ArabicNewSplits5_FineTuningAraBERT_run1_AugV5_k3_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: 0.6784
• Qwk: 0.7243
• Mse: 0.6784
• Rmse: 0.8237

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	0.0909	2	5.3334	-0.0349	5.3334	2.3094
No log	0.1818	4	3.0768	0.0786	3.0768	1.7541
No log	0.2727	6	1.9818	0.1627	1.9818	1.4078
No log	0.3636	8	1.8380	0.1429	1.8380	1.3557
No log	0.4545	10	1.8140	0.1733	1.8140	1.3468
No log	0.5455	12	1.7756	0.1816	1.7756	1.3325
No log	0.6364	14	1.5274	0.1979	1.5274	1.2359
No log	0.7273	16	1.2658	0.3731	1.2658	1.1251
No log	0.8182	18	1.7570	0.3194	1.7570	1.3255
No log	0.9091	20	1.7068	0.3431	1.7068	1.3064
No log	1.0	22	1.5386	0.3855	1.5386	1.2404
No log	1.0909	24	1.4374	0.4130	1.4374	1.1989
No log	1.1818	26	1.3035	0.4402	1.3035	1.1417
No log	1.2727	28	1.2854	0.4791	1.2854	1.1337
No log	1.3636	30	1.8391	0.3840	1.8391	1.3561
No log	1.4545	32	1.9487	0.3733	1.9487	1.3959
No log	1.5455	34	1.5374	0.4076	1.5374	1.2399
No log	1.6364	36	0.8981	0.5741	0.8981	0.9477
No log	1.7273	38	0.6609	0.6774	0.6609	0.8130
No log	1.8182	40	0.6345	0.6603	0.6345	0.7966
No log	1.9091	42	0.5974	0.6921	0.5974	0.7729
No log	2.0	44	0.6142	0.6770	0.6142	0.7837
No log	2.0909	46	0.8390	0.6055	0.8390	0.9160
No log	2.1818	48	1.2639	0.4385	1.2639	1.1242
No log	2.2727	50	1.0795	0.4853	1.0795	1.0390
No log	2.3636	52	0.8132	0.6622	0.8132	0.9018
No log	2.4545	54	0.6757	0.7017	0.6757	0.8220
No log	2.5455	56	0.7108	0.6952	0.7108	0.8431
No log	2.6364	58	0.7020	0.6928	0.7020	0.8379
No log	2.7273	60	0.7823	0.6932	0.7823	0.8845
No log	2.8182	62	1.1904	0.4771	1.1904	1.0911
No log	2.9091	64	1.5797	0.4183	1.5797	1.2568
No log	3.0	66	1.4718	0.4202	1.4718	1.2132
No log	3.0909	68	0.9715	0.5963	0.9715	0.9856
No log	3.1818	70	0.6219	0.7108	0.6219	0.7886
No log	3.2727	72	0.5974	0.7318	0.5974	0.7729
No log	3.3636	74	0.6166	0.7318	0.6166	0.7852
No log	3.4545	76	0.6546	0.7383	0.6546	0.8091
No log	3.5455	78	0.7142	0.7200	0.7142	0.8451
No log	3.6364	80	0.7155	0.7200	0.7155	0.8459
No log	3.7273	82	0.6790	0.7265	0.6790	0.8240
No log	3.8182	84	0.6695	0.7160	0.6695	0.8183
No log	3.9091	86	0.6670	0.7141	0.6670	0.8167
No log	4.0	88	0.6581	0.7084	0.6581	0.8112
No log	4.0909	90	0.6615	0.7171	0.6615	0.8133
No log	4.1818	92	0.7214	0.6988	0.7214	0.8494
No log	4.2727	94	0.6990	0.7028	0.6990	0.8361
No log	4.3636	96	0.6673	0.7331	0.6673	0.8169
No log	4.4545	98	0.6561	0.7422	0.6561	0.8100
No log	4.5455	100	0.6924	0.7056	0.6924	0.8321
No log	4.6364	102	0.6591	0.7386	0.6591	0.8118
No log	4.7273	104	0.6527	0.7378	0.6527	0.8079
No log	4.8182	106	0.6455	0.7480	0.6455	0.8034
No log	4.9091	108	0.7021	0.7318	0.7021	0.8379
No log	5.0	110	0.8543	0.6445	0.8543	0.9243
No log	5.0909	112	0.9474	0.6220	0.9474	0.9734
No log	5.1818	114	0.9031	0.6358	0.9031	0.9503
No log	5.2727	116	0.7187	0.7262	0.7187	0.8478
No log	5.3636	118	0.6387	0.7289	0.6387	0.7992
No log	5.4545	120	0.6426	0.7359	0.6426	0.8016
No log	5.5455	122	0.6419	0.7514	0.6419	0.8012
No log	5.6364	124	0.7038	0.6977	0.7038	0.8390
No log	5.7273	126	0.7440	0.6594	0.7440	0.8626
No log	5.8182	128	0.7353	0.6708	0.7353	0.8575
No log	5.9091	130	0.8099	0.6444	0.8099	0.8999
No log	6.0	132	0.9714	0.5903	0.9714	0.9856
No log	6.0909	134	0.9109	0.6194	0.9109	0.9544
No log	6.1818	136	0.8026	0.6658	0.8026	0.8959
No log	6.2727	138	0.7076	0.7049	0.7076	0.8412
No log	6.3636	140	0.7021	0.7085	0.7021	0.8379
No log	6.4545	142	0.6864	0.7244	0.6864	0.8285
No log	6.5455	144	0.6683	0.7466	0.6683	0.8175
No log	6.6364	146	0.7100	0.7108	0.7100	0.8426
No log	6.7273	148	0.7357	0.7183	0.7357	0.8577
No log	6.8182	150	0.7408	0.7226	0.7408	0.8607
No log	6.9091	152	0.7439	0.7353	0.7439	0.8625
No log	7.0	154	0.7395	0.7276	0.7395	0.8600
No log	7.0909	156	0.7259	0.7434	0.7259	0.8520
No log	7.1818	158	0.7073	0.7345	0.7073	0.8410
No log	7.2727	160	0.6800	0.7431	0.6800	0.8246
No log	7.3636	162	0.6563	0.7379	0.6563	0.8102
No log	7.4545	164	0.6431	0.7220	0.6431	0.8019
No log	7.5455	166	0.6394	0.7273	0.6394	0.7997
No log	7.6364	168	0.6602	0.7248	0.6602	0.8125
No log	7.7273	170	0.7104	0.7151	0.7104	0.8428
No log	7.8182	172	0.7943	0.6478	0.7943	0.8912
No log	7.9091	174	0.8135	0.6427	0.8135	0.9020
No log	8.0	176	0.7616	0.6640	0.7616	0.8727
No log	8.0909	178	0.6800	0.7176	0.6800	0.8246
No log	8.1818	180	0.6331	0.7235	0.6331	0.7957
No log	8.2727	182	0.6245	0.7328	0.6245	0.7902
No log	8.3636	184	0.6328	0.7359	0.6328	0.7955
No log	8.4545	186	0.6415	0.7245	0.6415	0.8010
No log	8.5455	188	0.6454	0.7328	0.6454	0.8034
No log	8.6364	190	0.6539	0.7344	0.6539	0.8086
No log	8.7273	192	0.6821	0.7234	0.6821	0.8259
No log	8.8182	194	0.7189	0.7131	0.7189	0.8479
No log	8.9091	196	0.7239	0.7228	0.7239	0.8508
No log	9.0	198	0.7063	0.7239	0.7063	0.8404
No log	9.0909	200	0.6944	0.7214	0.6944	0.8333
No log	9.1818	202	0.6876	0.7096	0.6876	0.8292
No log	9.2727	204	0.6866	0.7096	0.6866	0.8286
No log	9.3636	206	0.6884	0.7214	0.6884	0.8297
No log	9.4545	208	0.6849	0.7214	0.6849	0.8276
No log	9.5455	210	0.6829	0.7096	0.6829	0.8263
No log	9.6364	212	0.6828	0.7131	0.6828	0.8263
No log	9.7273	214	0.6824	0.7131	0.6824	0.8261
No log	9.8182	216	0.6796	0.7243	0.6796	0.8244
No log	9.9091	218	0.6788	0.7243	0.6788	0.8239
No log	10.0	220	0.6784	0.7243	0.6784	0.8237

Framework versions

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