ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run2_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: 0.7871
• Qwk: 0.6447
• Mse: 0.7871
• Rmse: 0.8872

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	2.2345	0.0485	2.2345	1.4948
No log	0.2857	4	1.4442	0.1786	1.4442	1.2018
No log	0.4286	6	1.3162	0.2189	1.3162	1.1472
No log	0.5714	8	1.3225	0.2743	1.3225	1.1500
No log	0.7143	10	1.3808	0.3360	1.3808	1.1751
No log	0.8571	12	1.4099	0.3960	1.4099	1.1874
No log	1.0	14	1.2761	0.3320	1.2761	1.1297
No log	1.1429	16	1.2043	0.3315	1.2043	1.0974
No log	1.2857	18	1.1086	0.3309	1.1086	1.0529
No log	1.4286	20	1.0880	0.3119	1.0880	1.0430
No log	1.5714	22	1.1417	0.3076	1.1417	1.0685
No log	1.7143	24	1.0875	0.3501	1.0875	1.0428
No log	1.8571	26	1.0207	0.3863	1.0207	1.0103
No log	2.0	28	1.1815	0.4540	1.1815	1.0870
No log	2.1429	30	1.2641	0.4353	1.2641	1.1243
No log	2.2857	32	1.2016	0.4807	1.2016	1.0962
No log	2.4286	34	0.9767	0.5649	0.9767	0.9883
No log	2.5714	36	0.9383	0.4861	0.9383	0.9686
No log	2.7143	38	1.0242	0.4078	1.0242	1.0120
No log	2.8571	40	0.9418	0.4743	0.9418	0.9705
No log	3.0	42	0.8639	0.5357	0.8639	0.9295
No log	3.1429	44	0.8816	0.5669	0.8816	0.9390
No log	3.2857	46	1.0140	0.5504	1.0140	1.0070
No log	3.4286	48	1.0228	0.5493	1.0228	1.0114
No log	3.5714	50	0.9324	0.5665	0.9324	0.9656
No log	3.7143	52	0.8924	0.5954	0.8924	0.9447
No log	3.8571	54	0.8557	0.6026	0.8557	0.9251
No log	4.0	56	0.8598	0.5994	0.8598	0.9273
No log	4.1429	58	0.9486	0.5855	0.9486	0.9740
No log	4.2857	60	0.9290	0.6558	0.9290	0.9639
No log	4.4286	62	0.7950	0.6566	0.7950	0.8916
No log	4.5714	64	0.7414	0.6548	0.7414	0.8611
No log	4.7143	66	0.7404	0.6674	0.7404	0.8604
No log	4.8571	68	0.7302	0.6628	0.7302	0.8545
No log	5.0	70	0.7414	0.6314	0.7414	0.8611
No log	5.1429	72	0.7538	0.6366	0.7538	0.8682
No log	5.2857	74	0.8105	0.5701	0.8105	0.9003
No log	5.4286	76	0.9958	0.5723	0.9958	0.9979
No log	5.5714	78	1.2290	0.5329	1.2290	1.1086
No log	5.7143	80	1.2617	0.5566	1.2617	1.1233
No log	5.8571	82	1.1464	0.5638	1.1464	1.0707
No log	6.0	84	0.9108	0.5735	0.9108	0.9544
No log	6.1429	86	0.7885	0.6078	0.7885	0.8880
No log	6.2857	88	0.7923	0.5813	0.7923	0.8901
No log	6.4286	90	0.7789	0.6050	0.7789	0.8826
No log	6.5714	92	0.8129	0.6186	0.8129	0.9016
No log	6.7143	94	0.9353	0.5853	0.9353	0.9671
No log	6.8571	96	1.0102	0.5625	1.0102	1.0051
No log	7.0	98	1.0066	0.5725	1.0066	1.0033
No log	7.1429	100	0.9486	0.5713	0.9486	0.9740
No log	7.2857	102	0.8563	0.6308	0.8563	0.9254
No log	7.4286	104	0.7767	0.6775	0.7767	0.8813
No log	7.5714	106	0.7706	0.6716	0.7706	0.8778
No log	7.7143	108	0.7719	0.6765	0.7719	0.8786
No log	7.8571	110	0.7734	0.6557	0.7734	0.8794
No log	8.0	112	0.7884	0.6559	0.7884	0.8879
No log	8.1429	114	0.8039	0.6513	0.8039	0.8966
No log	8.2857	116	0.8063	0.6496	0.8063	0.8979
No log	8.4286	118	0.8292	0.6451	0.8292	0.9106
No log	8.5714	120	0.8734	0.6255	0.8734	0.9346
No log	8.7143	122	0.9235	0.5992	0.9235	0.9610
No log	8.8571	124	0.9270	0.5869	0.9270	0.9628
No log	9.0	126	0.9003	0.6066	0.9003	0.9488
No log	9.1429	128	0.8617	0.6255	0.8617	0.9283
No log	9.2857	130	0.8261	0.6410	0.8261	0.9089
No log	9.4286	132	0.8052	0.6481	0.8052	0.8974
No log	9.5714	134	0.7918	0.6516	0.7918	0.8898
No log	9.7143	136	0.7877	0.6447	0.7877	0.8875
No log	9.8571	138	0.7860	0.6447	0.7860	0.8866
No log	10.0	140	0.7871	0.6447	0.7871	0.8872

Framework versions

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