ArabicNewSplits7_FineTuningAraBERT_run3_AugV5_k5_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: 0.9195
• Qwk: 0.0152
• Mse: 0.9195
• Rmse: 0.9589

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	0.125	2	3.5160	0.0048	3.5160	1.8751
No log	0.25	4	1.8904	0.0943	1.8904	1.3749
No log	0.375	6	1.4957	-0.0460	1.4957	1.2230
No log	0.5	8	1.3978	0.0731	1.3978	1.1823
No log	0.625	10	1.4233	0.0677	1.4233	1.1930
No log	0.75	12	1.2199	0.0493	1.2199	1.1045
No log	0.875	14	0.7043	0.0460	0.7043	0.8392
No log	1.0	16	0.7052	-0.0035	0.7052	0.8398
No log	1.125	18	0.9155	0.1374	0.9155	0.9568
No log	1.25	20	0.9848	0.0469	0.9848	0.9924
No log	1.375	22	0.8832	0.1589	0.8832	0.9398
No log	1.5	24	0.8492	0.1107	0.8492	0.9215
No log	1.625	26	1.2594	0.0048	1.2594	1.1222
No log	1.75	28	1.4924	0.0048	1.4924	1.2216
No log	1.875	30	1.2713	0.0048	1.2713	1.1275
No log	2.0	32	0.9138	0.0353	0.9138	0.9559
No log	2.125	34	0.8756	-0.0595	0.8756	0.9358
No log	2.25	36	0.8938	-0.0923	0.8938	0.9454
No log	2.375	38	1.0338	0.0006	1.0338	1.0168
No log	2.5	40	0.8572	-0.0408	0.8572	0.9258
No log	2.625	42	0.8017	-0.0725	0.8017	0.8954
No log	2.75	44	0.8572	0.0129	0.8572	0.9258
No log	2.875	46	1.0544	0.0810	1.0544	1.0268
No log	3.0	48	1.9534	0.0659	1.9534	1.3977
No log	3.125	50	2.2393	0.0408	2.2393	1.4964
No log	3.25	52	1.5814	0.0095	1.5814	1.2575
No log	3.375	54	0.8227	0.0225	0.8227	0.9070
No log	3.5	56	0.7734	-0.0069	0.7734	0.8794
No log	3.625	58	0.7791	-0.0188	0.7791	0.8827
No log	3.75	60	1.0952	-0.0331	1.0952	1.0465
No log	3.875	62	1.4573	-0.0479	1.4573	1.2072
No log	4.0	64	1.2023	-0.0398	1.2023	1.0965
No log	4.125	66	0.7852	0.1148	0.7852	0.8861
No log	4.25	68	0.7155	0.0857	0.7155	0.8459
No log	4.375	70	0.7732	0.0282	0.7732	0.8793
No log	4.5	72	1.0712	-0.0253	1.0712	1.0350
No log	4.625	74	1.1923	0.0025	1.1923	1.0919
No log	4.75	76	1.0888	-0.0471	1.0888	1.0434
No log	4.875	78	1.1665	-0.0156	1.1665	1.0801
No log	5.0	80	1.4049	-0.0511	1.4049	1.1853
No log	5.125	82	1.2603	-0.0434	1.2603	1.1226
No log	5.25	84	1.0434	-0.0225	1.0434	1.0215
No log	5.375	86	0.9533	0.0771	0.9533	0.9764
No log	5.5	88	1.0782	-0.0047	1.0782	1.0383
No log	5.625	90	1.2460	-0.0807	1.2460	1.1162
No log	5.75	92	1.1855	-0.1077	1.1855	1.0888
No log	5.875	94	1.0604	0.0365	1.0604	1.0298
No log	6.0	96	1.1585	0.0248	1.1585	1.0763
No log	6.125	98	1.2399	-0.0231	1.2399	1.1135
No log	6.25	100	0.9557	0.0081	0.9557	0.9776
No log	6.375	102	0.8727	0.1133	0.8727	0.9342
No log	6.5	104	0.8575	0.1263	0.8575	0.9260
No log	6.625	106	0.9604	0.0920	0.9604	0.9800
No log	6.75	108	1.3344	-0.0870	1.3344	1.1552
No log	6.875	110	1.3080	-0.0870	1.3080	1.1437
No log	7.0	112	1.2627	-0.0876	1.2627	1.1237
No log	7.125	114	1.1068	-0.1186	1.1068	1.0521
No log	7.25	116	0.9909	0.0109	0.9909	0.9954
No log	7.375	118	0.9873	-0.0184	0.9873	0.9936
No log	7.5	120	1.0590	0.0364	1.0590	1.0291
No log	7.625	122	1.0409	0.1103	1.0409	1.0202
No log	7.75	124	1.0723	0.0566	1.0723	1.0355
No log	7.875	126	0.9965	-0.0087	0.9965	0.9982
No log	8.0	128	0.9849	0.0627	0.9849	0.9924
No log	8.125	130	0.8645	0.0927	0.8645	0.9298
No log	8.25	132	0.8182	0.0816	0.8182	0.9045
No log	8.375	134	0.8653	0.0161	0.8653	0.9302
No log	8.5	136	0.9261	0.1145	0.9261	0.9623
No log	8.625	138	0.8542	0.0087	0.8542	0.9242
No log	8.75	140	0.8277	0.0038	0.8277	0.9098
No log	8.875	142	0.8814	-0.0101	0.8814	0.9388
No log	9.0	144	0.7352	0.1287	0.7352	0.8574
No log	9.125	146	0.7299	0.1453	0.7299	0.8544
No log	9.25	148	0.8611	0.0041	0.8611	0.9279
No log	9.375	150	0.9724	-0.1541	0.9724	0.9861
No log	9.5	152	1.0948	-0.0521	1.0948	1.0463
No log	9.625	154	0.8979	0.1440	0.8979	0.9476
No log	9.75	156	0.8893	0.0592	0.8893	0.9430
No log	9.875	158	0.8799	0.0592	0.8799	0.9380
No log	10.0	160	0.9910	-0.0438	0.9910	0.9955
No log	10.125	162	1.0511	-0.0526	1.0511	1.0252
No log	10.25	164	0.9184	-0.0440	0.9184	0.9583
No log	10.375	166	0.8159	-0.0163	0.8159	0.9033
No log	10.5	168	0.8753	-0.0052	0.8753	0.9356
No log	10.625	170	1.0402	0.0111	1.0402	1.0199
No log	10.75	172	1.1665	-0.0084	1.1665	1.0801
No log	10.875	174	0.8955	-0.0030	0.8955	0.9463
No log	11.0	176	0.8233	0.1408	0.8233	0.9074
No log	11.125	178	0.8387	0.1408	0.8387	0.9158
No log	11.25	180	0.8796	0.0892	0.8796	0.9379
No log	11.375	182	0.9183	0.0787	0.9183	0.9583
No log	11.5	184	0.9709	-0.0409	0.9709	0.9853
No log	11.625	186	0.9410	-0.0777	0.9410	0.9700
No log	11.75	188	0.9325	-0.0482	0.9325	0.9656
No log	11.875	190	0.9857	0.0486	0.9857	0.9928
No log	12.0	192	0.9762	0.0486	0.9762	0.9880
No log	12.125	194	0.8750	0.0250	0.8750	0.9354
No log	12.25	196	0.8317	0.0504	0.8317	0.9120
No log	12.375	198	0.9116	-0.0073	0.9116	0.9548
No log	12.5	200	0.8876	-0.0393	0.8876	0.9421
No log	12.625	202	0.8685	-0.0359	0.8685	0.9319
No log	12.75	204	0.9856	0.0556	0.9856	0.9928
No log	12.875	206	1.2826	0.0218	1.2826	1.1325
No log	13.0	208	1.3151	0.0972	1.3151	1.1468
No log	13.125	210	0.9838	-0.0513	0.9838	0.9918
No log	13.25	212	0.7786	0.1734	0.7786	0.8824
No log	13.375	214	0.7555	0.1443	0.7555	0.8692
No log	13.5	216	0.7518	0.0562	0.7518	0.8671
No log	13.625	218	0.9532	0.0134	0.9532	0.9763
No log	13.75	220	1.0667	0.0089	1.0667	1.0328
No log	13.875	222	0.8977	-0.0504	0.8977	0.9475
No log	14.0	224	0.7974	0.0030	0.7974	0.8929
No log	14.125	226	0.7927	0.0543	0.7927	0.8904
No log	14.25	228	0.8033	0.0289	0.8033	0.8963
No log	14.375	230	0.9368	0.0241	0.9368	0.9679
No log	14.5	232	1.1265	0.0045	1.1265	1.0614
No log	14.625	234	1.2298	0.0456	1.2298	1.1090
No log	14.75	236	1.0197	0.0067	1.0197	1.0098
No log	14.875	238	0.7640	0.1449	0.7640	0.8741
No log	15.0	240	0.7428	0.1080	0.7428	0.8618
No log	15.125	242	0.8911	0.0048	0.8911	0.9440
No log	15.25	244	0.8814	0.0364	0.8814	0.9388
No log	15.375	246	0.7897	0.2222	0.7897	0.8887
No log	15.5	248	1.1503	0.0041	1.1503	1.0725
No log	15.625	250	2.0538	0.0613	2.0538	1.4331
No log	15.75	252	2.2740	0.0016	2.2740	1.5080
No log	15.875	254	1.8237	0.0819	1.8237	1.3504
No log	16.0	256	1.1449	0.0101	1.1449	1.0700
No log	16.125	258	0.8842	0.1558	0.8842	0.9403
No log	16.25	260	0.8459	0.2019	0.8459	0.9197
No log	16.375	262	0.8571	0.0920	0.8571	0.9258
No log	16.5	264	1.2129	0.0252	1.2129	1.1013
No log	16.625	266	1.4645	0.0320	1.4645	1.2102
No log	16.75	268	1.2024	0.0456	1.2024	1.0965
No log	16.875	270	0.8226	0.1291	0.8226	0.9070
No log	17.0	272	0.7183	-0.0541	0.7183	0.8475
No log	17.125	274	0.7354	-0.0541	0.7354	0.8575
No log	17.25	276	0.7500	0.0303	0.7500	0.8660
No log	17.375	278	0.9188	0.0576	0.9188	0.9586
No log	17.5	280	1.3530	0.0431	1.3530	1.1632
No log	17.625	282	1.5368	0.0674	1.5368	1.2397
No log	17.75	284	1.4187	0.0772	1.4187	1.1911
No log	17.875	286	1.0696	0.0680	1.0696	1.0342
No log	18.0	288	0.8035	-0.0694	0.8035	0.8964
No log	18.125	290	0.7490	0.0449	0.7490	0.8654
No log	18.25	292	0.7400	-0.0059	0.7400	0.8602
No log	18.375	294	0.7798	0.1449	0.7798	0.8831
No log	18.5	296	0.9323	0.0710	0.9323	0.9655
No log	18.625	298	1.0192	0.0569	1.0192	1.0096
No log	18.75	300	0.9577	0.0458	0.9577	0.9786
No log	18.875	302	0.8427	0.1190	0.8427	0.9180
No log	19.0	304	0.7713	0.0412	0.7713	0.8782
No log	19.125	306	0.7581	0.0027	0.7581	0.8707
No log	19.25	308	0.7499	0.0376	0.7499	0.8660
No log	19.375	310	0.8976	0.0636	0.8976	0.9474
No log	19.5	312	1.1334	0.0786	1.1334	1.0646
No log	19.625	314	1.2039	0.0188	1.2039	1.0972
No log	19.75	316	1.0888	0.0687	1.0888	1.0435
No log	19.875	318	0.9120	0.0778	0.9120	0.9550
No log	20.0	320	0.8047	0.1716	0.8047	0.8971
No log	20.125	322	0.8064	0.1605	0.8064	0.8980
No log	20.25	324	0.8663	0.0182	0.8663	0.9307
No log	20.375	326	0.9535	0.0404	0.9535	0.9765
No log	20.5	328	1.0467	0.0319	1.0467	1.0231
No log	20.625	330	1.0567	0.0342	1.0567	1.0280
No log	20.75	332	0.9524	-0.0181	0.9524	0.9759
No log	20.875	334	0.8585	0.1105	0.8585	0.9265
No log	21.0	336	0.8030	0.1901	0.8030	0.8961
No log	21.125	338	0.8127	0.1727	0.8127	0.9015
No log	21.25	340	0.8916	0.1265	0.8916	0.9443
No log	21.375	342	1.0776	-0.0282	1.0776	1.0381
No log	21.5	344	1.1301	-0.0012	1.1301	1.0631
No log	21.625	346	1.0339	-0.0276	1.0339	1.0168
No log	21.75	348	0.8748	0.0755	0.8748	0.9353
No log	21.875	350	0.7854	0.2096	0.7854	0.8862
No log	22.0	352	0.7691	0.2471	0.7691	0.8770
No log	22.125	354	0.8372	0.0407	0.8372	0.9150
No log	22.25	356	1.0372	0.0527	1.0372	1.0184
No log	22.375	358	1.3195	0.0320	1.3195	1.1487
No log	22.5	360	1.3283	0.0339	1.3283	1.1525
No log	22.625	362	1.1036	-0.0297	1.1036	1.0505
No log	22.75	364	0.9029	0.0421	0.9029	0.9502
No log	22.875	366	0.8594	0.0876	0.8594	0.9270
No log	23.0	368	0.8280	0.1003	0.8280	0.9100
No log	23.125	370	0.8238	0.0424	0.8238	0.9077
No log	23.25	372	0.8479	-0.0031	0.8479	0.9208
No log	23.375	374	0.9022	-0.0157	0.9022	0.9498
No log	23.5	376	0.9651	0.0046	0.9651	0.9824
No log	23.625	378	0.9459	0.0152	0.9459	0.9726
No log	23.75	380	0.8742	0.0304	0.8742	0.9350
No log	23.875	382	0.8727	0.0333	0.8727	0.9342
No log	24.0	384	0.8765	0.0362	0.8765	0.9362
No log	24.125	386	0.9088	-0.0097	0.9088	0.9533
No log	24.25	388	1.0174	-0.0211	1.0174	1.0087
No log	24.375	390	1.2589	-0.0666	1.2589	1.1220
No log	24.5	392	1.3870	-0.0178	1.3870	1.1777
No log	24.625	394	1.2264	-0.0666	1.2264	1.1074
No log	24.75	396	1.1004	-0.0618	1.1004	1.0490
No log	24.875	398	0.9183	0.0175	0.9183	0.9583
No log	25.0	400	0.7666	-0.0660	0.7666	0.8756
No log	25.125	402	0.7314	0.0479	0.7314	0.8552
No log	25.25	404	0.7329	0.0479	0.7329	0.8561
No log	25.375	406	0.8205	-0.0054	0.8205	0.9058
No log	25.5	408	0.9576	0.0378	0.9576	0.9786
No log	25.625	410	1.0311	0.0378	1.0311	1.0154
No log	25.75	412	0.9679	0.0378	0.9679	0.9838
No log	25.875	414	0.8464	-0.0837	0.8464	0.9200
No log	26.0	416	0.7816	0.0	0.7816	0.8841
No log	26.125	418	0.8113	0.1032	0.8113	0.9007
No log	26.25	420	0.8173	0.1379	0.8173	0.9041
No log	26.375	422	0.8593	-0.1544	0.8593	0.9270
No log	26.5	424	0.8834	0.0224	0.8834	0.9399
No log	26.625	426	0.8302	-0.1155	0.8302	0.9111
No log	26.75	428	0.8119	-0.0218	0.8119	0.9011
No log	26.875	430	0.8342	-0.1952	0.8342	0.9134
No log	27.0	432	0.8993	-0.0157	0.8993	0.9483
No log	27.125	434	0.9324	0.0129	0.9324	0.9656
No log	27.25	436	0.9318	0.0129	0.9318	0.9653
No log	27.375	438	0.9323	0.0458	0.9323	0.9656
No log	27.5	440	0.8434	-0.0504	0.8434	0.9184
No log	27.625	442	0.7631	0.0821	0.7631	0.8736
No log	27.75	444	0.7479	0.0414	0.7479	0.8648
No log	27.875	446	0.7539	0.0414	0.7539	0.8683
No log	28.0	448	0.7606	0.0375	0.7606	0.8721
No log	28.125	450	0.8219	-0.0746	0.8219	0.9066
No log	28.25	452	0.9506	0.0515	0.9506	0.9750
No log	28.375	454	0.9843	0.0111	0.9843	0.9921
No log	28.5	456	0.8975	0.0224	0.8975	0.9473
No log	28.625	458	0.7781	0.1192	0.7781	0.8821
No log	28.75	460	0.7634	0.0412	0.7634	0.8737
No log	28.875	462	0.7975	0.0749	0.7975	0.8930
No log	29.0	464	0.8319	0.1004	0.8319	0.9121
No log	29.125	466	0.9215	-0.0094	0.9215	0.9599
No log	29.25	468	1.0840	-0.0316	1.0840	1.0411
No log	29.375	470	1.1274	-0.0067	1.1274	1.0618
No log	29.5	472	1.0358	0.0067	1.0358	1.0177
No log	29.625	474	0.8784	0.1065	0.8784	0.9372
No log	29.75	476	0.7898	0.0690	0.7898	0.8887
No log	29.875	478	0.7881	0.1541	0.7881	0.8878
No log	30.0	480	0.8251	0.0959	0.8251	0.9083
No log	30.125	482	0.9030	-0.0114	0.9030	0.9502
No log	30.25	484	1.0025	0.0391	1.0025	1.0012
No log	30.375	486	1.1322	0.0552	1.1322	1.0641
No log	30.5	488	1.0739	0.0274	1.0739	1.0363
No log	30.625	490	0.9279	0.0431	0.9279	0.9633
No log	30.75	492	0.8065	0.1243	0.8065	0.8981
No log	30.875	494	0.7860	0.2718	0.7860	0.8866
No log	31.0	496	0.7837	0.2248	0.7837	0.8852
No log	31.125	498	0.7991	0.1716	0.7991	0.8939
0.2843	31.25	500	0.8701	0.0515	0.8701	0.9328
0.2843	31.375	502	1.0315	0.0217	1.0315	1.0156
0.2843	31.5	504	1.1361	0.0423	1.1361	1.0659
0.2843	31.625	506	1.1254	0.0469	1.1254	1.0609
0.2843	31.75	508	1.0126	0.0305	1.0126	1.0063
0.2843	31.875	510	0.9195	0.0152	0.9195	0.9589

Framework versions

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