ArabicNewSplits5_FineTuningAraBERT_run1_AugV5_k2_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: 0.6769
• Qwk: 0.7119
• Mse: 0.6769
• Rmse: 0.8227

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	0.125	2	5.5699	-0.0294	5.5699	2.3601
No log	0.25	4	3.5628	0.0711	3.5628	1.8875
No log	0.375	6	2.0608	0.1163	2.0608	1.4356
No log	0.5	8	1.5821	-0.0031	1.5821	1.2578
No log	0.625	10	1.3989	0.1839	1.3989	1.1827
No log	0.75	12	1.2245	0.3124	1.2245	1.1066
No log	0.875	14	1.2244	0.2492	1.2244	1.1065
No log	1.0	16	1.1510	0.3679	1.1510	1.0728
No log	1.125	18	1.3392	0.1647	1.3392	1.1572
No log	1.25	20	1.8963	-0.0342	1.8963	1.3771
No log	1.375	22	2.3051	-0.0335	2.3051	1.5182
No log	1.5	24	1.7278	0.0729	1.7278	1.3145
No log	1.625	26	1.1493	0.2402	1.1493	1.0720
No log	1.75	28	1.0788	0.3573	1.0788	1.0387
No log	1.875	30	1.1913	0.3301	1.1913	1.0915
No log	2.0	32	1.2786	0.1211	1.2786	1.1307
No log	2.125	34	1.3259	0.0927	1.3259	1.1515
No log	2.25	36	1.3912	0.0428	1.3912	1.1795
No log	2.375	38	1.3741	0.1032	1.3741	1.1722
No log	2.5	40	1.4088	0.1117	1.4088	1.1869
No log	2.625	42	1.3344	0.1314	1.3344	1.1552
No log	2.75	44	1.2236	0.1513	1.2236	1.1062
No log	2.875	46	1.2397	0.1994	1.2397	1.1134
No log	3.0	48	1.2153	0.2276	1.2153	1.1024
No log	3.125	50	1.1248	0.3671	1.1248	1.0606
No log	3.25	52	1.1311	0.3888	1.1311	1.0635
No log	3.375	54	1.0116	0.4708	1.0116	1.0058
No log	3.5	56	0.8913	0.4679	0.8913	0.9441
No log	3.625	58	0.8537	0.4952	0.8537	0.9240
No log	3.75	60	0.8502	0.4759	0.8502	0.9221
No log	3.875	62	0.8600	0.4980	0.8600	0.9274
No log	4.0	64	0.9305	0.5503	0.9305	0.9646
No log	4.125	66	0.9184	0.5605	0.9184	0.9583
No log	4.25	68	0.8445	0.5581	0.8445	0.9190
No log	4.375	70	0.7494	0.5460	0.7494	0.8657
No log	4.5	72	0.6968	0.5764	0.6968	0.8347
No log	4.625	74	0.6862	0.6226	0.6862	0.8284
No log	4.75	76	0.6764	0.6686	0.6764	0.8224
No log	4.875	78	0.6657	0.6708	0.6657	0.8159
No log	5.0	80	0.6826	0.6748	0.6826	0.8262
No log	5.125	82	0.7007	0.6697	0.7007	0.8371
No log	5.25	84	0.6861	0.6799	0.6861	0.8283
No log	5.375	86	0.6389	0.6771	0.6389	0.7993
No log	5.5	88	0.6253	0.7039	0.6253	0.7908
No log	5.625	90	0.6566	0.6686	0.6566	0.8103
No log	5.75	92	0.7058	0.6387	0.7058	0.8401
No log	5.875	94	0.7066	0.6860	0.7066	0.8406
No log	6.0	96	0.6583	0.7253	0.6583	0.8114
No log	6.125	98	0.6689	0.6474	0.6689	0.8179
No log	6.25	100	0.7286	0.6412	0.7286	0.8536
No log	6.375	102	0.7135	0.6511	0.7135	0.8447
No log	6.5	104	0.6514	0.6599	0.6514	0.8071
No log	6.625	106	0.6215	0.7041	0.6215	0.7883
No log	6.75	108	0.6647	0.7017	0.6647	0.8153
No log	6.875	110	0.7270	0.6703	0.7270	0.8526
No log	7.0	112	0.7418	0.6703	0.7418	0.8613
No log	7.125	114	0.7211	0.6700	0.7211	0.8492
No log	7.25	116	0.7008	0.7227	0.7008	0.8371
No log	7.375	118	0.6878	0.7179	0.6878	0.8293
No log	7.5	120	0.6912	0.7221	0.6912	0.8314
No log	7.625	122	0.7018	0.7177	0.7018	0.8377
No log	7.75	124	0.7024	0.7053	0.7024	0.8381
No log	7.875	126	0.7019	0.7106	0.7019	0.8378
No log	8.0	128	0.7044	0.7033	0.7044	0.8393
No log	8.125	130	0.7068	0.7184	0.7068	0.8407
No log	8.25	132	0.7135	0.7321	0.7135	0.8447
No log	8.375	134	0.7053	0.7327	0.7053	0.8398
No log	8.5	136	0.7024	0.7305	0.7024	0.8381
No log	8.625	138	0.7004	0.7349	0.7004	0.8369
No log	8.75	140	0.6885	0.7250	0.6885	0.8297
No log	8.875	142	0.6758	0.7202	0.6758	0.8221
No log	9.0	144	0.6681	0.7179	0.6681	0.8174
No log	9.125	146	0.6674	0.7119	0.6674	0.8170
No log	9.25	148	0.6680	0.6982	0.6680	0.8173
No log	9.375	150	0.6715	0.7020	0.6715	0.8194
No log	9.5	152	0.6731	0.7081	0.6731	0.8204
No log	9.625	154	0.6739	0.7119	0.6739	0.8209
No log	9.75	156	0.6752	0.7119	0.6752	0.8217
No log	9.875	158	0.6763	0.7119	0.6763	0.8224
No log	10.0	160	0.6769	0.7119	0.6769	0.8227

Framework versions

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