ArabicNewSplits7_FineTuningAraBERT_run3_AugV5_k2_task3_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: 0.6977
• Qwk: 0.1769
• Mse: 0.6977
• Rmse: 0.8353

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	4.2515	0.0000	4.2515	2.0619
No log	0.5	4	1.9737	-0.0284	1.9737	1.4049
No log	0.75	6	1.2406	0.0279	1.2406	1.1138
No log	1.0	8	0.7407	0.0296	0.7407	0.8606
No log	1.25	10	0.7211	-0.0101	0.7211	0.8492
No log	1.5	12	0.7348	0.0909	0.7348	0.8572
No log	1.75	14	0.9060	0.1763	0.9060	0.9518
No log	2.0	16	0.8842	0.1577	0.8842	0.9403
No log	2.25	18	0.9739	0.1025	0.9739	0.9869
No log	2.5	20	0.7557	0.0129	0.7557	0.8693
No log	2.75	22	0.7612	0.0159	0.7612	0.8725
No log	3.0	24	1.0109	-0.0178	1.0109	1.0055
No log	3.25	26	1.1156	-0.0247	1.1156	1.0562
No log	3.5	28	1.0309	0.0100	1.0309	1.0153
No log	3.75	30	0.7548	0.1097	0.7548	0.8688
No log	4.0	32	0.7626	0.1097	0.7626	0.8733
No log	4.25	34	0.9160	0.0680	0.9160	0.9571
No log	4.5	36	0.8156	-0.0518	0.8156	0.9031
No log	4.75	38	0.6737	0.1828	0.6737	0.8208
No log	5.0	40	0.6937	0.1097	0.6937	0.8329
No log	5.25	42	1.0039	-0.0966	1.0039	1.0020
No log	5.5	44	0.8182	0.1107	0.8182	0.9046
No log	5.75	46	0.7110	0.2304	0.7110	0.8432
No log	6.0	48	0.6900	0.2209	0.6900	0.8307
No log	6.25	50	0.8970	0.0587	0.8970	0.9471
No log	6.5	52	0.7513	0.1903	0.7513	0.8668
No log	6.75	54	0.7781	0.1829	0.7781	0.8821
No log	7.0	56	0.7929	0.1807	0.7929	0.8904
No log	7.25	58	0.8077	0.1818	0.8077	0.8987
No log	7.5	60	0.9624	0.1281	0.9624	0.9810
No log	7.75	62	0.9185	0.1770	0.9185	0.9584
No log	8.0	64	0.9379	0.1850	0.9379	0.9685
No log	8.25	66	0.9327	0.1504	0.9327	0.9658
No log	8.5	68	0.7700	0.2325	0.7700	0.8775
No log	8.75	70	0.7874	0.2115	0.7874	0.8874
No log	9.0	72	0.7178	0.1943	0.7178	0.8472
No log	9.25	74	1.5187	0.1067	1.5187	1.2324
No log	9.5	76	1.3784	0.0887	1.3784	1.1741
No log	9.75	78	0.7103	0.1047	0.7103	0.8428
No log	10.0	80	0.8823	0.0233	0.8823	0.9393
No log	10.25	82	0.9958	0.0865	0.9958	0.9979
No log	10.5	84	0.7641	0.2961	0.7641	0.8741
No log	10.75	86	1.0972	0.1353	1.0972	1.0475
No log	11.0	88	1.1376	0.0440	1.1376	1.0666
No log	11.25	90	0.7967	0.1983	0.7967	0.8926
No log	11.5	92	0.8788	0.1586	0.8788	0.9374
No log	11.75	94	0.8221	0.1605	0.8221	0.9067
No log	12.0	96	0.7932	0.0538	0.7932	0.8906
No log	12.25	98	0.9070	0.0287	0.9070	0.9524
No log	12.5	100	0.7381	0.1097	0.7381	0.8591
No log	12.75	102	0.6769	0.1815	0.6769	0.8228
No log	13.0	104	0.6621	0.2271	0.6621	0.8137
No log	13.25	106	1.0443	0.0355	1.0443	1.0219
No log	13.5	108	1.4320	0.0881	1.4320	1.1967
No log	13.75	110	1.0084	0.1771	1.0084	1.0042
No log	14.0	112	0.7062	0.2899	0.7062	0.8404
No log	14.25	114	0.8244	0.1738	0.8244	0.9080
No log	14.5	116	0.7183	0.2486	0.7183	0.8475
No log	14.75	118	0.8732	0.1625	0.8732	0.9345
No log	15.0	120	1.2199	0.0695	1.2199	1.1045
No log	15.25	122	0.9635	0.2119	0.9635	0.9816
No log	15.5	124	0.6874	0.1047	0.6874	0.8291
No log	15.75	126	0.6518	0.1304	0.6518	0.8073
No log	16.0	128	0.6697	0.1675	0.6697	0.8183
No log	16.25	130	0.8507	0.0988	0.8507	0.9223
No log	16.5	132	0.9023	-0.0138	0.9023	0.9499
No log	16.75	134	0.7117	0.1599	0.7117	0.8436
No log	17.0	136	0.7177	0.2515	0.7177	0.8472
No log	17.25	138	0.7094	0.2005	0.7094	0.8422
No log	17.5	140	0.7732	0.1449	0.7732	0.8793
No log	17.75	142	0.8870	0.1025	0.8870	0.9418
No log	18.0	144	0.9035	0.0618	0.9035	0.9505
No log	18.25	146	0.7641	0.1449	0.7641	0.8741
No log	18.5	148	0.6808	0.2070	0.6808	0.8251
No log	18.75	150	0.7299	0.3331	0.7299	0.8543
No log	19.0	152	0.7069	0.2627	0.7069	0.8408
No log	19.25	154	0.7635	0.1440	0.7635	0.8738
No log	19.5	156	0.8737	0.0986	0.8737	0.9347
No log	19.75	158	0.8992	0.0556	0.8992	0.9482
No log	20.0	160	0.6874	0.1199	0.6874	0.8291
No log	20.25	162	0.6657	0.2747	0.6657	0.8159
No log	20.5	164	0.6780	0.1928	0.6780	0.8234
No log	20.75	166	0.6388	0.1740	0.6388	0.7992
No log	21.0	168	0.9742	-0.0301	0.9742	0.9870
No log	21.25	170	1.2251	0.0991	1.2251	1.1069
No log	21.5	172	0.9810	-0.0291	0.9810	0.9904
No log	21.75	174	0.7015	0.2166	0.7015	0.8375
No log	22.0	176	0.7051	0.2087	0.7051	0.8397
No log	22.25	178	0.7473	0.1599	0.7473	0.8645
No log	22.5	180	0.9468	0.0379	0.9468	0.9731
No log	22.75	182	0.8698	0.1360	0.8698	0.9326
No log	23.0	184	0.7128	0.1599	0.7128	0.8443
No log	23.25	186	0.6598	0.2550	0.6598	0.8123
No log	23.5	188	0.6591	0.2550	0.6591	0.8118
No log	23.75	190	0.7353	0.1485	0.7353	0.8575
No log	24.0	192	1.0324	0.0855	1.0324	1.0161
No log	24.25	194	1.0877	0.0726	1.0877	1.0429
No log	24.5	196	0.8110	0.2011	0.8110	0.9005
No log	24.75	198	0.7348	0.1846	0.7348	0.8572
No log	25.0	200	0.7385	0.2345	0.7385	0.8593
No log	25.25	202	0.7610	0.2572	0.7610	0.8723
No log	25.5	204	0.8155	0.1727	0.8155	0.9030
No log	25.75	206	0.7971	0.1783	0.7971	0.8928
No log	26.0	208	0.8026	0.1783	0.8026	0.8959
No log	26.25	210	0.7323	0.2243	0.7323	0.8558
No log	26.5	212	0.7225	0.2243	0.7225	0.8500
No log	26.75	214	0.7448	0.1431	0.7448	0.8630
No log	27.0	216	0.8563	-0.0101	0.8563	0.9254
No log	27.25	218	0.8431	-0.0079	0.8431	0.9182
No log	27.5	220	0.7169	0.1199	0.7169	0.8467
No log	27.75	222	0.7057	0.1807	0.7057	0.8401
No log	28.0	224	0.7311	0.1807	0.7311	0.8551
No log	28.25	226	0.7864	0.1553	0.7864	0.8868
No log	28.5	228	0.8654	0.0837	0.8654	0.9302
No log	28.75	230	0.8074	0.1986	0.8074	0.8985
No log	29.0	232	0.7822	0.2087	0.7822	0.8844
No log	29.25	234	0.8029	0.2437	0.8029	0.8961
No log	29.5	236	0.9352	0.1065	0.9352	0.9671
No log	29.75	238	0.9342	0.1027	0.9342	0.9665
No log	30.0	240	0.8303	0.0504	0.8303	0.9112
No log	30.25	242	0.8323	0.0504	0.8323	0.9123
No log	30.5	244	0.7465	0.1553	0.7465	0.8640
No log	30.75	246	0.7234	0.1254	0.7234	0.8505
No log	31.0	248	0.7571	0.0600	0.7571	0.8701
No log	31.25	250	0.7552	0.1506	0.7552	0.8690
No log	31.5	252	0.7280	0.1096	0.7280	0.8532
No log	31.75	254	0.7391	0.1495	0.7391	0.8597
No log	32.0	256	0.7475	0.1146	0.7475	0.8646
No log	32.25	258	0.8085	0.1440	0.8085	0.8991
No log	32.5	260	0.8555	0.0438	0.8555	0.9249
No log	32.75	262	0.9047	-0.0008	0.9047	0.9512
No log	33.0	264	0.8018	0.1440	0.8018	0.8954
No log	33.25	266	0.7367	0.2138	0.7367	0.8583
No log	33.5	268	0.7230	0.1196	0.7230	0.8503
No log	33.75	270	0.7218	0.1146	0.7218	0.8496
No log	34.0	272	0.8414	0.1196	0.8414	0.9173
No log	34.25	274	0.8940	0.0984	0.8940	0.9455
No log	34.5	276	0.7783	0.1836	0.7783	0.8822
No log	34.75	278	0.7246	0.1199	0.7246	0.8512
No log	35.0	280	0.7329	0.1199	0.7329	0.8561
No log	35.25	282	0.8316	0.1817	0.8316	0.9119
No log	35.5	284	0.8692	0.0826	0.8692	0.9323
No log	35.75	286	0.8378	0.1817	0.8378	0.9153
No log	36.0	288	0.7641	0.1495	0.7641	0.8741
No log	36.25	290	0.7068	0.1254	0.7068	0.8407
No log	36.5	292	0.7010	0.1254	0.7010	0.8372
No log	36.75	294	0.7654	0.1440	0.7654	0.8749
No log	37.0	296	0.9345	-0.0118	0.9345	0.9667
No log	37.25	298	1.0087	0.0440	1.0087	1.0044
No log	37.5	300	0.8799	0.0438	0.8799	0.9380
No log	37.75	302	0.7894	0.1921	0.7894	0.8885
No log	38.0	304	0.7499	0.1659	0.7499	0.8660
No log	38.25	306	0.7397	0.1298	0.7397	0.8600
No log	38.5	308	0.7408	0.1298	0.7408	0.8607
No log	38.75	310	0.7392	0.0732	0.7392	0.8597
No log	39.0	312	0.7688	0.1440	0.7688	0.8768
No log	39.25	314	0.8355	0.1387	0.8355	0.9141
No log	39.5	316	0.8384	0.1387	0.8384	0.9156
No log	39.75	318	0.7855	0.1049	0.7855	0.8863
No log	40.0	320	0.7665	0.2103	0.7665	0.8755
No log	40.25	322	0.7392	0.2181	0.7392	0.8598
No log	40.5	324	0.7406	0.1440	0.7406	0.8606
No log	40.75	326	0.9000	-0.0163	0.9000	0.9487
No log	41.0	328	1.0864	-0.0045	1.0864	1.0423
No log	41.25	330	1.0508	0.0006	1.0508	1.0251
No log	41.5	332	0.8710	0.1147	0.8710	0.9332
No log	41.75	334	0.7778	0.1923	0.7778	0.8819
No log	42.0	336	0.7775	0.2057	0.7775	0.8818
No log	42.25	338	0.7556	0.1983	0.7556	0.8693
No log	42.5	340	0.8036	0.0953	0.8036	0.8964
No log	42.75	342	0.9080	-0.0118	0.9080	0.9529
No log	43.0	344	0.9565	0.0152	0.9565	0.9780
No log	43.25	346	0.8742	-0.0441	0.8742	0.9350
No log	43.5	348	0.7522	0.1495	0.7522	0.8673
No log	43.75	350	0.7281	0.3014	0.7281	0.8533
No log	44.0	352	0.7809	0.3187	0.7809	0.8837
No log	44.25	354	0.7535	0.2490	0.7535	0.8680
No log	44.5	356	0.7063	0.3111	0.7063	0.8404
No log	44.75	358	0.7223	0.1986	0.7223	0.8499
No log	45.0	360	0.8013	0.1286	0.8013	0.8951
No log	45.25	362	0.7650	0.1506	0.7650	0.8747
No log	45.5	364	0.6943	0.1565	0.6943	0.8332
No log	45.75	366	0.6555	0.1878	0.6555	0.8097
No log	46.0	368	0.6503	0.1902	0.6503	0.8064
No log	46.25	370	0.6495	0.1902	0.6495	0.8059
No log	46.5	372	0.6792	0.2180	0.6792	0.8241
No log	46.75	374	0.7044	0.1506	0.7044	0.8393
No log	47.0	376	0.7049	0.1506	0.7049	0.8396
No log	47.25	378	0.7017	0.1659	0.7017	0.8377
No log	47.5	380	0.7323	0.1644	0.7323	0.8557
No log	47.75	382	0.7914	0.1964	0.7914	0.8896
No log	48.0	384	0.8950	-0.0393	0.8950	0.9460
No log	48.25	386	0.9220	0.0182	0.9220	0.9602
No log	48.5	388	0.8482	0.0091	0.8482	0.9210
No log	48.75	390	0.7857	0.0611	0.7857	0.8864
No log	49.0	392	0.7643	0.1541	0.7643	0.8742
No log	49.25	394	0.7688	0.0611	0.7688	0.8768
No log	49.5	396	0.7621	0.0562	0.7621	0.8730
No log	49.75	398	0.7194	0.1553	0.7194	0.8482
No log	50.0	400	0.6874	0.1553	0.6874	0.8291
No log	50.25	402	0.6810	0.1553	0.6810	0.8252
No log	50.5	404	0.6994	0.1495	0.6994	0.8363
No log	50.75	406	0.7097	0.1506	0.7097	0.8424
No log	51.0	408	0.7217	0.1449	0.7217	0.8495
No log	51.25	410	0.6974	0.1553	0.6974	0.8351
No log	51.5	412	0.6829	0.1541	0.6829	0.8264
No log	51.75	414	0.6807	0.1942	0.6807	0.8251
No log	52.0	416	0.6808	0.3317	0.6808	0.8251
No log	52.25	418	0.6839	0.2934	0.6839	0.8270
No log	52.5	420	0.6940	0.2471	0.6940	0.8330
No log	52.75	422	0.7397	0.1495	0.7397	0.8601
No log	53.0	424	0.8129	0.0871	0.8129	0.9016
No log	53.25	426	0.8142	0.0867	0.8142	0.9024
No log	53.5	428	0.7517	0.1387	0.7517	0.8670
No log	53.75	430	0.7022	0.2078	0.7022	0.8380
No log	54.0	432	0.7315	0.2080	0.7315	0.8553
No log	54.25	434	0.7702	0.1649	0.7702	0.8776
No log	54.5	436	0.7786	0.1209	0.7786	0.8824
No log	54.75	438	0.8042	0.1050	0.8042	0.8968
No log	55.0	440	0.8732	0.0816	0.8732	0.9345
No log	55.25	442	0.8908	0.0747	0.8908	0.9438
No log	55.5	444	0.8485	0.1901	0.8485	0.9211
No log	55.75	446	0.7788	0.2336	0.7788	0.8825
No log	56.0	448	0.7525	0.2476	0.7525	0.8675
No log	56.25	450	0.7369	0.2096	0.7369	0.8584
No log	56.5	452	0.7383	0.2009	0.7383	0.8593
No log	56.75	454	0.7611	0.1495	0.7611	0.8724
No log	57.0	456	0.7704	0.1495	0.7704	0.8777
No log	57.25	458	0.7566	0.1495	0.7566	0.8698
No log	57.5	460	0.7267	0.2009	0.7267	0.8525
No log	57.75	462	0.6967	0.2078	0.6967	0.8347
No log	58.0	464	0.6836	0.2078	0.6836	0.8268
No log	58.25	466	0.6834	0.2078	0.6834	0.8267
No log	58.5	468	0.6925	0.2150	0.6925	0.8322
No log	58.75	470	0.7066	0.2053	0.7066	0.8406
No log	59.0	472	0.7285	0.2009	0.7285	0.8535
No log	59.25	474	0.7614	0.1440	0.7614	0.8726
No log	59.5	476	0.7561	0.1440	0.7561	0.8696
No log	59.75	478	0.7617	0.1440	0.7617	0.8728
No log	60.0	480	0.7440	0.1440	0.7440	0.8626
No log	60.25	482	0.7165	0.1506	0.7165	0.8465
No log	60.5	484	0.7107	0.1506	0.7107	0.8430
No log	60.75	486	0.7037	0.1506	0.7037	0.8389
No log	61.0	488	0.6974	0.1506	0.6974	0.8351
No log	61.25	490	0.6985	0.1506	0.6985	0.8358
No log	61.5	492	0.6861	0.1553	0.6861	0.8283
No log	61.75	494	0.6844	0.2078	0.6844	0.8273
No log	62.0	496	0.6829	0.1807	0.6829	0.8264
No log	62.25	498	0.6822	0.1304	0.6822	0.8260
0.2249	62.5	500	0.6853	0.2078	0.6853	0.8278
0.2249	62.75	502	0.7075	0.1565	0.7075	0.8411
0.2249	63.0	504	0.7229	0.1506	0.7229	0.8502
0.2249	63.25	506	0.7064	0.1565	0.7064	0.8405
0.2249	63.5	508	0.6853	0.1659	0.6853	0.8279
0.2249	63.75	510	0.6977	0.1769	0.6977	0.8353

Framework versions

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