ArabicNewSplits7_FineTuningAraBERT_run2_AugV5_k2_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: 0.4352
• Qwk: 0.6132
• Mse: 0.4352
• Rmse: 0.6597

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	0.25	2	2.4023	0.0052	2.4023	1.5499
No log	0.5	4	1.1655	0.1259	1.1655	1.0796
No log	0.75	6	0.7541	0.1372	0.7541	0.8684
No log	1.0	8	0.7285	0.2319	0.7285	0.8535
No log	1.25	10	0.6985	0.2783	0.6985	0.8357
No log	1.5	12	0.8415	0.2375	0.8415	0.9173
No log	1.75	14	0.6713	0.3029	0.6713	0.8194
No log	2.0	16	0.6546	0.3690	0.6546	0.8091
No log	2.25	18	0.7346	0.4074	0.7346	0.8571
No log	2.5	20	0.6949	0.4172	0.6949	0.8336
No log	2.75	22	0.5933	0.4569	0.5933	0.7702
No log	3.0	24	0.5082	0.5046	0.5082	0.7129
No log	3.25	26	0.4662	0.4656	0.4662	0.6828
No log	3.5	28	0.4559	0.5488	0.4559	0.6752
No log	3.75	30	0.4866	0.6052	0.4866	0.6976
No log	4.0	32	0.5134	0.5950	0.5134	0.7165
No log	4.25	34	0.4695	0.4471	0.4695	0.6852
No log	4.5	36	0.6077	0.5112	0.6077	0.7795
No log	4.75	38	0.4910	0.5373	0.4910	0.7007
No log	5.0	40	0.4398	0.5939	0.4398	0.6632
No log	5.25	42	0.4167	0.6517	0.4167	0.6456
No log	5.5	44	0.4541	0.5779	0.4541	0.6738
No log	5.75	46	0.5788	0.5587	0.5788	0.7608
No log	6.0	48	0.8070	0.4464	0.8070	0.8983
No log	6.25	50	0.7189	0.4633	0.7189	0.8479
No log	6.5	52	0.4613	0.6017	0.4613	0.6792
No log	6.75	54	0.3903	0.6929	0.3903	0.6247
No log	7.0	56	0.4025	0.6598	0.4025	0.6345
No log	7.25	58	0.4032	0.6339	0.4032	0.6349
No log	7.5	60	0.4184	0.6946	0.4184	0.6468
No log	7.75	62	0.4942	0.6047	0.4942	0.7030
No log	8.0	64	0.4830	0.6114	0.4830	0.6950
No log	8.25	66	0.4558	0.6010	0.4558	0.6751
No log	8.5	68	0.4602	0.6068	0.4602	0.6784
No log	8.75	70	0.4621	0.5985	0.4621	0.6798
No log	9.0	72	0.4760	0.6514	0.4760	0.6899
No log	9.25	74	0.4554	0.6828	0.4554	0.6748
No log	9.5	76	0.4965	0.5514	0.4965	0.7046
No log	9.75	78	0.4453	0.6241	0.4453	0.6673
No log	10.0	80	0.4359	0.7012	0.4359	0.6602
No log	10.25	82	0.4308	0.7012	0.4308	0.6564
No log	10.5	84	0.4461	0.6252	0.4461	0.6679
No log	10.75	86	0.4604	0.5723	0.4604	0.6786
No log	11.0	88	0.4782	0.5723	0.4782	0.6916
No log	11.25	90	0.4452	0.6542	0.4452	0.6672
No log	11.5	92	0.4517	0.5549	0.4517	0.6721
No log	11.75	94	0.4680	0.5723	0.4680	0.6841
No log	12.0	96	0.4415	0.6078	0.4415	0.6645
No log	12.25	98	0.4280	0.7041	0.4280	0.6542
No log	12.5	100	0.4186	0.6847	0.4186	0.6470
No log	12.75	102	0.4381	0.6307	0.4381	0.6619
No log	13.0	104	0.4293	0.6484	0.4293	0.6552
No log	13.25	106	0.4267	0.6928	0.4267	0.6532
No log	13.5	108	0.4959	0.6206	0.4959	0.7042
No log	13.75	110	0.4875	0.6379	0.4875	0.6982
No log	14.0	112	0.4718	0.6547	0.4718	0.6869
No log	14.25	114	0.4790	0.6547	0.4790	0.6921
No log	14.5	116	0.4202	0.6900	0.4202	0.6483
No log	14.75	118	0.4143	0.6570	0.4143	0.6437
No log	15.0	120	0.4101	0.6639	0.4101	0.6404
No log	15.25	122	0.4147	0.6772	0.4147	0.6440
No log	15.5	124	0.3999	0.7032	0.3999	0.6324
No log	15.75	126	0.4301	0.6349	0.4301	0.6558
No log	16.0	128	0.5999	0.4617	0.5999	0.7745
No log	16.25	130	0.5035	0.6658	0.5035	0.7096
No log	16.5	132	0.4140	0.6890	0.4140	0.6435
No log	16.75	134	0.4616	0.6259	0.4616	0.6794
No log	17.0	136	0.4258	0.6552	0.4258	0.6525
No log	17.25	138	0.4281	0.5929	0.4281	0.6543
No log	17.5	140	0.4265	0.5875	0.4265	0.6531
No log	17.75	142	0.4325	0.5999	0.4325	0.6576
No log	18.0	144	0.4263	0.6068	0.4263	0.6529
No log	18.25	146	0.4761	0.6091	0.4761	0.6900
No log	18.5	148	0.5250	0.6038	0.5250	0.7246
No log	18.75	150	0.4500	0.6162	0.4500	0.6708
No log	19.0	152	0.4550	0.5657	0.4550	0.6746
No log	19.25	154	0.4499	0.5841	0.4499	0.6707
No log	19.5	156	0.4387	0.6530	0.4387	0.6623
No log	19.75	158	0.4610	0.6309	0.4610	0.6790
No log	20.0	160	0.4242	0.6448	0.4242	0.6513
No log	20.25	162	0.4733	0.5384	0.4733	0.6879
No log	20.5	164	0.5117	0.4969	0.5117	0.7154
No log	20.75	166	0.4441	0.6047	0.4441	0.6664
No log	21.0	168	0.4198	0.6344	0.4198	0.6479
No log	21.25	170	0.4222	0.6389	0.4222	0.6498
No log	21.5	172	0.4226	0.6125	0.4226	0.6501
No log	21.75	174	0.4213	0.6125	0.4213	0.6490
No log	22.0	176	0.4237	0.6125	0.4237	0.6509
No log	22.25	178	0.4467	0.5999	0.4467	0.6683
No log	22.5	180	0.4398	0.5985	0.4398	0.6632
No log	22.75	182	0.4332	0.5765	0.4332	0.6582
No log	23.0	184	0.4306	0.5831	0.4306	0.6562
No log	23.25	186	0.4306	0.5831	0.4306	0.6562
No log	23.5	188	0.4300	0.6448	0.4300	0.6558
No log	23.75	190	0.4559	0.5765	0.4559	0.6752
No log	24.0	192	0.4698	0.5779	0.4698	0.6855
No log	24.25	194	0.4195	0.6567	0.4195	0.6477
No log	24.5	196	0.4269	0.6919	0.4269	0.6534
No log	24.75	198	0.4420	0.6381	0.4420	0.6648
No log	25.0	200	0.4068	0.6747	0.4068	0.6378
No log	25.25	202	0.4562	0.6457	0.4562	0.6754
No log	25.5	204	0.5250	0.6042	0.5250	0.7245
No log	25.75	206	0.4513	0.6445	0.4513	0.6718
No log	26.0	208	0.4405	0.7025	0.4405	0.6637
No log	26.25	210	0.5498	0.6275	0.5498	0.7415
No log	26.5	212	0.5134	0.6200	0.5134	0.7165
No log	26.75	214	0.4229	0.6388	0.4229	0.6503
No log	27.0	216	0.4346	0.6419	0.4346	0.6592
No log	27.25	218	0.4205	0.6101	0.4205	0.6485
No log	27.5	220	0.4118	0.6357	0.4118	0.6417
No log	27.75	222	0.4378	0.6518	0.4378	0.6617
No log	28.0	224	0.4250	0.6431	0.4250	0.6519
No log	28.25	226	0.4136	0.5890	0.4136	0.6431
No log	28.5	228	0.4245	0.6305	0.4245	0.6515
No log	28.75	230	0.4131	0.6448	0.4131	0.6427
No log	29.0	232	0.4334	0.6809	0.4334	0.6583
No log	29.25	234	0.4306	0.6503	0.4306	0.6562
No log	29.5	236	0.4177	0.6448	0.4177	0.6463
No log	29.75	238	0.4224	0.6448	0.4224	0.6499
No log	30.0	240	0.4304	0.6068	0.4304	0.6560
No log	30.25	242	0.4300	0.5555	0.4300	0.6558
No log	30.5	244	0.4118	0.6154	0.4118	0.6417
No log	30.75	246	0.4114	0.6832	0.4114	0.6414
No log	31.0	248	0.4104	0.7055	0.4104	0.6406
No log	31.25	250	0.4232	0.6154	0.4232	0.6505
No log	31.5	252	0.4238	0.6370	0.4238	0.6510
No log	31.75	254	0.4069	0.6843	0.4069	0.6379
No log	32.0	256	0.4122	0.6611	0.4122	0.6421
No log	32.25	258	0.4061	0.6344	0.4061	0.6373
No log	32.5	260	0.4093	0.6344	0.4093	0.6398
No log	32.75	262	0.4081	0.6448	0.4081	0.6389
No log	33.0	264	0.4130	0.6542	0.4130	0.6427
No log	33.25	266	0.4134	0.6542	0.4134	0.6429
No log	33.5	268	0.4097	0.6448	0.4097	0.6401
No log	33.75	270	0.4173	0.6243	0.4173	0.6460
No log	34.0	272	0.4377	0.5208	0.4377	0.6616
No log	34.25	274	0.4245	0.6032	0.4245	0.6516
No log	34.5	276	0.4351	0.5899	0.4351	0.6597
No log	34.75	278	0.4475	0.6032	0.4475	0.6690
No log	35.0	280	0.4400	0.6241	0.4400	0.6633
No log	35.25	282	0.4339	0.6553	0.4339	0.6587
No log	35.5	284	0.4344	0.6289	0.4344	0.6591
No log	35.75	286	0.4354	0.6105	0.4354	0.6599
No log	36.0	288	0.4335	0.6154	0.4335	0.6584
No log	36.25	290	0.4489	0.6446	0.4489	0.6700
No log	36.5	292	0.4909	0.5944	0.4909	0.7006
No log	36.75	294	0.4658	0.5859	0.4658	0.6825
No log	37.0	296	0.4247	0.6357	0.4247	0.6517
No log	37.25	298	0.4252	0.6770	0.4252	0.6521
No log	37.5	300	0.4181	0.6357	0.4181	0.6466
No log	37.75	302	0.4199	0.6344	0.4199	0.6480
No log	38.0	304	0.4211	0.6435	0.4211	0.6490
No log	38.25	306	0.4346	0.6293	0.4346	0.6592
No log	38.5	308	0.4235	0.6215	0.4235	0.6508
No log	38.75	310	0.4264	0.5986	0.4264	0.6530
No log	39.0	312	0.4368	0.5836	0.4368	0.6609
No log	39.25	314	0.4633	0.5715	0.4633	0.6807
No log	39.5	316	0.4949	0.5802	0.4949	0.7035
No log	39.75	318	0.4754	0.5955	0.4754	0.6895
No log	40.0	320	0.4618	0.5693	0.4618	0.6796
No log	40.25	322	0.4533	0.5693	0.4533	0.6733
No log	40.5	324	0.4483	0.5693	0.4483	0.6696
No log	40.75	326	0.4428	0.5587	0.4428	0.6654
No log	41.0	328	0.4310	0.5836	0.4310	0.6565
No log	41.25	330	0.4209	0.5749	0.4209	0.6488
No log	41.5	332	0.4207	0.5749	0.4207	0.6486
No log	41.75	334	0.4378	0.6602	0.4378	0.6617
No log	42.0	336	0.4589	0.5939	0.4589	0.6774
No log	42.25	338	0.4726	0.5939	0.4726	0.6875
No log	42.5	340	0.4620	0.5939	0.4620	0.6797
No log	42.75	342	0.4261	0.6111	0.4261	0.6528
No log	43.0	344	0.4149	0.6542	0.4149	0.6442
No log	43.25	346	0.4128	0.6542	0.4128	0.6425
No log	43.5	348	0.4161	0.6830	0.4161	0.6451
No log	43.75	350	0.4240	0.6818	0.4240	0.6511
No log	44.0	352	0.4307	0.6627	0.4307	0.6563
No log	44.25	354	0.4357	0.6627	0.4357	0.6600
No log	44.5	356	0.4334	0.6542	0.4334	0.6583
No log	44.75	358	0.4296	0.6344	0.4296	0.6554
No log	45.0	360	0.4345	0.6096	0.4345	0.6592
No log	45.25	362	0.4430	0.6395	0.4430	0.6656
No log	45.5	364	0.4426	0.6395	0.4426	0.6653
No log	45.75	366	0.4290	0.6344	0.4290	0.6549
No log	46.0	368	0.4280	0.6243	0.4280	0.6542
No log	46.25	370	0.4249	0.6448	0.4249	0.6518
No log	46.5	372	0.4231	0.6448	0.4231	0.6505
No log	46.75	374	0.4246	0.6154	0.4246	0.6516
No log	47.0	376	0.4255	0.6489	0.4255	0.6523
No log	47.25	378	0.4195	0.6448	0.4195	0.6477
No log	47.5	380	0.4281	0.5883	0.4281	0.6543
No log	47.75	382	0.4340	0.5970	0.4340	0.6588
No log	48.0	384	0.4241	0.6001	0.4241	0.6512
No log	48.25	386	0.4213	0.6448	0.4213	0.6491
No log	48.5	388	0.4204	0.6229	0.4204	0.6484
No log	48.75	390	0.4198	0.6229	0.4198	0.6479
No log	49.0	392	0.4251	0.6370	0.4251	0.6520
No log	49.25	394	0.4347	0.6127	0.4347	0.6593
No log	49.5	396	0.4502	0.6260	0.4502	0.6710
No log	49.75	398	0.4312	0.6370	0.4312	0.6567
No log	50.0	400	0.4326	0.6240	0.4326	0.6577
No log	50.25	402	0.4485	0.5947	0.4485	0.6697
No log	50.5	404	0.4378	0.5631	0.4378	0.6616
No log	50.75	406	0.4292	0.5550	0.4292	0.6552
No log	51.0	408	0.4166	0.6111	0.4166	0.6454
No log	51.25	410	0.4168	0.6111	0.4168	0.6456
No log	51.5	412	0.4217	0.5550	0.4217	0.6494
No log	51.75	414	0.4245	0.6241	0.4245	0.6515
No log	52.0	416	0.4365	0.6518	0.4365	0.6607
No log	52.25	418	0.4304	0.6252	0.4304	0.6561
No log	52.5	420	0.4290	0.6678	0.4290	0.6550
No log	52.75	422	0.4385	0.6601	0.4385	0.6622
No log	53.0	424	0.4374	0.6601	0.4374	0.6614
No log	53.25	426	0.4276	0.6370	0.4276	0.6539
No log	53.5	428	0.4336	0.5883	0.4336	0.6585
No log	53.75	430	0.4437	0.5631	0.4437	0.6661
No log	54.0	432	0.4411	0.5897	0.4411	0.6642
No log	54.25	434	0.4297	0.5883	0.4297	0.6555
No log	54.5	436	0.4288	0.6282	0.4288	0.6548
No log	54.75	438	0.4365	0.6601	0.4365	0.6607
No log	55.0	440	0.4310	0.6601	0.4310	0.6565
No log	55.25	442	0.4246	0.6542	0.4246	0.6516
No log	55.5	444	0.4390	0.6321	0.4390	0.6625
No log	55.75	446	0.4444	0.6321	0.4444	0.6666
No log	56.0	448	0.4368	0.6313	0.4368	0.6609
No log	56.25	450	0.4242	0.6632	0.4242	0.6513
No log	56.5	452	0.4239	0.6242	0.4239	0.6511
No log	56.75	454	0.4261	0.5831	0.4261	0.6528
No log	57.0	456	0.4248	0.6214	0.4248	0.6518
No log	57.25	458	0.4347	0.6401	0.4347	0.6593
No log	57.5	460	0.4594	0.6388	0.4594	0.6778
No log	57.75	462	0.4710	0.6399	0.4710	0.6863
No log	58.0	464	0.4591	0.6388	0.4591	0.6775
No log	58.25	466	0.4352	0.6401	0.4352	0.6597
No log	58.5	468	0.4237	0.6111	0.4237	0.6510
No log	58.75	470	0.4304	0.6101	0.4304	0.6561
No log	59.0	472	0.4409	0.6330	0.4409	0.6640
No log	59.25	474	0.4321	0.5781	0.4321	0.6574
No log	59.5	476	0.4213	0.6024	0.4213	0.6491
No log	59.75	478	0.4270	0.5550	0.4270	0.6535
No log	60.0	480	0.4382	0.6492	0.4382	0.6620
No log	60.25	482	0.4518	0.6388	0.4518	0.6722
No log	60.5	484	0.4583	0.6173	0.4583	0.6770
No log	60.75	486	0.4425	0.5980	0.4425	0.6652
No log	61.0	488	0.4300	0.6303	0.4300	0.6558
No log	61.25	490	0.4195	0.6330	0.4195	0.6477
No log	61.5	492	0.4219	0.6068	0.4219	0.6495
No log	61.75	494	0.4225	0.6068	0.4225	0.6500
No log	62.0	496	0.4186	0.6377	0.4186	0.6470
No log	62.25	498	0.4158	0.6830	0.4158	0.6449
0.1924	62.5	500	0.4253	0.6517	0.4253	0.6521
0.1924	62.75	502	0.4523	0.6388	0.4523	0.6725
0.1924	63.0	504	0.4734	0.5802	0.4734	0.6880
0.1924	63.25	506	0.4719	0.5802	0.4719	0.6870
0.1924	63.5	508	0.4543	0.5980	0.4543	0.6740
0.1924	63.75	510	0.4352	0.6132	0.4352	0.6597

Framework versions

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