ArabicNewSplits6_FineTuningAraBERTFreeze_run2_AugV5_k1_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: 0.9130
• Qwk: 0.4878
• Mse: 0.9130
• Rmse: 0.9555

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: 16
• eval_batch_size: 16
• 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	0.5	2	6.3720	-0.0278	6.3720	2.5243
No log	1.0	4	4.1663	-0.0186	4.1663	2.0412
No log	1.5	6	2.8586	0.0180	2.8586	1.6908
No log	2.0	8	2.0237	0.1231	2.0237	1.4226
No log	2.5	10	1.3556	0.1257	1.3556	1.1643
No log	3.0	12	0.9069	0.1147	0.9069	0.9523
No log	3.5	14	0.6983	0.2123	0.6983	0.8356
No log	4.0	16	0.6593	0.2424	0.6593	0.8120
No log	4.5	18	0.6943	0.2196	0.6943	0.8333
No log	5.0	20	0.6991	0.1918	0.6991	0.8361
No log	5.5	22	0.6589	0.2470	0.6589	0.8117
No log	6.0	24	0.6454	0.3461	0.6454	0.8033
No log	6.5	26	0.6382	0.3555	0.6382	0.7989
No log	7.0	28	0.6260	0.3346	0.6260	0.7912
No log	7.5	30	0.6277	0.2985	0.6277	0.7923
No log	8.0	32	0.5905	0.3674	0.5905	0.7684
No log	8.5	34	0.5696	0.4074	0.5696	0.7547
No log	9.0	36	0.5987	0.3301	0.5987	0.7737
No log	9.5	38	0.7052	0.3327	0.7052	0.8398
No log	10.0	40	0.7584	0.3537	0.7584	0.8708
No log	10.5	42	0.7086	0.3520	0.7086	0.8418
No log	11.0	44	0.6441	0.3566	0.6441	0.8026
No log	11.5	46	0.6386	0.4324	0.6386	0.7991
No log	12.0	48	0.6581	0.4742	0.6581	0.8112
No log	12.5	50	0.6831	0.4574	0.6831	0.8265
No log	13.0	52	0.6978	0.4561	0.6978	0.8354
No log	13.5	54	0.7407	0.4432	0.7407	0.8606
No log	14.0	56	0.7822	0.4292	0.7822	0.8844
No log	14.5	58	0.8365	0.4469	0.8365	0.9146
No log	15.0	60	0.8042	0.4634	0.8042	0.8968
No log	15.5	62	0.7491	0.4773	0.7491	0.8655
No log	16.0	64	0.7156	0.4632	0.7156	0.8459
No log	16.5	66	0.7185	0.4657	0.7185	0.8476
No log	17.0	68	0.7293	0.4657	0.7293	0.8540
No log	17.5	70	0.7428	0.4887	0.7428	0.8619
No log	18.0	72	0.7458	0.4886	0.7458	0.8636
No log	18.5	74	0.7490	0.5049	0.7490	0.8654
No log	19.0	76	0.7588	0.4739	0.7588	0.8711
No log	19.5	78	0.7644	0.4978	0.7644	0.8743
No log	20.0	80	0.7727	0.4739	0.7727	0.8790
No log	20.5	82	0.7996	0.4823	0.7996	0.8942
No log	21.0	84	0.8139	0.4940	0.8139	0.9021
No log	21.5	86	0.8087	0.5106	0.8087	0.8993
No log	22.0	88	0.7923	0.4953	0.7923	0.8901
No log	22.5	90	0.7693	0.4987	0.7693	0.8771
No log	23.0	92	0.7725	0.4883	0.7725	0.8789
No log	23.5	94	0.7753	0.4987	0.7753	0.8805
No log	24.0	96	0.7848	0.5265	0.7848	0.8859
No log	24.5	98	0.7979	0.4660	0.7979	0.8933
No log	25.0	100	0.8064	0.4585	0.8064	0.8980
No log	25.5	102	0.7972	0.4892	0.7972	0.8928
No log	26.0	104	0.8225	0.4948	0.8225	0.9069
No log	26.5	106	0.8413	0.4977	0.8413	0.9172
No log	27.0	108	0.8322	0.4949	0.8322	0.9122
No log	27.5	110	0.8144	0.5039	0.8144	0.9024
No log	28.0	112	0.8194	0.5144	0.8194	0.9052
No log	28.5	114	0.8215	0.5144	0.8215	0.9063
No log	29.0	116	0.8278	0.5077	0.8278	0.9098
No log	29.5	118	0.8461	0.5140	0.8461	0.9198
No log	30.0	120	0.8676	0.4628	0.8676	0.9315
No log	30.5	122	0.8738	0.4505	0.8738	0.9348
No log	31.0	124	0.8618	0.4846	0.8618	0.9283
No log	31.5	126	0.8771	0.5	0.8771	0.9365
No log	32.0	128	0.8888	0.4992	0.8888	0.9428
No log	32.5	130	0.8761	0.4992	0.8761	0.9360
No log	33.0	132	0.8517	0.4366	0.8517	0.9229
No log	33.5	134	0.8631	0.4602	0.8631	0.9291
No log	34.0	136	0.9087	0.4754	0.9087	0.9533
No log	34.5	138	0.8870	0.4618	0.8870	0.9418
No log	35.0	140	0.8382	0.4818	0.8382	0.9156
No log	35.5	142	0.8309	0.4751	0.8309	0.9115
No log	36.0	144	0.8693	0.4918	0.8693	0.9323
No log	36.5	146	0.8819	0.4918	0.8819	0.9391
No log	37.0	148	0.8663	0.4964	0.8663	0.9307
No log	37.5	150	0.8557	0.4831	0.8557	0.9251
No log	38.0	152	0.8644	0.4964	0.8644	0.9298
No log	38.5	154	0.8552	0.4969	0.8552	0.9248
No log	39.0	156	0.8608	0.4616	0.8608	0.9278
No log	39.5	158	0.9010	0.4867	0.9010	0.9492
No log	40.0	160	0.9286	0.4762	0.9286	0.9636
No log	40.5	162	0.9059	0.4882	0.9059	0.9518
No log	41.0	164	0.8804	0.4964	0.8804	0.9383
No log	41.5	166	0.8801	0.4668	0.8801	0.9381
No log	42.0	168	0.8829	0.4508	0.8829	0.9396
No log	42.5	170	0.8816	0.4444	0.8816	0.9389
No log	43.0	172	0.8835	0.4704	0.8835	0.9399
No log	43.5	174	0.8866	0.4688	0.8866	0.9416
No log	44.0	176	0.8910	0.4752	0.8910	0.9439
No log	44.5	178	0.8929	0.4706	0.8929	0.9450
No log	45.0	180	0.8936	0.4856	0.8936	0.9453
No log	45.5	182	0.8980	0.5224	0.8980	0.9476
No log	46.0	184	0.9058	0.4809	0.9058	0.9518
No log	46.5	186	0.9023	0.4763	0.9023	0.9499
No log	47.0	188	0.9082	0.4739	0.9082	0.9530
No log	47.5	190	0.8998	0.4778	0.8998	0.9486
No log	48.0	192	0.8994	0.4932	0.8994	0.9484
No log	48.5	194	0.8971	0.4924	0.8971	0.9471
No log	49.0	196	0.8867	0.4924	0.8867	0.9416
No log	49.5	198	0.8826	0.5023	0.8826	0.9395
No log	50.0	200	0.8766	0.4830	0.8766	0.9363
No log	50.5	202	0.8707	0.4882	0.8707	0.9331
No log	51.0	204	0.8566	0.4949	0.8566	0.9255
No log	51.5	206	0.8562	0.4949	0.8562	0.9253
No log	52.0	208	0.8665	0.4882	0.8665	0.9309
No log	52.5	210	0.8630	0.4973	0.8630	0.9290
No log	53.0	212	0.8618	0.4798	0.8618	0.9283
No log	53.5	214	0.8652	0.5169	0.8652	0.9302
No log	54.0	216	0.8824	0.5015	0.8824	0.9394
No log	54.5	218	0.9168	0.4971	0.9168	0.9575
No log	55.0	220	0.9495	0.4934	0.9495	0.9744
No log	55.5	222	0.9310	0.4873	0.9310	0.9649
No log	56.0	224	0.8857	0.4986	0.8857	0.9411
No log	56.5	226	0.8784	0.5070	0.8784	0.9372
No log	57.0	228	0.8657	0.5072	0.8657	0.9304
No log	57.5	230	0.8675	0.4922	0.8675	0.9314
No log	58.0	232	0.8747	0.4922	0.8747	0.9352
No log	58.5	234	0.8918	0.5010	0.8918	0.9444
No log	59.0	236	0.9199	0.5	0.9199	0.9591
No log	59.5	238	0.9346	0.5	0.9346	0.9667
No log	60.0	240	0.9282	0.5007	0.9282	0.9634
No log	60.5	242	0.9158	0.4968	0.9158	0.9570
No log	61.0	244	0.9184	0.4755	0.9184	0.9583
No log	61.5	246	0.9245	0.4792	0.9245	0.9615
No log	62.0	248	0.9395	0.4967	0.9395	0.9693
No log	62.5	250	0.9404	0.4967	0.9404	0.9697
No log	63.0	252	0.9353	0.4967	0.9353	0.9671
No log	63.5	254	0.9248	0.4943	0.9248	0.9617
No log	64.0	256	0.9125	0.4905	0.9125	0.9552
No log	64.5	258	0.9134	0.5269	0.9134	0.9557
No log	65.0	260	0.9109	0.4919	0.9109	0.9544
No log	65.5	262	0.9107	0.4840	0.9107	0.9543
No log	66.0	264	0.9233	0.5073	0.9233	0.9609
No log	66.5	266	0.9427	0.4904	0.9427	0.9709
No log	67.0	268	0.9411	0.4904	0.9411	0.9701
No log	67.5	270	0.9214	0.4933	0.9214	0.9599
No log	68.0	272	0.9071	0.5082	0.9071	0.9524
No log	68.5	274	0.8975	0.4760	0.8975	0.9474
No log	69.0	276	0.8917	0.4827	0.8917	0.9443
No log	69.5	278	0.8918	0.4889	0.8918	0.9444
No log	70.0	280	0.8938	0.4827	0.8938	0.9454
No log	70.5	282	0.8970	0.4755	0.8970	0.9471
No log	71.0	284	0.9018	0.4896	0.9018	0.9496
No log	71.5	286	0.9145	0.5181	0.9145	0.9563
No log	72.0	288	0.9313	0.5040	0.9313	0.9650
No log	72.5	290	0.9372	0.4972	0.9372	0.9681
No log	73.0	292	0.9527	0.4890	0.9527	0.9761
No log	73.5	294	0.9601	0.4829	0.9601	0.9798
No log	74.0	296	0.9526	0.5005	0.9526	0.9760
No log	74.5	298	0.9338	0.4925	0.9338	0.9663
No log	75.0	300	0.9108	0.5186	0.9108	0.9544
No log	75.5	302	0.9023	0.4988	0.9023	0.9499
No log	76.0	304	0.9053	0.4981	0.9053	0.9515
No log	76.5	306	0.9113	0.5007	0.9113	0.9546
No log	77.0	308	0.9298	0.4925	0.9298	0.9643
No log	77.5	310	0.9420	0.4864	0.9420	0.9706
No log	78.0	312	0.9440	0.4904	0.9440	0.9716
No log	78.5	314	0.9474	0.4951	0.9474	0.9733
No log	79.0	316	0.9417	0.4904	0.9417	0.9704
No log	79.5	318	0.9262	0.4858	0.9262	0.9624
No log	80.0	320	0.9055	0.4785	0.9055	0.9516
No log	80.5	322	0.8985	0.4832	0.8985	0.9479
No log	81.0	324	0.8979	0.4896	0.8979	0.9476
No log	81.5	326	0.9008	0.4905	0.9008	0.9491
No log	82.0	328	0.9040	0.4905	0.9040	0.9508
No log	82.5	330	0.9047	0.4905	0.9047	0.9512
No log	83.0	332	0.9074	0.4842	0.9074	0.9526
No log	83.5	334	0.9139	0.4863	0.9139	0.9560
No log	84.0	336	0.9254	0.4933	0.9254	0.9620
No log	84.5	338	0.9324	0.4872	0.9324	0.9656
No log	85.0	340	0.9386	0.4872	0.9386	0.9688
No log	85.5	342	0.9382	0.4872	0.9382	0.9686
No log	86.0	344	0.9300	0.4946	0.9300	0.9644
No log	86.5	346	0.9250	0.4967	0.9250	0.9617
No log	87.0	348	0.9201	0.4940	0.9201	0.9592
No log	87.5	350	0.9166	0.4940	0.9166	0.9574
No log	88.0	352	0.9134	0.5092	0.9134	0.9557
No log	88.5	354	0.9122	0.5092	0.9122	0.9551
No log	89.0	356	0.9106	0.5084	0.9106	0.9543
No log	89.5	358	0.9124	0.4947	0.9124	0.9552
No log	90.0	360	0.9182	0.4933	0.9182	0.9582
No log	90.5	362	0.9247	0.4933	0.9247	0.9616
No log	91.0	364	0.9261	0.4933	0.9261	0.9623
No log	91.5	366	0.9284	0.4926	0.9284	0.9635
No log	92.0	368	0.9271	0.4933	0.9271	0.9629
No log	92.5	370	0.9259	0.4933	0.9259	0.9623
No log	93.0	372	0.9239	0.4980	0.9239	0.9612
No log	93.5	374	0.9210	0.5039	0.9210	0.9597
No log	94.0	376	0.9193	0.5151	0.9193	0.9588
No log	94.5	378	0.9180	0.5108	0.9180	0.9581
No log	95.0	380	0.9155	0.5076	0.9155	0.9568
No log	95.5	382	0.9130	0.4869	0.9130	0.9555
No log	96.0	384	0.9108	0.4836	0.9108	0.9544
No log	96.5	386	0.9105	0.4788	0.9105	0.9542
No log	97.0	388	0.9103	0.4893	0.9103	0.9541
No log	97.5	390	0.9104	0.4893	0.9104	0.9541
No log	98.0	392	0.9112	0.4893	0.9112	0.9545
No log	98.5	394	0.9118	0.4940	0.9118	0.9549
No log	99.0	396	0.9126	0.4878	0.9126	0.9553
No log	99.5	398	0.9129	0.4878	0.9129	0.9555
No log	100.0	400	0.9130	0.4878	0.9130	0.9555

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.2.0
• Tokenizers 0.19.1