ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run3_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: 1.1167
• Qwk: 0.1032
• Mse: 1.1167
• Rmse: 1.0567

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	3.2743	-0.0149	3.2743	1.8095
No log	0.3077	4	1.5006	-0.0070	1.5006	1.2250
No log	0.4615	6	0.7575	0.0108	0.7575	0.8704
No log	0.6154	8	0.6126	0.0569	0.6126	0.7827
No log	0.7692	10	0.7166	0.0	0.7166	0.8465
No log	0.9231	12	0.6134	0.0080	0.6134	0.7832
No log	1.0769	14	0.8368	0.1238	0.8368	0.9148
No log	1.2308	16	0.9829	0.0201	0.9829	0.9914
No log	1.3846	18	0.8342	-0.0175	0.8342	0.9133
No log	1.5385	20	0.6245	-0.0233	0.6245	0.7903
No log	1.6923	22	0.6481	0.0	0.6481	0.8050
No log	1.8462	24	0.6660	-0.0732	0.6660	0.8161
No log	2.0	26	0.6671	-0.0732	0.6671	0.8167
No log	2.1538	28	0.6374	-0.0159	0.6374	0.7983
No log	2.3077	30	0.6216	-0.0233	0.6216	0.7884
No log	2.4615	32	0.6284	-0.0963	0.6284	0.7927
No log	2.6154	34	0.6611	-0.0556	0.6611	0.8131
No log	2.7692	36	0.7495	-0.0424	0.7495	0.8657
No log	2.9231	38	0.9286	0.1705	0.9286	0.9637
No log	3.0769	40	1.3911	0.1455	1.3911	1.1794
No log	3.2308	42	1.1335	0.1565	1.1335	1.0647
No log	3.3846	44	0.6707	0.0886	0.6707	0.8190
No log	3.5385	46	0.6983	0.0504	0.6983	0.8356
No log	3.6923	48	0.6734	0.0423	0.6734	0.8206
No log	3.8462	50	0.6330	0.1529	0.6330	0.7956
No log	4.0	52	0.9775	0.1803	0.9775	0.9887
No log	4.1538	54	1.0033	0.1803	1.0033	1.0016
No log	4.3077	56	0.7142	0.2000	0.7142	0.8451
No log	4.4615	58	0.6375	0.0769	0.6375	0.7984
No log	4.6154	60	0.7100	0.1716	0.7100	0.8426
No log	4.7692	62	0.7385	0.1724	0.7385	0.8594
No log	4.9231	64	0.6904	0.2258	0.6904	0.8309
No log	5.0769	66	0.7025	0.2523	0.7025	0.8382
No log	5.2308	68	0.8426	0.2263	0.8426	0.9179
No log	5.3846	70	0.7181	0.2579	0.7181	0.8474
No log	5.5385	72	0.7410	0.2811	0.7410	0.8608
No log	5.6923	74	1.0574	0.0075	1.0574	1.0283
No log	5.8462	76	1.1195	0.0769	1.1195	1.0581
No log	6.0	78	0.9271	0.1405	0.9271	0.9628
No log	6.1538	80	0.6640	0.3973	0.6640	0.8149
No log	6.3077	82	0.6967	0.2000	0.6967	0.8347
No log	6.4615	84	0.8112	0.2605	0.8112	0.9007
No log	6.6154	86	0.7469	0.1538	0.7469	0.8642
No log	6.7692	88	0.7300	0.2811	0.7300	0.8544
No log	6.9231	90	0.9681	0.0968	0.9681	0.9839
No log	7.0769	92	1.1426	0.0108	1.1426	1.0689
No log	7.2308	94	1.1645	0.0423	1.1645	1.0791
No log	7.3846	96	1.0991	0.0606	1.0991	1.0484
No log	7.5385	98	0.9259	0.2000	0.9259	0.9622
No log	7.6923	100	0.8357	0.3333	0.8357	0.9142
No log	7.8462	102	0.8143	0.3333	0.8143	0.9024
No log	8.0	104	0.8376	0.3305	0.8376	0.9152
No log	8.1538	106	0.8800	0.2941	0.8800	0.9381
No log	8.3077	108	0.9431	0.2903	0.9431	0.9711
No log	8.4615	110	1.0136	0.1161	1.0136	1.0068
No log	8.6154	112	1.0887	0.0722	1.0887	1.0434
No log	8.7692	114	1.1625	0.0861	1.1625	1.0782
No log	8.9231	116	1.1925	0.0882	1.1925	1.0920
No log	9.0769	118	1.2081	0.0882	1.2081	1.0991
No log	9.2308	120	1.2275	0.0882	1.2275	1.1079
No log	9.3846	122	1.2133	0.0882	1.2133	1.1015
No log	9.5385	124	1.1696	0.0861	1.1696	1.0815
No log	9.6923	126	1.1347	0.0861	1.1347	1.0652
No log	9.8462	128	1.1222	0.1111	1.1222	1.0593
No log	10.0	130	1.1167	0.1032	1.1167	1.0567

Framework versions

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