ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run3_AugV5_k3_task5_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: 1.1678
• Qwk: 0.5818
• Mse: 1.1678
• Rmse: 1.0806

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	0.1429	2	2.3380	0.0392	2.3380	1.5291
No log	0.2857	4	1.4509	0.1988	1.4509	1.2045
No log	0.4286	6	1.4480	0.1672	1.4480	1.2033
No log	0.5714	8	1.6235	0.1822	1.6235	1.2741
No log	0.7143	10	1.6591	0.1954	1.6591	1.2881
No log	0.8571	12	1.6976	0.2897	1.6976	1.3029
No log	1.0	14	1.6739	0.3007	1.6739	1.2938
No log	1.1429	16	1.6049	0.3649	1.6049	1.2669
No log	1.2857	18	1.5255	0.3906	1.5255	1.2351
No log	1.4286	20	1.4423	0.3833	1.4423	1.2010
No log	1.5714	22	1.1887	0.3955	1.1887	1.0903
No log	1.7143	24	1.1526	0.4659	1.1526	1.0736
No log	1.8571	26	1.2656	0.4606	1.2656	1.1250
No log	2.0	28	1.4015	0.4548	1.4015	1.1838
No log	2.1429	30	1.4590	0.4361	1.4590	1.2079
No log	2.2857	32	1.2075	0.5077	1.2075	1.0989
No log	2.4286	34	1.0858	0.4871	1.0858	1.0420
No log	2.5714	36	1.0713	0.4604	1.0713	1.0350
No log	2.7143	38	1.0695	0.4995	1.0695	1.0342
No log	2.8571	40	1.3381	0.5489	1.3381	1.1568
No log	3.0	42	1.7076	0.4512	1.7076	1.3068
No log	3.1429	44	1.6518	0.4473	1.6518	1.2852
No log	3.2857	46	1.2447	0.5426	1.2447	1.1156
No log	3.4286	48	1.0705	0.4773	1.0705	1.0347
No log	3.5714	50	1.0083	0.4922	1.0083	1.0041
No log	3.7143	52	1.0299	0.4647	1.0299	1.0149
No log	3.8571	54	1.0353	0.4901	1.0353	1.0175
No log	4.0	56	0.9936	0.4569	0.9936	0.9968
No log	4.1429	58	1.0390	0.5072	1.0390	1.0193
No log	4.2857	60	1.0480	0.5247	1.0480	1.0237
No log	4.4286	62	1.1248	0.5746	1.1248	1.0606
No log	4.5714	64	1.2700	0.5537	1.2700	1.1269
No log	4.7143	66	1.1689	0.5847	1.1689	1.0812
No log	4.8571	68	1.1262	0.5857	1.1262	1.0612
No log	5.0	70	1.1676	0.6049	1.1676	1.0806
No log	5.1429	72	1.2461	0.5830	1.2461	1.1163
No log	5.2857	74	1.2099	0.5988	1.2099	1.1000
No log	5.4286	76	1.1051	0.6065	1.1051	1.0512
No log	5.5714	78	1.1965	0.5861	1.1965	1.0938
No log	5.7143	80	1.5401	0.5539	1.5401	1.2410
No log	5.8571	82	1.9874	0.4608	1.9874	1.4098
No log	6.0	84	1.9450	0.4747	1.9450	1.3946
No log	6.1429	86	1.6307	0.5494	1.6307	1.2770
No log	6.2857	88	1.2160	0.5632	1.2160	1.1027
No log	6.4286	90	1.0648	0.5889	1.0648	1.0319
No log	6.5714	92	1.1238	0.5778	1.1238	1.0601
No log	6.7143	94	1.3563	0.5677	1.3563	1.1646
No log	6.8571	96	1.6199	0.5509	1.6199	1.2728
No log	7.0	98	1.5464	0.5423	1.5464	1.2435
No log	7.1429	100	1.2806	0.5675	1.2806	1.1316
No log	7.2857	102	1.0761	0.5822	1.0761	1.0374
No log	7.4286	104	1.0052	0.5720	1.0052	1.0026
No log	7.5714	106	1.0447	0.5844	1.0447	1.0221
No log	7.7143	108	1.1263	0.5970	1.1263	1.0613
No log	7.8571	110	1.1961	0.5903	1.1961	1.0937
No log	8.0	112	1.1659	0.5914	1.1659	1.0798
No log	8.1429	114	1.1466	0.5891	1.1466	1.0708
No log	8.2857	116	1.1552	0.5982	1.1552	1.0748
No log	8.4286	118	1.1997	0.5908	1.1997	1.0953
No log	8.5714	120	1.2351	0.5838	1.2351	1.1114
No log	8.7143	122	1.2542	0.5798	1.2542	1.1199
No log	8.8571	124	1.1996	0.5838	1.1996	1.0952
No log	9.0	126	1.1119	0.5861	1.1119	1.0545
No log	9.1429	128	1.0846	0.5905	1.0846	1.0414
No log	9.2857	130	1.0982	0.5808	1.0982	1.0480
No log	9.4286	132	1.1271	0.5829	1.1271	1.0616
No log	9.5714	134	1.1317	0.5807	1.1317	1.0638
No log	9.7143	136	1.1458	0.5735	1.1458	1.0704
No log	9.8571	138	1.1610	0.5818	1.1610	1.0775
No log	10.0	140	1.1678	0.5818	1.1678	1.0806

Framework versions

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