ArabicNewSplits6_FineTuningAraBERT_run3_AugV5_k6_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: 0.8252
• Qwk: 0.5490
• Mse: 0.8252
• Rmse: 0.9084

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	4.1221	-0.0222	4.1221	2.0303
No log	0.125	4	2.1222	0.0301	2.1222	1.4568
No log	0.1875	6	1.6562	-0.0135	1.6562	1.2870
No log	0.25	8	1.1894	0.0126	1.1894	1.0906
No log	0.3125	10	0.7638	0.0985	0.7638	0.8740
No log	0.375	12	0.7768	-0.0118	0.7768	0.8814
No log	0.4375	14	0.9413	0.0500	0.9413	0.9702
No log	0.5	16	1.3215	0.0296	1.3215	1.1496
No log	0.5625	18	1.3128	0.0296	1.3128	1.1458
No log	0.625	20	0.9333	0.0296	0.9333	0.9661
No log	0.6875	22	0.7658	0.2137	0.7658	0.8751
No log	0.75	24	0.7059	0.2993	0.7059	0.8402
No log	0.8125	26	0.7748	0.2603	0.7748	0.8803
No log	0.875	28	1.0284	0.0133	1.0284	1.0141
No log	0.9375	30	0.9650	0.1284	0.9650	0.9823
No log	1.0	32	1.0168	0.0970	1.0168	1.0084
No log	1.0625	34	1.0722	0.1075	1.0722	1.0355
No log	1.125	36	1.0471	0.1525	1.0471	1.0233
No log	1.1875	38	0.8811	0.2206	0.8811	0.9387
No log	1.25	40	0.8880	0.2279	0.8880	0.9423
No log	1.3125	42	0.8204	0.2278	0.8204	0.9057
No log	1.375	44	0.6378	0.3487	0.6378	0.7986
No log	1.4375	46	0.5988	0.4692	0.5988	0.7738
No log	1.5	48	0.6289	0.3733	0.6289	0.7930
No log	1.5625	50	0.5861	0.4338	0.5861	0.7656
No log	1.625	52	0.6885	0.3683	0.6885	0.8298
No log	1.6875	54	1.3175	0.2155	1.3175	1.1478
No log	1.75	56	1.6094	0.2217	1.6094	1.2686
No log	1.8125	58	1.1333	0.1806	1.1333	1.0646
No log	1.875	60	0.7489	0.3089	0.7489	0.8654
No log	1.9375	62	0.6336	0.4070	0.6336	0.7960
No log	2.0	64	0.6244	0.4153	0.6244	0.7902
No log	2.0625	66	0.7070	0.3285	0.7070	0.8409
No log	2.125	68	0.9178	0.3025	0.9178	0.9580
No log	2.1875	70	0.9249	0.2581	0.9249	0.9617
No log	2.25	72	0.9059	0.3291	0.9059	0.9518
No log	2.3125	74	1.0110	0.2968	1.0110	1.0055
No log	2.375	76	1.2025	0.2458	1.2025	1.0966
No log	2.4375	78	1.2030	0.2561	1.2030	1.0968
No log	2.5	80	0.9473	0.3571	0.9473	0.9733
No log	2.5625	82	0.6787	0.4270	0.6787	0.8238
No log	2.625	84	0.6530	0.4388	0.6530	0.8081
No log	2.6875	86	0.6275	0.5266	0.6275	0.7921
No log	2.75	88	0.6761	0.3748	0.6761	0.8222
No log	2.8125	90	0.8489	0.4017	0.8489	0.9213
No log	2.875	92	0.8982	0.4162	0.8982	0.9477
No log	2.9375	94	0.7892	0.4111	0.7892	0.8884
No log	3.0	96	0.7403	0.4159	0.7403	0.8604
No log	3.0625	98	0.6643	0.4865	0.6643	0.8150
No log	3.125	100	0.6618	0.5456	0.6618	0.8135
No log	3.1875	102	0.7127	0.5274	0.7127	0.8442
No log	3.25	104	0.7472	0.5296	0.7472	0.8644
No log	3.3125	106	0.8100	0.4992	0.8100	0.9000
No log	3.375	108	0.8626	0.4748	0.8626	0.9288
No log	3.4375	110	0.8920	0.4906	0.8920	0.9444
No log	3.5	112	0.9027	0.4814	0.9027	0.9501
No log	3.5625	114	0.9270	0.4801	0.9270	0.9628
No log	3.625	116	1.0144	0.4585	1.0144	1.0072
No log	3.6875	118	1.0606	0.4725	1.0606	1.0298
No log	3.75	120	1.0212	0.4893	1.0212	1.0105
No log	3.8125	122	0.9772	0.4920	0.9772	0.9886
No log	3.875	124	0.9654	0.4826	0.9654	0.9825
No log	3.9375	126	1.0846	0.4368	1.0846	1.0415
No log	4.0	128	1.2125	0.3699	1.2125	1.1011
No log	4.0625	130	1.0627	0.4124	1.0627	1.0309
No log	4.125	132	0.8931	0.4924	0.8931	0.9450
No log	4.1875	134	0.8764	0.5043	0.8764	0.9362
No log	4.25	136	0.8932	0.5257	0.8932	0.9451
No log	4.3125	138	0.9999	0.4505	0.9999	1.0000
No log	4.375	140	1.3704	0.3545	1.3704	1.1706
No log	4.4375	142	1.4235	0.3506	1.4235	1.1931
No log	4.5	144	1.1387	0.3762	1.1387	1.0671
No log	4.5625	146	0.8644	0.5290	0.8644	0.9297
No log	4.625	148	0.7901	0.5240	0.7901	0.8889
No log	4.6875	150	0.7669	0.5736	0.7669	0.8757
No log	4.75	152	0.8304	0.5148	0.8304	0.9112
No log	4.8125	154	1.0202	0.3878	1.0202	1.0101
No log	4.875	156	1.1297	0.3564	1.1297	1.0629
No log	4.9375	158	1.0324	0.3837	1.0324	1.0161
No log	5.0	160	0.8050	0.4057	0.8050	0.8972
No log	5.0625	162	0.6335	0.4881	0.6335	0.7959
No log	5.125	164	0.6360	0.5148	0.6360	0.7975
No log	5.1875	166	0.6998	0.4751	0.6998	0.8365
No log	5.25	168	0.7068	0.4809	0.7068	0.8407
No log	5.3125	170	0.6675	0.5217	0.6675	0.8170
No log	5.375	172	0.6713	0.5723	0.6713	0.8194
No log	5.4375	174	0.7573	0.4746	0.7573	0.8702
No log	5.5	176	0.9103	0.4469	0.9103	0.9541
No log	5.5625	178	1.0183	0.4524	1.0183	1.0091
No log	5.625	180	1.0570	0.4036	1.0570	1.0281
No log	5.6875	182	0.9717	0.4422	0.9717	0.9858
No log	5.75	184	0.9281	0.5331	0.9281	0.9634
No log	5.8125	186	0.9966	0.4831	0.9966	0.9983
No log	5.875	188	1.0830	0.4818	1.0830	1.0407
No log	5.9375	190	1.0837	0.4843	1.0837	1.0410
No log	6.0	192	1.0456	0.4613	1.0456	1.0226
No log	6.0625	194	1.0275	0.4802	1.0275	1.0136
No log	6.125	196	1.0037	0.5061	1.0037	1.0018
No log	6.1875	198	0.9708	0.4926	0.9708	0.9853
No log	6.25	200	0.9186	0.5016	0.9186	0.9584
No log	6.3125	202	0.8557	0.5479	0.8557	0.9250
No log	6.375	204	0.8530	0.5479	0.8530	0.9236
No log	6.4375	206	0.8771	0.5531	0.8771	0.9366
No log	6.5	208	0.9064	0.5152	0.9064	0.9521
No log	6.5625	210	0.8853	0.5639	0.8853	0.9409
No log	6.625	212	0.8600	0.5279	0.8600	0.9273
No log	6.6875	214	0.8636	0.5328	0.8636	0.9293
No log	6.75	216	0.8616	0.5477	0.8616	0.9282
No log	6.8125	218	0.8439	0.4964	0.8439	0.9187
No log	6.875	220	0.8324	0.5108	0.8324	0.9124
No log	6.9375	222	0.8686	0.5185	0.8686	0.9320
No log	7.0	224	0.8840	0.5099	0.8840	0.9402
No log	7.0625	226	0.8439	0.5029	0.8439	0.9186
No log	7.125	228	0.7912	0.5129	0.7912	0.8895
No log	7.1875	230	0.7456	0.5040	0.7456	0.8635
No log	7.25	232	0.7336	0.5336	0.7336	0.8565
No log	7.3125	234	0.7334	0.5336	0.7334	0.8564
No log	7.375	236	0.7418	0.5159	0.7418	0.8613
No log	7.4375	238	0.7663	0.5043	0.7663	0.8754
No log	7.5	240	0.8072	0.5154	0.8072	0.8985
No log	7.5625	242	0.8307	0.4946	0.8307	0.9114
No log	7.625	244	0.8279	0.5224	0.8279	0.9099
No log	7.6875	246	0.8171	0.5066	0.8171	0.9039
No log	7.75	248	0.8017	0.4778	0.8017	0.8954
No log	7.8125	250	0.8080	0.4824	0.8080	0.8989
No log	7.875	252	0.8428	0.5346	0.8428	0.9180
No log	7.9375	254	0.8690	0.5361	0.8690	0.9322
No log	8.0	256	0.8711	0.5361	0.8711	0.9333
No log	8.0625	258	0.8786	0.5366	0.8786	0.9373
No log	8.125	260	0.8849	0.5366	0.8849	0.9407
No log	8.1875	262	0.8830	0.5366	0.8830	0.9397
No log	8.25	264	0.8636	0.5161	0.8636	0.9293
No log	8.3125	266	0.8625	0.5161	0.8625	0.9287
No log	8.375	268	0.8816	0.5366	0.8816	0.9389
No log	8.4375	270	0.8936	0.5377	0.8936	0.9453
No log	8.5	272	0.8920	0.5355	0.8920	0.9445
No log	8.5625	274	0.8917	0.5355	0.8917	0.9443
No log	8.625	276	0.8990	0.5327	0.8990	0.9481
No log	8.6875	278	0.8975	0.5287	0.8975	0.9474
No log	8.75	280	0.8884	0.5287	0.8884	0.9426
No log	8.8125	282	0.8677	0.5338	0.8677	0.9315
No log	8.875	284	0.8493	0.5490	0.8493	0.9216
No log	8.9375	286	0.8482	0.5503	0.8482	0.9210
No log	9.0	288	0.8630	0.5182	0.8630	0.9290
No log	9.0625	290	0.8782	0.5140	0.8782	0.9371
No log	9.125	292	0.8791	0.5140	0.8791	0.9376
No log	9.1875	294	0.8883	0.5146	0.8883	0.9425
No log	9.25	296	0.8826	0.5146	0.8826	0.9395
No log	9.3125	298	0.8669	0.5140	0.8669	0.9311
No log	9.375	300	0.8449	0.5074	0.8449	0.9192
No log	9.4375	302	0.8248	0.5451	0.8248	0.9082
No log	9.5	304	0.8185	0.5503	0.8185	0.9047
No log	9.5625	306	0.8205	0.5503	0.8205	0.9058
No log	9.625	308	0.8250	0.5503	0.8250	0.9083
No log	9.6875	310	0.8272	0.5503	0.8272	0.9095
No log	9.75	312	0.8296	0.5503	0.8296	0.9108
No log	9.8125	314	0.8287	0.5503	0.8287	0.9103
No log	9.875	316	0.8261	0.5490	0.8261	0.9089
No log	9.9375	318	0.8253	0.5490	0.8253	0.9085
No log	10.0	320	0.8252	0.5490	0.8252	0.9084

Framework versions

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