ArabicNewSplits6_FineTuningAraBERT_run1_AugV5_k2_task3_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: 0.7997
• Qwk: 0.1724
• Mse: 0.7997
• Rmse: 0.8943

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	0.1667	2	3.1916	-0.0053	3.1916	1.7865
No log	0.3333	4	1.6059	-0.0070	1.6059	1.2672
No log	0.5	6	1.4059	0.0294	1.4059	1.1857
No log	0.6667	8	0.9984	0.0661	0.9984	0.9992
No log	0.8333	10	0.6967	-0.1156	0.6967	0.8347
No log	1.0	12	1.0106	-0.0000	1.0106	1.0053
No log	1.1667	14	1.1531	0.0	1.1531	1.0738
No log	1.3333	16	0.9787	0.0140	0.9787	0.9893
No log	1.5	18	0.6957	-0.0350	0.6957	0.8341
No log	1.6667	20	0.6888	0.0448	0.6888	0.8299
No log	1.8333	22	0.6904	-0.0072	0.6904	0.8309
No log	2.0	24	0.6256	0.1448	0.6256	0.7910
No log	2.1667	26	0.6621	0.1905	0.6621	0.8137
No log	2.3333	28	0.6149	0.2099	0.6149	0.7842
No log	2.5	30	0.6447	0.2215	0.6447	0.8029
No log	2.6667	32	0.8259	0.1588	0.8259	0.9088
No log	2.8333	34	0.7486	0.2146	0.7486	0.8652
No log	3.0	36	0.6662	0.3231	0.6662	0.8162
No log	3.1667	38	0.7327	0.1636	0.7327	0.8560
No log	3.3333	40	0.7363	0.1464	0.7363	0.8581
No log	3.5	42	0.6735	0.3761	0.6735	0.8207
No log	3.6667	44	0.8324	0.2066	0.8324	0.9124
No log	3.8333	46	0.9862	0.1331	0.9862	0.9931
No log	4.0	48	0.9003	0.1235	0.9003	0.9488
No log	4.1667	50	0.6919	0.2146	0.6919	0.8318
No log	4.3333	52	0.7885	0.1545	0.7885	0.8880
No log	4.5	54	0.7488	0.1525	0.7488	0.8653
No log	4.6667	56	0.9428	0.1027	0.9428	0.9710
No log	4.8333	58	1.0903	0.1655	1.0903	1.0442
No log	5.0	60	0.9721	0.1027	0.9721	0.9860
No log	5.1667	62	0.7543	0.1730	0.7543	0.8685
No log	5.3333	64	0.6909	0.2744	0.6909	0.8312
No log	5.5	66	0.7945	0.1799	0.7945	0.8913
No log	5.6667	68	0.7324	0.2356	0.7324	0.8558
No log	5.8333	70	0.7639	0.1786	0.7639	0.8740
No log	6.0	72	0.9415	0.0947	0.9415	0.9703
No log	6.1667	74	0.8902	0.0947	0.8902	0.9435
No log	6.3333	76	0.9078	0.0947	0.9078	0.9528
No log	6.5	78	0.8132	0.1464	0.8132	0.9018
No log	6.6667	80	0.7434	0.1402	0.7434	0.8622
No log	6.8333	82	0.7880	0.1730	0.7880	0.8877
No log	7.0	84	0.8799	0.1545	0.8799	0.9380
No log	7.1667	86	1.0600	0.0769	1.0600	1.0295
No log	7.3333	88	1.0070	0.1579	1.0070	1.0035
No log	7.5	90	0.8681	0.1799	0.8681	0.9317
No log	7.6667	92	0.7406	0.2632	0.7406	0.8606
No log	7.8333	94	0.7633	0.2348	0.7633	0.8737
No log	8.0	96	0.8369	0.1795	0.8369	0.9148
No log	8.1667	98	0.9645	0.1807	0.9645	0.9821
No log	8.3333	100	1.0741	0.1317	1.0741	1.0364
No log	8.5	102	1.1648	0.1622	1.1648	1.0793
No log	8.6667	104	1.1049	0.1622	1.1049	1.0511
No log	8.8333	106	0.9734	0.1545	0.9734	0.9866
No log	9.0	108	0.9232	0.1799	0.9232	0.9608
No log	9.1667	110	0.8710	0.1795	0.8710	0.9332
No log	9.3333	112	0.8142	0.1790	0.8142	0.9023
No log	9.5	114	0.7797	0.2605	0.7797	0.8830
No log	9.6667	116	0.7866	0.2000	0.7866	0.8869
No log	9.8333	118	0.7905	0.1724	0.7905	0.8891
No log	10.0	120	0.7997	0.1724	0.7997	0.8943

Framework versions

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