ArabicNewSplits6_FineTuningAraBERT_run1_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: 0.8555
• Qwk: 0.6963
• Mse: 0.8555
• Rmse: 0.9249

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	2.1844	0.0446	2.1844	1.4780
No log	0.3077	4	1.4821	0.1945	1.4821	1.2174
No log	0.4615	6	1.3793	0.1711	1.3793	1.1744
No log	0.6154	8	1.5486	0.2677	1.5486	1.2444
No log	0.7692	10	1.7001	0.2507	1.7001	1.3039
No log	0.9231	12	1.6164	0.2958	1.6164	1.2714
No log	1.0769	14	1.4029	0.3973	1.4029	1.1844
No log	1.2308	16	1.2207	0.2497	1.2207	1.1049
No log	1.3846	18	1.1455	0.2689	1.1455	1.0703
No log	1.5385	20	1.1056	0.3200	1.1056	1.0515
No log	1.6923	22	1.1009	0.3912	1.1009	1.0492
No log	1.8462	24	1.0975	0.4201	1.0975	1.0476
No log	2.0	26	1.0554	0.4446	1.0554	1.0273
No log	2.1538	28	1.0154	0.5112	1.0154	1.0076
No log	2.3077	30	0.9745	0.5384	0.9745	0.9872
No log	2.4615	32	1.0780	0.5620	1.0780	1.0383
No log	2.6154	34	1.0782	0.5620	1.0782	1.0383
No log	2.7692	36	0.9194	0.5927	0.9194	0.9589
No log	2.9231	38	0.8388	0.6158	0.8388	0.9159
No log	3.0769	40	0.8256	0.6093	0.8256	0.9086
No log	3.2308	42	0.8112	0.6261	0.8112	0.9007
No log	3.3846	44	0.8062	0.6108	0.8062	0.8979
No log	3.5385	46	0.8090	0.5885	0.8090	0.8994
No log	3.6923	48	0.8115	0.5953	0.8115	0.9009
No log	3.8462	50	0.8455	0.6218	0.8455	0.9195
No log	4.0	52	0.9307	0.6062	0.9307	0.9647
No log	4.1538	54	0.9616	0.5965	0.9616	0.9806
No log	4.3077	56	0.9293	0.6157	0.9293	0.9640
No log	4.4615	58	0.8720	0.6154	0.8720	0.9338
No log	4.6154	60	0.8730	0.6224	0.8730	0.9344
No log	4.7692	62	0.8428	0.6266	0.8428	0.9181
No log	4.9231	64	0.8252	0.6490	0.8252	0.9084
No log	5.0769	66	0.8515	0.6648	0.8515	0.9228
No log	5.2308	68	0.8552	0.6411	0.8552	0.9247
No log	5.3846	70	0.8265	0.6609	0.8265	0.9091
No log	5.5385	72	0.8956	0.6659	0.8956	0.9464
No log	5.6923	74	1.0355	0.6312	1.0355	1.0176
No log	5.8462	76	1.0577	0.6223	1.0577	1.0285
No log	6.0	78	0.9528	0.6451	0.9528	0.9761
No log	6.1538	80	0.8924	0.6519	0.8924	0.9447
No log	6.3077	82	0.8547	0.6397	0.8547	0.9245
No log	6.4615	84	0.8633	0.6595	0.8633	0.9292
No log	6.6154	86	0.9348	0.6813	0.9348	0.9668
No log	6.7692	88	0.9883	0.6824	0.9883	0.9941
No log	6.9231	90	1.0505	0.6440	1.0505	1.0249
No log	7.0769	92	1.0494	0.6565	1.0494	1.0244
No log	7.2308	94	0.9800	0.6435	0.9800	0.9899
No log	7.3846	96	0.8715	0.6775	0.8715	0.9335
No log	7.5385	98	0.7948	0.6847	0.7948	0.8915
No log	7.6923	100	0.7820	0.6847	0.7820	0.8843
No log	7.8462	102	0.8113	0.6773	0.8113	0.9007
No log	8.0	104	0.8981	0.6789	0.8981	0.9477
No log	8.1538	106	0.9710	0.6613	0.9710	0.9854
No log	8.3077	108	1.0228	0.6507	1.0228	1.0113
No log	8.4615	110	1.0312	0.6270	1.0312	1.0155
No log	8.6154	112	1.0066	0.6366	1.0066	1.0033
No log	8.7692	114	0.9597	0.6561	0.9597	0.9796
No log	8.9231	116	0.9050	0.6789	0.9050	0.9513
No log	9.0769	118	0.8539	0.6712	0.8539	0.9241
No log	9.2308	120	0.8389	0.6755	0.8389	0.9159
No log	9.3846	122	0.8382	0.6783	0.8382	0.9155
No log	9.5385	124	0.8379	0.6928	0.8379	0.9154
No log	9.6923	126	0.8449	0.6963	0.8449	0.9192
No log	9.8462	128	0.8528	0.6963	0.8528	0.9235
No log	10.0	130	0.8555	0.6963	0.8555	0.9249

Framework versions

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