ArabicNewSplits7_FineTuningAraBERT_run3_AugV5_k3_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: 0.4424
• Qwk: 0.5662
• Mse: 0.4424
• Rmse: 0.6651

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	0.2	2	2.6123	-0.1213	2.6123	1.6162
No log	0.4	4	1.2733	0.0495	1.2733	1.1284
No log	0.6	6	0.8888	0.0535	0.8888	0.9427
No log	0.8	8	0.8559	0.2328	0.8559	0.9251
No log	1.0	10	0.7017	0.2621	0.7017	0.8377
No log	1.2	12	0.6839	0.3169	0.6839	0.8270
No log	1.4	14	0.7108	0.2885	0.7108	0.8431
No log	1.6	16	0.6370	0.2063	0.6370	0.7981
No log	1.8	18	0.6868	0.3312	0.6868	0.8287
No log	2.0	20	0.6771	0.3312	0.6771	0.8228
No log	2.2	22	0.6451	0.2471	0.6451	0.8032
No log	2.4	24	0.6144	0.1327	0.6144	0.7838
No log	2.6	26	0.7481	0.3090	0.7481	0.8649
No log	2.8	28	0.7556	0.3051	0.7556	0.8693
No log	3.0	30	0.6302	0.3060	0.6302	0.7938
No log	3.2	32	0.5401	0.4420	0.5401	0.7349
No log	3.4	34	0.5822	0.4997	0.5822	0.7630
No log	3.6	36	0.5125	0.4468	0.5125	0.7159
No log	3.8	38	0.5040	0.4061	0.5040	0.7100
No log	4.0	40	0.4915	0.4561	0.4915	0.7011
No log	4.2	42	0.4976	0.5228	0.4976	0.7054
No log	4.4	44	0.4834	0.5386	0.4834	0.6953
No log	4.6	46	0.4585	0.5617	0.4585	0.6771
No log	4.8	48	0.4640	0.5918	0.4640	0.6812
No log	5.0	50	0.4477	0.6305	0.4477	0.6691
No log	5.2	52	0.4755	0.5779	0.4755	0.6896
No log	5.4	54	0.4755	0.5513	0.4755	0.6896
No log	5.6	56	0.4098	0.6601	0.4098	0.6401
No log	5.8	58	0.4967	0.5460	0.4967	0.7047
No log	6.0	60	0.6808	0.5670	0.6808	0.8251
No log	6.2	62	0.4878	0.5687	0.4878	0.6984
No log	6.4	64	0.4226	0.6446	0.4226	0.6501
No log	6.6	66	0.4205	0.6849	0.4205	0.6484
No log	6.8	68	0.7067	0.4867	0.7067	0.8407
No log	7.0	70	1.0648	0.3424	1.0648	1.0319
No log	7.2	72	1.0912	0.3303	1.0912	1.0446
No log	7.4	74	0.6416	0.5047	0.6416	0.8010
No log	7.6	76	0.3946	0.6977	0.3946	0.6281
No log	7.8	78	0.4495	0.6787	0.4495	0.6704
No log	8.0	80	0.3844	0.6847	0.3844	0.6200
No log	8.2	82	0.6298	0.5818	0.6298	0.7936
No log	8.4	84	0.9882	0.3676	0.9882	0.9941
No log	8.6	86	0.8814	0.4956	0.8814	0.9388
No log	8.8	88	0.5773	0.5862	0.5773	0.7598
No log	9.0	90	0.4577	0.5633	0.4577	0.6765
No log	9.2	92	0.4356	0.5114	0.4356	0.6600
No log	9.4	94	0.4416	0.4681	0.4416	0.6645
No log	9.6	96	0.4612	0.5577	0.4612	0.6791
No log	9.8	98	0.4991	0.5308	0.4991	0.7065
No log	10.0	100	0.6313	0.5570	0.6313	0.7945
No log	10.2	102	0.8276	0.4851	0.8276	0.9097
No log	10.4	104	0.6275	0.5259	0.6275	0.7921
No log	10.6	106	0.4443	0.6197	0.4443	0.6665
No log	10.8	108	0.5147	0.5836	0.5147	0.7174
No log	11.0	110	0.4674	0.5569	0.4674	0.6836
No log	11.2	112	0.4333	0.6201	0.4333	0.6582
No log	11.4	114	0.5515	0.5659	0.5515	0.7426
No log	11.6	116	0.5175	0.5470	0.5175	0.7194
No log	11.8	118	0.5130	0.5445	0.5130	0.7162
No log	12.0	120	0.4534	0.6341	0.4534	0.6733
No log	12.2	122	0.4400	0.6739	0.4400	0.6633
No log	12.4	124	0.4394	0.6301	0.4394	0.6628
No log	12.6	126	0.4540	0.6434	0.4540	0.6738
No log	12.8	128	0.4367	0.6292	0.4367	0.6608
No log	13.0	130	0.4506	0.5814	0.4506	0.6712
No log	13.2	132	0.4350	0.6518	0.4350	0.6595
No log	13.4	134	0.4498	0.6223	0.4498	0.6707
No log	13.6	136	0.5490	0.5595	0.5490	0.7410
No log	13.8	138	0.4917	0.5393	0.4917	0.7012
No log	14.0	140	0.4777	0.5922	0.4777	0.6912
No log	14.2	142	0.5272	0.5678	0.5272	0.7261
No log	14.4	144	0.4496	0.6248	0.4496	0.6705
No log	14.6	146	0.4162	0.6530	0.4162	0.6451
No log	14.8	148	0.4866	0.6135	0.4866	0.6975
No log	15.0	150	0.4420	0.6612	0.4420	0.6648
No log	15.2	152	0.4215	0.6672	0.4215	0.6492
No log	15.4	154	0.4367	0.6349	0.4367	0.6608
No log	15.6	156	0.4803	0.5922	0.4803	0.6930
No log	15.8	158	0.4289	0.6526	0.4289	0.6549
No log	16.0	160	0.4167	0.6530	0.4167	0.6455
No log	16.2	162	0.4359	0.6101	0.4359	0.6603
No log	16.4	164	0.4094	0.6530	0.4094	0.6398
No log	16.6	166	0.4527	0.6236	0.4527	0.6728
No log	16.8	168	0.5654	0.5153	0.5654	0.7519
No log	17.0	170	0.5670	0.5153	0.5670	0.7530
No log	17.2	172	0.4777	0.5735	0.4777	0.6912
No log	17.4	174	0.4087	0.6503	0.4087	0.6393
No log	17.6	176	0.4271	0.6359	0.4271	0.6535
No log	17.8	178	0.4298	0.6553	0.4298	0.6556
No log	18.0	180	0.4224	0.6183	0.4224	0.6499
No log	18.2	182	0.4155	0.5750	0.4155	0.6446
No log	18.4	184	0.4134	0.5846	0.4134	0.6430
No log	18.6	186	0.4142	0.5633	0.4142	0.6436
No log	18.8	188	0.4056	0.5539	0.4056	0.6369
No log	19.0	190	0.4105	0.6186	0.4105	0.6407
No log	19.2	192	0.4287	0.6599	0.4287	0.6547
No log	19.4	194	0.4103	0.6632	0.4103	0.6406
No log	19.6	196	0.4265	0.6337	0.4265	0.6531
No log	19.8	198	0.4774	0.6294	0.4774	0.6910
No log	20.0	200	0.5563	0.5678	0.5563	0.7458
No log	20.2	202	0.5007	0.6206	0.5007	0.7076
No log	20.4	204	0.4201	0.5672	0.4201	0.6482
No log	20.6	206	0.4461	0.6381	0.4461	0.6679
No log	20.8	208	0.5152	0.5086	0.5152	0.7178
No log	21.0	210	0.4872	0.5681	0.4872	0.6980
No log	21.2	212	0.4406	0.4729	0.4406	0.6638
No log	21.4	214	0.4298	0.5367	0.4298	0.6556
No log	21.6	216	0.4255	0.5405	0.4255	0.6523
No log	21.8	218	0.4208	0.6171	0.4208	0.6487
No log	22.0	220	0.4250	0.6210	0.4250	0.6520
No log	22.2	222	0.4331	0.6284	0.4331	0.6581
No log	22.4	224	0.4277	0.5910	0.4277	0.6540
No log	22.6	226	0.4195	0.5996	0.4195	0.6477
No log	22.8	228	0.4133	0.6052	0.4133	0.6429
No log	23.0	230	0.4092	0.5937	0.4092	0.6397
No log	23.2	232	0.4003	0.5883	0.4003	0.6327
No log	23.4	234	0.3895	0.6125	0.3895	0.6241
No log	23.6	236	0.3751	0.6716	0.3751	0.6125
No log	23.8	238	0.3827	0.6506	0.3827	0.6186
No log	24.0	240	0.4049	0.6593	0.4049	0.6363
No log	24.2	242	0.4547	0.6840	0.4547	0.6744
No log	24.4	244	0.4348	0.6853	0.4348	0.6594
No log	24.6	246	0.3823	0.6786	0.3823	0.6183
No log	24.8	248	0.3756	0.6255	0.3756	0.6128
No log	25.0	250	0.3948	0.6383	0.3948	0.6283
No log	25.2	252	0.3726	0.6154	0.3726	0.6104
No log	25.4	254	0.3605	0.6505	0.3605	0.6004
No log	25.6	256	0.4068	0.6863	0.4068	0.6378
No log	25.8	258	0.4810	0.6386	0.4810	0.6935
No log	26.0	260	0.4427	0.6019	0.4427	0.6653
No log	26.2	262	0.3836	0.6492	0.3836	0.6194
No log	26.4	264	0.3697	0.7043	0.3697	0.6081
No log	26.6	266	0.4174	0.5619	0.4174	0.6461
No log	26.8	268	0.4369	0.5544	0.4369	0.6610
No log	27.0	270	0.4040	0.5449	0.4040	0.6356
No log	27.2	272	0.3746	0.6832	0.3746	0.6120
No log	27.4	274	0.4162	0.6755	0.4162	0.6452
No log	27.6	276	0.4347	0.6392	0.4347	0.6593
No log	27.8	278	0.4059	0.6399	0.4059	0.6371
No log	28.0	280	0.3680	0.6624	0.3680	0.6066
No log	28.2	282	0.4071	0.5907	0.4071	0.6381
No log	28.4	284	0.4716	0.5897	0.4716	0.6867
No log	28.6	286	0.4526	0.5897	0.4526	0.6728
No log	28.8	288	0.3866	0.6587	0.3866	0.6218
No log	29.0	290	0.3807	0.6518	0.3807	0.6170
No log	29.2	292	0.4187	0.6226	0.4187	0.6471
No log	29.4	294	0.4093	0.6227	0.4093	0.6398
No log	29.6	296	0.3864	0.6808	0.3864	0.6216
No log	29.8	298	0.3878	0.6156	0.3878	0.6227
No log	30.0	300	0.3997	0.5580	0.3997	0.6322
No log	30.2	302	0.4011	0.5797	0.4011	0.6333
No log	30.4	304	0.3946	0.5656	0.3946	0.6282
No log	30.6	306	0.4043	0.6724	0.4043	0.6358
No log	30.8	308	0.4118	0.6808	0.4118	0.6417
No log	31.0	310	0.4151	0.6797	0.4151	0.6443
No log	31.2	312	0.4470	0.6394	0.4470	0.6686
No log	31.4	314	0.5186	0.5697	0.5186	0.7201
No log	31.6	316	0.5571	0.5598	0.5571	0.7464
No log	31.8	318	0.5021	0.6025	0.5021	0.7086
No log	32.0	320	0.4492	0.5831	0.4492	0.6702
No log	32.2	322	0.4239	0.5937	0.4239	0.6511
No log	32.4	324	0.4099	0.5379	0.4099	0.6402
No log	32.6	326	0.4065	0.5379	0.4065	0.6376
No log	32.8	328	0.4033	0.5719	0.4033	0.6351
No log	33.0	330	0.4040	0.6530	0.4040	0.6356
No log	33.2	332	0.4061	0.6712	0.4061	0.6373
No log	33.4	334	0.4078	0.6377	0.4078	0.6386
No log	33.6	336	0.4149	0.6087	0.4149	0.6441
No log	33.8	338	0.4118	0.6087	0.4118	0.6417
No log	34.0	340	0.4108	0.7339	0.4108	0.6409
No log	34.2	342	0.4285	0.6846	0.4285	0.6546
No log	34.4	344	0.4474	0.6124	0.4474	0.6689
No log	34.6	346	0.4463	0.6352	0.4463	0.6681
No log	34.8	348	0.4111	0.7212	0.4111	0.6412
No log	35.0	350	0.4071	0.7166	0.4071	0.6380
No log	35.2	352	0.4169	0.7432	0.4169	0.6457
No log	35.4	354	0.4198	0.64	0.4198	0.6479
No log	35.6	356	0.4236	0.6130	0.4236	0.6508
No log	35.8	358	0.4132	0.6377	0.4132	0.6428
No log	36.0	360	0.4187	0.6612	0.4187	0.6471
No log	36.2	362	0.4220	0.6698	0.4220	0.6496
No log	36.4	364	0.4338	0.6411	0.4338	0.6586
No log	36.6	366	0.4393	0.6593	0.4393	0.6628
No log	36.8	368	0.4326	0.5852	0.4326	0.6577
No log	37.0	370	0.4292	0.5715	0.4292	0.6551
No log	37.2	372	0.4229	0.6060	0.4229	0.6503
No log	37.4	374	0.4241	0.5305	0.4241	0.6512
No log	37.6	376	0.4154	0.5324	0.4154	0.6445
No log	37.8	378	0.4018	0.5227	0.4018	0.6339
No log	38.0	380	0.4074	0.6530	0.4074	0.6383
No log	38.2	382	0.4231	0.6873	0.4231	0.6504
No log	38.4	384	0.4245	0.6325	0.4245	0.6516
No log	38.6	386	0.4056	0.7054	0.4056	0.6369
No log	38.8	388	0.3939	0.6919	0.3939	0.6276
No log	39.0	390	0.4123	0.5877	0.4123	0.6421
No log	39.2	392	0.4103	0.5877	0.4103	0.6406
No log	39.4	394	0.4022	0.5633	0.4022	0.6342
No log	39.6	396	0.3958	0.5457	0.3958	0.6291
No log	39.8	398	0.3925	0.6215	0.3925	0.6265
No log	40.0	400	0.3921	0.6215	0.3921	0.6262
No log	40.2	402	0.3921	0.6935	0.3921	0.6262
No log	40.4	404	0.3912	0.5890	0.3912	0.6254
No log	40.6	406	0.3973	0.7247	0.3973	0.6303
No log	40.8	408	0.4372	0.6346	0.4372	0.6612
No log	41.0	410	0.4641	0.6499	0.4641	0.6812
No log	41.2	412	0.4476	0.6333	0.4476	0.6690
No log	41.4	414	0.4090	0.6860	0.4090	0.6395
No log	41.6	416	0.3827	0.6946	0.3827	0.6186
No log	41.8	418	0.3922	0.6317	0.3922	0.6262
No log	42.0	420	0.4048	0.5765	0.4048	0.6362
No log	42.2	422	0.4024	0.5765	0.4024	0.6344
No log	42.4	424	0.3883	0.6197	0.3883	0.6231
No log	42.6	426	0.3860	0.7128	0.3860	0.6213
No log	42.8	428	0.3957	0.6819	0.3957	0.6291
No log	43.0	430	0.4078	0.6616	0.4078	0.6386
No log	43.2	432	0.4295	0.6399	0.4295	0.6554
No log	43.4	434	0.4416	0.6495	0.4416	0.6645
No log	43.6	436	0.4405	0.6495	0.4405	0.6637
No log	43.8	438	0.4313	0.6018	0.4313	0.6567
No log	44.0	440	0.4270	0.5584	0.4270	0.6535
No log	44.2	442	0.4269	0.6200	0.4269	0.6534
No log	44.4	444	0.4369	0.6173	0.4369	0.6610
No log	44.6	446	0.4480	0.5950	0.4480	0.6693
No log	44.8	448	0.4603	0.5975	0.4603	0.6785
No log	45.0	450	0.4467	0.5544	0.4467	0.6684
No log	45.2	452	0.4223	0.6503	0.4223	0.6498
No log	45.4	454	0.4109	0.5986	0.4109	0.6410
No log	45.6	456	0.4156	0.5475	0.4156	0.6447
No log	45.8	458	0.4164	0.5574	0.4164	0.6453
No log	46.0	460	0.4105	0.5574	0.4105	0.6407
No log	46.2	462	0.4075	0.5227	0.4075	0.6384
No log	46.4	464	0.4138	0.6326	0.4138	0.6433
No log	46.6	466	0.4181	0.6530	0.4181	0.6466
No log	46.8	468	0.4181	0.6727	0.4181	0.6466
No log	47.0	470	0.4151	0.6326	0.4151	0.6443
No log	47.2	472	0.4255	0.6611	0.4255	0.6523
No log	47.4	474	0.4495	0.6596	0.4495	0.6704
No log	47.6	476	0.4771	0.5787	0.4771	0.6908
No log	47.8	478	0.4892	0.5787	0.4892	0.6994
No log	48.0	480	0.4804	0.5920	0.4804	0.6931
No log	48.2	482	0.4565	0.6495	0.4565	0.6756
No log	48.4	484	0.4257	0.6503	0.4257	0.6525
No log	48.6	486	0.4207	0.6503	0.4207	0.6486
No log	48.8	488	0.4113	0.5681	0.4113	0.6413
No log	49.0	490	0.4103	0.5152	0.4103	0.6405
No log	49.2	492	0.4110	0.5457	0.4110	0.6411
No log	49.4	494	0.4132	0.5734	0.4132	0.6428
No log	49.6	496	0.4192	0.6024	0.4192	0.6475
No log	49.8	498	0.4276	0.5800	0.4276	0.6539
0.2315	50.0	500	0.4303	0.5800	0.4303	0.6559
0.2315	50.2	502	0.4309	0.5625	0.4309	0.6565
0.2315	50.4	504	0.4348	0.5505	0.4348	0.6594
0.2315	50.6	506	0.4430	0.5662	0.4430	0.6655
0.2315	50.8	508	0.4454	0.5662	0.4454	0.6674
0.2315	51.0	510	0.4424	0.5662	0.4424	0.6651

Framework versions

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