ArabicNewSplits4_WithDuplicationsForScore5_FineTuningAraBERT_run2_AugV5_k2_task2_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: 1.0534
• Qwk: 0.4686
• Mse: 1.0534
• Rmse: 1.0264

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	0.1538	2	4.1348	-0.0084	4.1348	2.0334
No log	0.3077	4	2.1811	0.0681	2.1811	1.4768
No log	0.4615	6	1.0728	0.1031	1.0728	1.0358
No log	0.6154	8	0.8160	0.0266	0.8160	0.9033
No log	0.7692	10	0.7032	0.2377	0.7032	0.8385
No log	0.9231	12	0.6623	0.2730	0.6623	0.8138
No log	1.0769	14	0.6419	0.2378	0.6419	0.8012
No log	1.2308	16	0.6154	0.2598	0.6154	0.7845
No log	1.3846	18	0.6020	0.2577	0.6020	0.7759
No log	1.5385	20	0.5872	0.3497	0.5872	0.7663
No log	1.6923	22	0.5817	0.3147	0.5817	0.7627
No log	1.8462	24	0.6064	0.3114	0.6064	0.7787
No log	2.0	26	0.6017	0.3336	0.6017	0.7757
No log	2.1538	28	0.6110	0.3454	0.6110	0.7817
No log	2.3077	30	0.6669	0.3188	0.6669	0.8167
No log	2.4615	32	0.8718	0.3719	0.8718	0.9337
No log	2.6154	34	0.8568	0.3530	0.8568	0.9256
No log	2.7692	36	0.6757	0.3556	0.6757	0.8220
No log	2.9231	38	0.6166	0.3987	0.6166	0.7853
No log	3.0769	40	0.7121	0.3108	0.7121	0.8438
No log	3.2308	42	0.6923	0.3185	0.6923	0.8321
No log	3.3846	44	0.6044	0.4100	0.6044	0.7774
No log	3.5385	46	0.5683	0.4250	0.5683	0.7538
No log	3.6923	48	0.5858	0.4173	0.5858	0.7654
No log	3.8462	50	0.6309	0.4174	0.6309	0.7943
No log	4.0	52	0.6436	0.4180	0.6436	0.8023
No log	4.1538	54	0.6514	0.4587	0.6514	0.8071
No log	4.3077	56	0.6655	0.4590	0.6655	0.8158
No log	4.4615	58	0.7179	0.4776	0.7179	0.8473
No log	4.6154	60	0.7384	0.4820	0.7384	0.8593
No log	4.7692	62	0.7250	0.4949	0.7250	0.8515
No log	4.9231	64	0.7873	0.4392	0.7873	0.8873
No log	5.0769	66	0.8974	0.3880	0.8974	0.9473
No log	5.2308	68	0.8890	0.3867	0.8890	0.9429
No log	5.3846	70	0.8074	0.4080	0.8074	0.8985
No log	5.5385	72	0.7414	0.4923	0.7414	0.8611
No log	5.6923	74	0.7242	0.5743	0.7242	0.8510
No log	5.8462	76	0.7177	0.5569	0.7177	0.8472
No log	6.0	78	0.7331	0.5267	0.7331	0.8562
No log	6.1538	80	0.7538	0.5014	0.7538	0.8682
No log	6.3077	82	0.8456	0.4874	0.8456	0.9196
No log	6.4615	84	0.9902	0.4031	0.9902	0.9951
No log	6.6154	86	1.0198	0.3989	1.0198	1.0099
No log	6.7692	88	0.9480	0.4784	0.9480	0.9737
No log	6.9231	90	0.8386	0.4671	0.8386	0.9157
No log	7.0769	92	0.8050	0.4900	0.8050	0.8972
No log	7.2308	94	0.8042	0.5054	0.8042	0.8968
No log	7.3846	96	0.8236	0.4938	0.8236	0.9075
No log	7.5385	98	0.8765	0.4909	0.8765	0.9362
No log	7.6923	100	0.9456	0.4852	0.9456	0.9724
No log	7.8462	102	0.9827	0.4852	0.9827	0.9913
No log	8.0	104	1.0079	0.4995	1.0079	1.0040
No log	8.1538	106	1.0054	0.4841	1.0054	1.0027
No log	8.3077	108	1.0068	0.4800	1.0068	1.0034
No log	8.4615	110	1.0035	0.4935	1.0035	1.0018
No log	8.6154	112	1.0099	0.4888	1.0099	1.0050
No log	8.7692	114	1.0142	0.4888	1.0142	1.0071
No log	8.9231	116	1.0269	0.4757	1.0269	1.0134
No log	9.0769	118	1.0426	0.4800	1.0426	1.0211
No log	9.2308	120	1.0508	0.4580	1.0508	1.0251
No log	9.3846	122	1.0499	0.4580	1.0499	1.0247
No log	9.5385	124	1.0448	0.4850	1.0448	1.0221
No log	9.6923	126	1.0471	0.4634	1.0471	1.0233
No log	9.8462	128	1.0522	0.4634	1.0522	1.0258
No log	10.0	130	1.0534	0.4686	1.0534	1.0264

Framework versions

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