ArabicNewSplits6_FineTuningAraBERTFreeze_run3_AugV5_k1_task1_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: 0.8619
• Qwk: 0.6422
• Mse: 0.8619
• Rmse: 0.9284

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	7.7877	-0.0408	7.7877	2.7906
No log	1.0	4	5.4589	-0.0366	5.4589	2.3364
No log	1.5	6	3.8453	0.0371	3.8453	1.9609
No log	2.0	8	2.8719	0.1111	2.8719	1.6947
No log	2.5	10	2.1515	0.1874	2.1515	1.4668
No log	3.0	12	1.5877	0.1624	1.5877	1.2600
No log	3.5	14	1.2366	0.3191	1.2366	1.1120
No log	4.0	16	1.1142	0.2882	1.1142	1.0556
No log	4.5	18	1.0901	0.1961	1.0901	1.0441
No log	5.0	20	1.0637	0.2332	1.0637	1.0314
No log	5.5	22	1.0383	0.2759	1.0383	1.0189
No log	6.0	24	1.0128	0.3334	1.0128	1.0064
No log	6.5	26	1.0300	0.3873	1.0300	1.0149
No log	7.0	28	0.9954	0.4311	0.9954	0.9977
No log	7.5	30	0.9057	0.4590	0.9057	0.9517
No log	8.0	32	0.8513	0.5180	0.8513	0.9227
No log	8.5	34	0.8327	0.5662	0.8327	0.9125
No log	9.0	36	0.7915	0.5443	0.7915	0.8897
No log	9.5	38	0.7592	0.5488	0.7592	0.8713
No log	10.0	40	0.8160	0.5215	0.8160	0.9033
No log	10.5	42	0.8628	0.5291	0.8628	0.9289
No log	11.0	44	0.7989	0.5373	0.7989	0.8938
No log	11.5	46	0.7222	0.5893	0.7222	0.8498
No log	12.0	48	0.7151	0.6427	0.7151	0.8456
No log	12.5	50	0.7216	0.6299	0.7216	0.8494
No log	13.0	52	0.7018	0.6225	0.7018	0.8377
No log	13.5	54	0.6812	0.6397	0.6812	0.8254
No log	14.0	56	0.6981	0.6259	0.6981	0.8356
No log	14.5	58	0.6987	0.6366	0.6987	0.8359
No log	15.0	60	0.7002	0.6496	0.7002	0.8368
No log	15.5	62	0.6868	0.6430	0.6868	0.8288
No log	16.0	64	0.6828	0.6595	0.6828	0.8263
No log	16.5	66	0.6892	0.6587	0.6892	0.8302
No log	17.0	68	0.6882	0.6934	0.6882	0.8296
No log	17.5	70	0.6832	0.6790	0.6832	0.8265
No log	18.0	72	0.6827	0.6824	0.6827	0.8263
No log	18.5	74	0.6934	0.6890	0.6934	0.8327
No log	19.0	76	0.7049	0.6500	0.7049	0.8396
No log	19.5	78	0.7224	0.6919	0.7224	0.8500
No log	20.0	80	0.7521	0.6984	0.7521	0.8672
No log	20.5	82	0.7854	0.6905	0.7854	0.8862
No log	21.0	84	0.7788	0.6740	0.7788	0.8825
No log	21.5	86	0.7888	0.6984	0.7888	0.8882
No log	22.0	88	0.7833	0.7022	0.7833	0.8850
No log	22.5	90	0.7612	0.6716	0.7612	0.8725
No log	23.0	92	0.7527	0.6678	0.7527	0.8676
No log	23.5	94	0.7454	0.6678	0.7454	0.8634
No log	24.0	96	0.7413	0.6833	0.7413	0.8610
No log	24.5	98	0.7433	0.6999	0.7433	0.8621
No log	25.0	100	0.7571	0.7087	0.7571	0.8701
No log	25.5	102	0.7936	0.6957	0.7936	0.8908
No log	26.0	104	0.8128	0.6836	0.8128	0.9015
No log	26.5	106	0.8045	0.6620	0.8045	0.8970
No log	27.0	108	0.8241	0.6280	0.8241	0.9078
No log	27.5	110	0.8195	0.6420	0.8195	0.9053
No log	28.0	112	0.8048	0.6582	0.8048	0.8971
No log	28.5	114	0.8082	0.6817	0.8082	0.8990
No log	29.0	116	0.7969	0.7076	0.7969	0.8927
No log	29.5	118	0.7667	0.6772	0.7667	0.8756
No log	30.0	120	0.7517	0.6723	0.7517	0.8670
No log	30.5	122	0.7532	0.6941	0.7532	0.8679
No log	31.0	124	0.7676	0.6756	0.7676	0.8761
No log	31.5	126	0.8067	0.6714	0.8067	0.8982
No log	32.0	128	0.8354	0.6446	0.8354	0.9140
No log	32.5	130	0.8188	0.6522	0.8188	0.9049
No log	33.0	132	0.8261	0.6681	0.8261	0.9089
No log	33.5	134	0.8582	0.6332	0.8582	0.9264
No log	34.0	136	0.8335	0.6594	0.8335	0.9130
No log	34.5	138	0.7855	0.6699	0.7855	0.8863
No log	35.0	140	0.7608	0.6567	0.7608	0.8722
No log	35.5	142	0.7567	0.6666	0.7567	0.8699
No log	36.0	144	0.7669	0.6866	0.7669	0.8757
No log	36.5	146	0.7901	0.6702	0.7901	0.8889
No log	37.0	148	0.8037	0.6666	0.8037	0.8965
No log	37.5	150	0.7800	0.6767	0.7800	0.8832
No log	38.0	152	0.7708	0.6760	0.7708	0.8779
No log	38.5	154	0.7802	0.6650	0.7802	0.8833
No log	39.0	156	0.8014	0.6729	0.8014	0.8952
No log	39.5	158	0.8393	0.6252	0.8393	0.9162
No log	40.0	160	0.9020	0.6432	0.9020	0.9497
No log	40.5	162	0.9673	0.6102	0.9673	0.9835
No log	41.0	164	0.9658	0.6102	0.9658	0.9827
No log	41.5	166	0.9069	0.6319	0.9069	0.9523
No log	42.0	168	0.8314	0.6438	0.8314	0.9118
No log	42.5	170	0.7998	0.6684	0.7998	0.8943
No log	43.0	172	0.7979	0.6565	0.7979	0.8933
No log	43.5	174	0.8322	0.6446	0.8322	0.9123
No log	44.0	176	0.8495	0.6594	0.8495	0.9217
No log	44.5	178	0.8260	0.6446	0.8260	0.9088
No log	45.0	180	0.8010	0.6494	0.8010	0.8950
No log	45.5	182	0.8164	0.6379	0.8164	0.9036
No log	46.0	184	0.8428	0.6632	0.8428	0.9181
No log	46.5	186	0.8683	0.6558	0.8683	0.9318
No log	47.0	188	0.8708	0.6407	0.8708	0.9331
No log	47.5	190	0.8427	0.6491	0.8427	0.9180
No log	48.0	192	0.8261	0.6534	0.8261	0.9089
No log	48.5	194	0.8245	0.6534	0.8245	0.9080
No log	49.0	196	0.8407	0.6527	0.8407	0.9169
No log	49.5	198	0.8605	0.6666	0.8605	0.9277
No log	50.0	200	0.8836	0.6607	0.8836	0.9400
No log	50.5	202	0.8618	0.6603	0.8618	0.9283
No log	51.0	204	0.8410	0.6674	0.8410	0.9170
No log	51.5	206	0.8332	0.6531	0.8332	0.9128
No log	52.0	208	0.8328	0.6606	0.8328	0.9126
No log	52.5	210	0.8453	0.6649	0.8453	0.9194
No log	53.0	212	0.8647	0.6526	0.8647	0.9299
No log	53.5	214	0.8841	0.6589	0.8841	0.9403
No log	54.0	216	0.9081	0.6690	0.9081	0.9529
No log	54.5	218	0.9239	0.6446	0.9239	0.9612
No log	55.0	220	0.9267	0.6411	0.9267	0.9627
No log	55.5	222	0.9255	0.6332	0.9255	0.9620
No log	56.0	224	0.9033	0.6332	0.9033	0.9504
No log	56.5	226	0.8608	0.6520	0.8608	0.9278
No log	57.0	228	0.8327	0.6551	0.8327	0.9125
No log	57.5	230	0.8305	0.6708	0.8305	0.9113
No log	58.0	232	0.8433	0.6850	0.8433	0.9183
No log	58.5	234	0.8643	0.6761	0.8643	0.9297
No log	59.0	236	0.8663	0.6857	0.8663	0.9307
No log	59.5	238	0.8726	0.6674	0.8726	0.9341
No log	60.0	240	0.8921	0.6569	0.8921	0.9445
No log	60.5	242	0.8922	0.6534	0.8922	0.9446
No log	61.0	244	0.8904	0.6543	0.8904	0.9436
No log	61.5	246	0.8776	0.6578	0.8776	0.9368
No log	62.0	248	0.8512	0.6788	0.8512	0.9226
No log	62.5	250	0.8391	0.6788	0.8391	0.9160
No log	63.0	252	0.8211	0.6769	0.8211	0.9062
No log	63.5	254	0.8159	0.6769	0.8159	0.9032
No log	64.0	256	0.8336	0.6532	0.8336	0.9130
No log	64.5	258	0.8613	0.6508	0.8613	0.9281
No log	65.0	260	0.8889	0.6426	0.8889	0.9428
No log	65.5	262	0.9086	0.6426	0.9086	0.9532
No log	66.0	264	0.9193	0.6255	0.9193	0.9588
No log	66.5	266	0.8973	0.6297	0.8973	0.9473
No log	67.0	268	0.8718	0.6454	0.8718	0.9337
No log	67.5	270	0.8574	0.6380	0.8574	0.9260
No log	68.0	272	0.8536	0.6499	0.8536	0.9239
No log	68.5	274	0.8508	0.6499	0.8508	0.9224
No log	69.0	276	0.8701	0.6422	0.8701	0.9328
No log	69.5	278	0.8963	0.6413	0.8963	0.9468
No log	70.0	280	0.9273	0.6426	0.9273	0.9630
No log	70.5	282	0.9460	0.6237	0.9460	0.9726
No log	71.0	284	0.9586	0.6251	0.9586	0.9791
No log	71.5	286	0.9563	0.6309	0.9563	0.9779
No log	72.0	288	0.9364	0.6237	0.9364	0.9677
No log	72.5	290	0.9144	0.6354	0.9144	0.9563
No log	73.0	292	0.8830	0.6378	0.8830	0.9397
No log	73.5	294	0.8635	0.6454	0.8635	0.9292
No log	74.0	296	0.8495	0.6523	0.8495	0.9217
No log	74.5	298	0.8563	0.6654	0.8563	0.9253
No log	75.0	300	0.8804	0.6422	0.8804	0.9383
No log	75.5	302	0.9021	0.6297	0.9021	0.9498
No log	76.0	304	0.9107	0.6298	0.9107	0.9543
No log	76.5	306	0.9107	0.6298	0.9107	0.9543
No log	77.0	308	0.9117	0.6298	0.9117	0.9548
No log	77.5	310	0.8927	0.6426	0.8927	0.9448
No log	78.0	312	0.8665	0.6457	0.8665	0.9308
No log	78.5	314	0.8431	0.6516	0.8431	0.9182
No log	79.0	316	0.8288	0.6520	0.8288	0.9104
No log	79.5	318	0.8180	0.6681	0.8180	0.9044
No log	80.0	320	0.8087	0.6616	0.8087	0.8993
No log	80.5	322	0.8071	0.6736	0.8071	0.8984
No log	81.0	324	0.8141	0.6556	0.8141	0.9023
No log	81.5	326	0.8294	0.6556	0.8294	0.9107
No log	82.0	328	0.8492	0.6430	0.8492	0.9215
No log	82.5	330	0.8659	0.6587	0.8659	0.9306
No log	83.0	332	0.8750	0.6580	0.8750	0.9354
No log	83.5	334	0.8835	0.6520	0.8835	0.9399
No log	84.0	336	0.8917	0.6426	0.8917	0.9443
No log	84.5	338	0.8895	0.6426	0.8895	0.9432
No log	85.0	340	0.8869	0.6426	0.8869	0.9417
No log	85.5	342	0.8851	0.6426	0.8851	0.9408
No log	86.0	344	0.8866	0.6426	0.8866	0.9416
No log	86.5	346	0.8874	0.6426	0.8874	0.9420
No log	87.0	348	0.8833	0.6297	0.8833	0.9398
No log	87.5	350	0.8846	0.6297	0.8846	0.9405
No log	88.0	352	0.8887	0.6297	0.8887	0.9427
No log	88.5	354	0.8848	0.6341	0.8848	0.9406
No log	89.0	356	0.8811	0.6341	0.8811	0.9387
No log	89.5	358	0.8843	0.6341	0.8843	0.9404
No log	90.0	360	0.8878	0.6312	0.8878	0.9423
No log	90.5	362	0.8949	0.6255	0.8949	0.9460
No log	91.0	364	0.8940	0.6255	0.8940	0.9455
No log	91.5	366	0.8933	0.6255	0.8933	0.9451
No log	92.0	368	0.8889	0.6112	0.8889	0.9428
No log	92.5	370	0.8825	0.6392	0.8825	0.9394
No log	93.0	372	0.8773	0.6436	0.8773	0.9366
No log	93.5	374	0.8725	0.6422	0.8725	0.9341
No log	94.0	376	0.8657	0.6422	0.8657	0.9304
No log	94.5	378	0.8624	0.6422	0.8624	0.9286
No log	95.0	380	0.8588	0.6422	0.8588	0.9267
No log	95.5	382	0.8580	0.6356	0.8580	0.9263
No log	96.0	384	0.8559	0.6356	0.8559	0.9251
No log	96.5	386	0.8548	0.6356	0.8548	0.9246
No log	97.0	388	0.8559	0.6356	0.8559	0.9252
No log	97.5	390	0.8560	0.6356	0.8560	0.9252
No log	98.0	392	0.8569	0.6356	0.8569	0.9257
No log	98.5	394	0.8587	0.6356	0.8587	0.9267
No log	99.0	396	0.8602	0.6356	0.8602	0.9275
No log	99.5	398	0.8614	0.6422	0.8614	0.9281
No log	100.0	400	0.8619	0.6422	0.8619	0.9284

Framework versions

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