ArabicNewSplits7_FineTuningAraBERT_run3_AugV5_k20_task7_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: 0.6566
• Qwk: 0.4890
• Mse: 0.6566
• Rmse: 0.8103

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: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	0.04	2	2.5990	-0.0262	2.5990	1.6121
No log	0.08	4	1.4044	0.0540	1.4044	1.1851
No log	0.12	6	0.9350	-0.0228	0.9350	0.9670
No log	0.16	8	0.9408	-0.0860	0.9408	0.9700
No log	0.2	10	0.8959	-0.0079	0.8959	0.9465
No log	0.24	12	0.8302	-0.0027	0.8302	0.9111
No log	0.28	14	0.8096	0.0	0.8096	0.8998
No log	0.32	16	0.8459	0.0	0.8459	0.9198
No log	0.36	18	0.8157	-0.0444	0.8157	0.9031
No log	0.4	20	0.7359	0.0717	0.7359	0.8578
No log	0.44	22	0.7123	0.2407	0.7123	0.8440
No log	0.48	24	0.6686	0.2963	0.6686	0.8177
No log	0.52	26	0.6946	0.2041	0.6946	0.8334
No log	0.56	28	0.9998	0.0975	0.9998	0.9999
No log	0.6	30	1.1047	0.1265	1.1047	1.0510
No log	0.64	32	0.9848	0.0651	0.9848	0.9924
No log	0.68	34	0.8076	0.1714	0.8076	0.8987
No log	0.72	36	0.6670	0.2412	0.6670	0.8167
No log	0.76	38	0.7052	0.3238	0.7052	0.8398
No log	0.8	40	0.7228	0.3099	0.7228	0.8502
No log	0.84	42	0.6196	0.3673	0.6196	0.7871
No log	0.88	44	0.5745	0.4709	0.5745	0.7579
No log	0.92	46	0.5651	0.4538	0.5651	0.7517
No log	0.96	48	0.5628	0.5470	0.5628	0.7502
No log	1.0	50	0.7374	0.3693	0.7374	0.8587
No log	1.04	52	1.0917	0.2214	1.0917	1.0449
No log	1.08	54	1.1289	0.2439	1.1289	1.0625
No log	1.12	56	0.9948	0.2658	0.9948	0.9974
No log	1.16	58	0.9209	0.2273	0.9209	0.9596
No log	1.2	60	0.8068	0.2817	0.8068	0.8982
No log	1.24	62	0.6364	0.3829	0.6364	0.7978
No log	1.28	64	0.6023	0.3633	0.6023	0.7761
No log	1.32	66	0.5990	0.2641	0.5990	0.7740
No log	1.3600	68	0.5737	0.4007	0.5737	0.7575
No log	1.4	70	0.5646	0.4238	0.5646	0.7514
No log	1.44	72	0.5500	0.3274	0.5500	0.7417
No log	1.48	74	0.5849	0.2817	0.5849	0.7648
No log	1.52	76	0.6114	0.2851	0.6114	0.7819
No log	1.56	78	0.6346	0.3506	0.6346	0.7966
No log	1.6	80	0.6331	0.3477	0.6331	0.7957
No log	1.6400	82	0.6272	0.3446	0.6272	0.7919
No log	1.6800	84	0.6046	0.2890	0.6046	0.7775
No log	1.72	86	0.6001	0.2890	0.6001	0.7747
No log	1.76	88	0.6039	0.3866	0.6039	0.7771
No log	1.8	90	0.6017	0.3701	0.6017	0.7757
No log	1.8400	92	0.5704	0.4729	0.5704	0.7553
No log	1.88	94	0.5891	0.4677	0.5891	0.7675
No log	1.92	96	0.6263	0.4537	0.6263	0.7914
No log	1.96	98	0.6160	0.4969	0.6160	0.7849
No log	2.0	100	0.5127	0.4358	0.5127	0.7160
No log	2.04	102	0.5279	0.4855	0.5279	0.7266
No log	2.08	104	0.5255	0.4769	0.5255	0.7249
No log	2.12	106	0.5069	0.4973	0.5069	0.7120
No log	2.16	108	0.6200	0.5267	0.6200	0.7874
No log	2.2	110	0.8134	0.5167	0.8134	0.9019
No log	2.24	112	0.8133	0.4993	0.8133	0.9018
No log	2.2800	114	0.8409	0.3942	0.8409	0.9170
No log	2.32	116	0.7301	0.4383	0.7301	0.8545
No log	2.36	118	0.5747	0.5400	0.5747	0.7581
No log	2.4	120	0.4916	0.5022	0.4916	0.7011
No log	2.44	122	0.6211	0.4303	0.6211	0.7881
No log	2.48	124	0.6746	0.4716	0.6746	0.8213
No log	2.52	126	0.6074	0.4568	0.6074	0.7794
No log	2.56	128	0.5660	0.5050	0.5660	0.7523
No log	2.6	130	0.5394	0.4817	0.5394	0.7344
No log	2.64	132	0.5561	0.5117	0.5561	0.7457
No log	2.68	134	0.6166	0.6154	0.6166	0.7853
No log	2.7200	136	0.5806	0.6154	0.5806	0.7620
No log	2.76	138	0.5291	0.5518	0.5291	0.7274
No log	2.8	140	0.6548	0.5614	0.6548	0.8092
No log	2.84	142	0.7484	0.4199	0.7484	0.8651
No log	2.88	144	0.6987	0.4353	0.6987	0.8359
No log	2.92	146	0.5894	0.4356	0.5894	0.7677
No log	2.96	148	0.5429	0.5463	0.5429	0.7368
No log	3.0	150	0.5429	0.4762	0.5429	0.7368
No log	3.04	152	0.5881	0.3890	0.5881	0.7669
No log	3.08	154	0.7048	0.3305	0.7048	0.8395
No log	3.12	156	0.7130	0.3475	0.7130	0.8444
No log	3.16	158	0.6015	0.3914	0.6015	0.7756
No log	3.2	160	0.5330	0.5357	0.5330	0.7301
No log	3.24	162	0.5635	0.5779	0.5635	0.7507
No log	3.2800	164	0.6024	0.5765	0.6024	0.7762
No log	3.32	166	0.6626	0.5441	0.6626	0.8140
No log	3.36	168	0.6894	0.5441	0.6894	0.8303
No log	3.4	170	0.6918	0.5502	0.6918	0.8318
No log	3.44	172	0.5847	0.6445	0.5847	0.7647
No log	3.48	174	0.5332	0.6156	0.5332	0.7302
No log	3.52	176	0.5292	0.5956	0.5292	0.7274
No log	3.56	178	0.5661	0.5779	0.5661	0.7524
No log	3.6	180	0.6489	0.5752	0.6489	0.8055
No log	3.64	182	0.6117	0.5696	0.6117	0.7821
No log	3.68	184	0.5219	0.5826	0.5219	0.7225
No log	3.7200	186	0.5343	0.5184	0.5343	0.7310
No log	3.76	188	0.6267	0.4556	0.6267	0.7916
No log	3.8	190	0.6250	0.5195	0.6250	0.7905
No log	3.84	192	0.6173	0.4794	0.6173	0.7857
No log	3.88	194	0.5674	0.4340	0.5674	0.7533
No log	3.92	196	0.5695	0.4478	0.5695	0.7546
No log	3.96	198	0.5560	0.5044	0.5560	0.7456
No log	4.0	200	0.5479	0.5214	0.5479	0.7402
No log	4.04	202	0.5409	0.5165	0.5409	0.7354
No log	4.08	204	0.5493	0.4772	0.5493	0.7411
No log	4.12	206	0.5523	0.4374	0.5523	0.7431
No log	4.16	208	0.5666	0.4644	0.5666	0.7527
No log	4.2	210	0.5274	0.4601	0.5274	0.7262
No log	4.24	212	0.5524	0.5268	0.5524	0.7432
No log	4.28	214	0.6265	0.5793	0.6265	0.7915
No log	4.32	216	0.6184	0.5793	0.6184	0.7864
No log	4.36	218	0.6912	0.5186	0.6912	0.8314
No log	4.4	220	0.6006	0.5283	0.6006	0.7750
No log	4.44	222	0.5609	0.5359	0.5609	0.7489
No log	4.48	224	0.5705	0.5717	0.5705	0.7553
No log	4.52	226	0.5613	0.5717	0.5613	0.7492
No log	4.5600	228	0.5204	0.5878	0.5204	0.7214
No log	4.6	230	0.5446	0.5226	0.5446	0.7380
No log	4.64	232	0.5585	0.5275	0.5585	0.7474
No log	4.68	234	0.5242	0.5341	0.5242	0.7240
No log	4.72	236	0.5010	0.5930	0.5010	0.7078
No log	4.76	238	0.4960	0.5698	0.4960	0.7043
No log	4.8	240	0.4983	0.5326	0.4983	0.7059
No log	4.84	242	0.5088	0.4883	0.5088	0.7133
No log	4.88	244	0.5046	0.4986	0.5046	0.7103
No log	4.92	246	0.4877	0.5625	0.4877	0.6983
No log	4.96	248	0.4897	0.6503	0.4897	0.6998
No log	5.0	250	0.4784	0.5915	0.4784	0.6917
No log	5.04	252	0.4721	0.5565	0.4721	0.6871
No log	5.08	254	0.5047	0.5945	0.5047	0.7104
No log	5.12	256	0.5624	0.5765	0.5624	0.7499
No log	5.16	258	0.5078	0.5445	0.5078	0.7126
No log	5.2	260	0.4717	0.5904	0.4717	0.6868
No log	5.24	262	0.4646	0.6154	0.4646	0.6816
No log	5.28	264	0.4639	0.5457	0.4639	0.6811
No log	5.32	266	0.4713	0.4746	0.4713	0.6865
No log	5.36	268	0.4849	0.4746	0.4849	0.6963
No log	5.4	270	0.5242	0.5449	0.5242	0.7240
No log	5.44	272	0.5075	0.5432	0.5075	0.7124
No log	5.48	274	0.4706	0.5010	0.4706	0.6860
No log	5.52	276	0.4722	0.5246	0.4722	0.6871
No log	5.5600	278	0.4673	0.5831	0.4673	0.6836
No log	5.6	280	0.4655	0.5800	0.4655	0.6823
No log	5.64	282	0.4669	0.5440	0.4669	0.6833
No log	5.68	284	0.4945	0.5947	0.4945	0.7032
No log	5.72	286	0.6726	0.5223	0.6726	0.8201
No log	5.76	288	0.6738	0.5281	0.6738	0.8208
No log	5.8	290	0.5238	0.5650	0.5238	0.7238
No log	5.84	292	0.4784	0.5320	0.4784	0.6916
No log	5.88	294	0.4983	0.4838	0.4983	0.7059
No log	5.92	296	0.4824	0.5003	0.4824	0.6945
No log	5.96	298	0.4814	0.6053	0.4814	0.6938
No log	6.0	300	0.5050	0.6342	0.5050	0.7106
No log	6.04	302	0.4937	0.6141	0.4937	0.7026
No log	6.08	304	0.4940	0.6342	0.4940	0.7028
No log	6.12	306	0.4685	0.5548	0.4685	0.6845
No log	6.16	308	0.5418	0.6005	0.5418	0.7361
No log	6.2	310	0.6455	0.5309	0.6455	0.8034
No log	6.24	312	0.5924	0.5494	0.5924	0.7697
No log	6.28	314	0.4890	0.5708	0.4890	0.6993
No log	6.32	316	0.4769	0.5571	0.4769	0.6906
No log	6.36	318	0.5212	0.5513	0.5212	0.7219
No log	6.4	320	0.5715	0.5862	0.5715	0.7560
No log	6.44	322	0.5227	0.5621	0.5227	0.7230
No log	6.48	324	0.4817	0.5840	0.4817	0.6940
No log	6.52	326	0.5440	0.5072	0.5440	0.7376
No log	6.5600	328	0.5574	0.5126	0.5574	0.7466
No log	6.6	330	0.4992	0.4847	0.4992	0.7066
No log	6.64	332	0.4937	0.5512	0.4937	0.7026
No log	6.68	334	0.6117	0.5543	0.6117	0.7821
No log	6.72	336	0.6986	0.4844	0.6986	0.8358
No log	6.76	338	0.6354	0.5281	0.6354	0.7971
No log	6.8	340	0.5170	0.5512	0.5170	0.7190
No log	6.84	342	0.4617	0.5286	0.4617	0.6795
No log	6.88	344	0.4616	0.5095	0.4616	0.6794
No log	6.92	346	0.4567	0.5929	0.4567	0.6758
No log	6.96	348	0.4984	0.5779	0.4984	0.7059
No log	7.0	350	0.6477	0.5209	0.6477	0.8048
No log	7.04	352	0.7219	0.5226	0.7219	0.8497
No log	7.08	354	0.5991	0.5516	0.5991	0.7740
No log	7.12	356	0.5062	0.6716	0.5062	0.7115
No log	7.16	358	0.5068	0.6361	0.5068	0.7119
No log	7.2	360	0.4679	0.6514	0.4679	0.6841
No log	7.24	362	0.4439	0.5953	0.4439	0.6662
No log	7.28	364	0.4523	0.5853	0.4523	0.6726
No log	7.32	366	0.4553	0.6040	0.4553	0.6748
No log	7.36	368	0.4854	0.6141	0.4854	0.6967
No log	7.4	370	0.4878	0.5733	0.4878	0.6984
No log	7.44	372	0.4710	0.5283	0.4710	0.6863
No log	7.48	374	0.4586	0.5475	0.4586	0.6772
No log	7.52	376	0.4691	0.5687	0.4691	0.6849
No log	7.5600	378	0.4931	0.5692	0.4931	0.7022
No log	7.6	380	0.4821	0.5756	0.4821	0.6943
No log	7.64	382	0.4581	0.6572	0.4581	0.6768
No log	7.68	384	0.4743	0.6515	0.4743	0.6887
No log	7.72	386	0.4918	0.6445	0.4918	0.7013
No log	7.76	388	0.5228	0.6470	0.5228	0.7230
No log	7.8	390	0.4867	0.6783	0.4867	0.6976
No log	7.84	392	0.4723	0.5913	0.4723	0.6873
No log	7.88	394	0.5149	0.5876	0.5149	0.7176
No log	7.92	396	0.5154	0.5761	0.5154	0.7179
No log	7.96	398	0.4777	0.5756	0.4777	0.6911
No log	8.0	400	0.4897	0.6146	0.4897	0.6998
No log	8.04	402	0.5183	0.6633	0.5183	0.7199
No log	8.08	404	0.5019	0.6793	0.5019	0.7084
No log	8.12	406	0.4572	0.5939	0.4572	0.6762
No log	8.16	408	0.4813	0.6018	0.4813	0.6937
No log	8.2	410	0.5077	0.6101	0.5077	0.7125
No log	8.24	412	0.4823	0.6083	0.4823	0.6945
No log	8.28	414	0.4778	0.6009	0.4778	0.6912
No log	8.32	416	0.4936	0.5865	0.4936	0.7026
No log	8.36	418	0.5456	0.5396	0.5456	0.7387
No log	8.4	420	0.5782	0.5354	0.5782	0.7604
No log	8.44	422	0.5427	0.5705	0.5427	0.7367
No log	8.48	424	0.4997	0.5272	0.4997	0.7069
No log	8.52	426	0.4630	0.6156	0.4630	0.6804
No log	8.56	428	0.5106	0.5621	0.5106	0.7146
No log	8.6	430	0.5387	0.5445	0.5387	0.7339
No log	8.64	432	0.5006	0.5627	0.5006	0.7075
No log	8.68	434	0.4790	0.6330	0.4790	0.6921
No log	8.72	436	0.5265	0.6080	0.5265	0.7256
No log	8.76	438	0.5393	0.6269	0.5393	0.7344
No log	8.8	440	0.5121	0.5985	0.5121	0.7156
No log	8.84	442	0.5141	0.6184	0.5141	0.7170
No log	8.88	444	0.5104	0.5692	0.5104	0.7145
No log	8.92	446	0.4743	0.6477	0.4743	0.6887
No log	8.96	448	0.4673	0.6330	0.4673	0.6836
No log	9.0	450	0.4655	0.6017	0.4655	0.6823
No log	9.04	452	0.4667	0.6068	0.4667	0.6831
No log	9.08	454	0.4743	0.5999	0.4743	0.6887
No log	9.12	456	0.4597	0.5875	0.4597	0.6780
No log	9.16	458	0.4568	0.5979	0.4568	0.6758
No log	9.2	460	0.4666	0.5783	0.4666	0.6831
No log	9.24	462	0.5043	0.5861	0.5043	0.7101
No log	9.28	464	0.5210	0.5683	0.5210	0.7218
No log	9.32	466	0.4673	0.6182	0.4673	0.6836
No log	9.36	468	0.4505	0.6071	0.4505	0.6712
No log	9.4	470	0.4449	0.6121	0.4449	0.6670
No log	9.44	472	0.4507	0.5669	0.4507	0.6713
No log	9.48	474	0.4888	0.6004	0.4888	0.6991
No log	9.52	476	0.5194	0.6030	0.5194	0.7207
No log	9.56	478	0.4999	0.5895	0.4999	0.7070
No log	9.6	480	0.4755	0.6186	0.4755	0.6896
No log	9.64	482	0.4889	0.5991	0.4889	0.6992
No log	9.68	484	0.4789	0.5923	0.4789	0.6920
No log	9.72	486	0.4791	0.5672	0.4791	0.6922
No log	9.76	488	0.4849	0.5248	0.4849	0.6963
No log	9.8	490	0.4842	0.4942	0.4842	0.6959
No log	9.84	492	0.4981	0.4960	0.4981	0.7058
No log	9.88	494	0.5194	0.5513	0.5194	0.7207
No log	9.92	496	0.5231	0.5513	0.5231	0.7233
No log	9.96	498	0.5218	0.5553	0.5218	0.7224
0.3903	10.0	500	0.5106	0.5335	0.5106	0.7146
0.3903	10.04	502	0.5077	0.5217	0.5077	0.7125
0.3903	10.08	504	0.5159	0.5647	0.5159	0.7182
0.3903	10.12	506	0.5111	0.4613	0.5111	0.7149
0.3903	10.16	508	0.4971	0.5010	0.4971	0.7051
0.3903	10.2	510	0.5194	0.5144	0.5194	0.7207
0.3903	10.24	512	0.6029	0.4909	0.6029	0.7764
0.3903	10.28	514	0.6566	0.4890	0.6566	0.8103

Framework versions

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