ArabicNewSplits4_WithDuplicationsForScore5_FineTuningAraBERT_run1_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: 0.8923
• Qwk: 0.5146
• Mse: 0.8923
• Rmse: 0.9446

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	4.0144	0.0194	4.0144	2.0036
No log	0.3077	4	2.7717	0.0597	2.7717	1.6649
No log	0.4615	6	1.3383	0.0951	1.3383	1.1569
No log	0.6154	8	0.9628	0.0250	0.9628	0.9812
No log	0.7692	10	1.0457	-0.0802	1.0457	1.0226
No log	0.9231	12	0.7361	0.2069	0.7361	0.8579
No log	1.0769	14	0.7430	0.1469	0.7430	0.8620
No log	1.2308	16	0.7424	0.1261	0.7424	0.8616
No log	1.3846	18	0.7168	0.1777	0.7168	0.8466
No log	1.5385	20	0.6743	0.2108	0.6743	0.8211
No log	1.6923	22	0.6584	0.3128	0.6584	0.8114
No log	1.8462	24	0.6677	0.3270	0.6677	0.8171
No log	2.0	26	0.6739	0.3321	0.6739	0.8209
No log	2.1538	28	0.6089	0.3357	0.6089	0.7803
No log	2.3077	30	0.6385	0.2493	0.6385	0.7990
No log	2.4615	32	0.6599	0.2231	0.6599	0.8124
No log	2.6154	34	0.6449	0.2231	0.6449	0.8030
No log	2.7692	36	0.5909	0.3423	0.5909	0.7687
No log	2.9231	38	0.6063	0.3566	0.6063	0.7787
No log	3.0769	40	0.6448	0.3863	0.6448	0.8030
No log	3.2308	42	0.6881	0.3426	0.6881	0.8295
No log	3.3846	44	0.6065	0.3611	0.6065	0.7788
No log	3.5385	46	0.5506	0.4255	0.5506	0.7421
No log	3.6923	48	0.5805	0.4977	0.5805	0.7619
No log	3.8462	50	0.6709	0.5250	0.6709	0.8191
No log	4.0	52	0.7688	0.4631	0.7688	0.8768
No log	4.1538	54	0.7568	0.4758	0.7568	0.8699
No log	4.3077	56	0.6911	0.5194	0.6911	0.8313
No log	4.4615	58	0.6922	0.5178	0.6922	0.8320
No log	4.6154	60	0.7021	0.5224	0.7021	0.8379
No log	4.7692	62	0.7134	0.5219	0.7134	0.8446
No log	4.9231	64	0.7573	0.4912	0.7573	0.8702
No log	5.0769	66	0.7881	0.5042	0.7881	0.8878
No log	5.2308	68	0.8487	0.5370	0.8487	0.9213
No log	5.3846	70	0.8931	0.5274	0.8931	0.9450
No log	5.5385	72	0.9242	0.5164	0.9242	0.9613
No log	5.6923	74	0.9675	0.4608	0.9675	0.9836
No log	5.8462	76	0.9331	0.4626	0.9331	0.9660
No log	6.0	78	0.8574	0.4713	0.8574	0.9259
No log	6.1538	80	0.7944	0.5026	0.7944	0.8913
No log	6.3077	82	0.8113	0.5121	0.8113	0.9007
No log	6.4615	84	0.9064	0.4730	0.9064	0.9520
No log	6.6154	86	0.9760	0.4403	0.9760	0.9879
No log	6.7692	88	0.9304	0.4740	0.9304	0.9646
No log	6.9231	90	0.8666	0.4812	0.8666	0.9309
No log	7.0769	92	0.8457	0.5330	0.8457	0.9196
No log	7.2308	94	0.8679	0.4981	0.8679	0.9316
No log	7.3846	96	0.9112	0.4792	0.9112	0.9546
No log	7.5385	98	0.9357	0.4744	0.9357	0.9673
No log	7.6923	100	0.9652	0.4879	0.9652	0.9824
No log	7.8462	102	0.9416	0.4652	0.9416	0.9703
No log	8.0	104	0.9040	0.4960	0.9040	0.9508
No log	8.1538	106	0.8942	0.5062	0.8942	0.9456
No log	8.3077	108	0.8924	0.4859	0.8924	0.9447
No log	8.4615	110	0.8840	0.5021	0.8840	0.9402
No log	8.6154	112	0.8787	0.4954	0.8787	0.9374
No log	8.7692	114	0.8806	0.5014	0.8806	0.9384
No log	8.9231	116	0.9005	0.5086	0.9005	0.9490
No log	9.0769	118	0.9086	0.4884	0.9086	0.9532
No log	9.2308	120	0.9190	0.5016	0.9190	0.9586
No log	9.3846	122	0.9130	0.4977	0.9130	0.9555
No log	9.5385	124	0.9051	0.5213	0.9051	0.9514
No log	9.6923	126	0.8954	0.5086	0.8954	0.9463
No log	9.8462	128	0.8934	0.5086	0.8934	0.9452
No log	10.0	130	0.8923	0.5146	0.8923	0.9446

Framework versions

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