0.9843
No log	7.36	184	0.9792	0.4727	0.9792	0.9895
No log	7.44	186	1.0011	0.4482	1.0011	1.0005
No log	7.52	188	1.0194	0.4250	1.0194	1.0097
No log	7.6	190	1.0323	0.4299	1.0323	1.0160
No log	7.68	192	1.0453	0.4487	1.0453	1.0224
No log	7.76	194	1.0526	0.4299	1.0526	1.0260
No log	7.84	196	1.0696	0.4281	1.0696	1.0342
No log	7.92	198	1.0846	0.3983	1.0846	1.0414
No log	8.0	200	1.0866	0.4087	1.0866	1.0424
No log	8.08	202	1.0820	0.4281	1.0820	1.0402
No log	8.16	204	1.0864	0.4369	1.0864	1.0423
No log	8.24	206	1.0972	0.4193	1.0972	1.0475
No log	8.32	208	1.0999	0.4262	1.0999	1.0487
No log	8.4	210	1.0891	0.4191	1.0891	1.0436
No log	8.48	212	1.0891	0.4235	1.0891	1.0436
No log	8.56	214	1.0922	0.3932	1.0922	1.0451
No log	8.64	216	1.1021	0.3976	1.1021	1.0498
No log	8.72	218	1.0865	0.3932	1.0865	1.0424
No log	8.8	220	1.0708	0.4368	1.0708	1.0348
No log	8.88	222	1.0635	0.4190	1.0635	1.0313
No log	8.96	224	1.0560	0.4190	1.0560	1.0276
No log	9.04	226	1.0498	0.4097	1.0498	1.0246
No log	9.12	228	1.0376	0.4097	1.0376	1.0186
No log	9.2	230	1.0254	0.4094	1.0254	1.0126
No log	9.28	232	1.0185	0.4171	1.0185	1.0092
No log	9.36	234	1.0148	0.4343	1.0148	1.0074
No log	9.44	236	1.0106	0.4641	1.0106	1.0053
No log	9.52	238	1.0094	0.4550	1.0094	1.0047
No log	9.6	240	1.0090	0.4311	1.0090	1.0045
No log	9.68	242	1.0075	0.4311	1.0075	1.0037
No log	9.76	244	1.0054	0.4437	1.0054	1.0027
No log	9.84	246	1.0037	0.4610	1.0037	1.0018
No log	9.92	248	1.0039	0.4610	1.0039	1.0020
No log	10.0	250	1.0043	0.4610	1.0043	1.0021

Framework versions

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