ArabicNewSplits6_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: 0.5863
• Qwk: 0.7383
• Mse: 0.5863
• Rmse: 0.7657

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	2.1424	0.0392	2.1424	1.4637
No log	0.3077	4	1.4228	0.2019	1.4228	1.1928
No log	0.4615	6	1.3512	0.2156	1.3512	1.1624
No log	0.6154	8	1.5465	0.3246	1.5465	1.2436
No log	0.7692	10	1.4248	0.1797	1.4248	1.1937
No log	0.9231	12	1.4367	0.1549	1.4367	1.1986
No log	1.0769	14	1.4150	0.1639	1.4150	1.1895
No log	1.2308	16	1.5234	0.3116	1.5234	1.2343
No log	1.3846	18	1.3895	0.3454	1.3895	1.1788
No log	1.5385	20	1.2064	0.3051	1.2064	1.0984
No log	1.6923	22	1.1051	0.3611	1.1051	1.0512
No log	1.8462	24	1.0332	0.4010	1.0332	1.0165
No log	2.0	26	0.9844	0.4430	0.9844	0.9922
No log	2.1538	28	0.9614	0.5330	0.9614	0.9805
No log	2.3077	30	0.9503	0.5319	0.9503	0.9749
No log	2.4615	32	0.8952	0.5312	0.8952	0.9462
No log	2.6154	34	0.8392	0.6225	0.8392	0.9161
No log	2.7692	36	0.8736	0.5780	0.8736	0.9346
No log	2.9231	38	0.8712	0.5494	0.8712	0.9334
No log	3.0769	40	0.9476	0.4966	0.9476	0.9734
No log	3.2308	42	0.8074	0.6178	0.8074	0.8985
No log	3.3846	44	0.7086	0.6766	0.7086	0.8418
No log	3.5385	46	0.6797	0.6711	0.6797	0.8244
No log	3.6923	48	0.6767	0.6776	0.6767	0.8226
No log	3.8462	50	0.6620	0.6891	0.6620	0.8136
No log	4.0	52	0.6544	0.6882	0.6544	0.8090
No log	4.1538	54	0.6656	0.6988	0.6656	0.8158
No log	4.3077	56	0.6465	0.7035	0.6465	0.8041
No log	4.4615	58	0.6427	0.7130	0.6427	0.8017
No log	4.6154	60	0.6437	0.6957	0.6437	0.8023
No log	4.7692	62	0.6245	0.7088	0.6245	0.7902
No log	4.9231	64	0.6212	0.7432	0.6212	0.7882
No log	5.0769	66	0.6527	0.7450	0.6527	0.8079
No log	5.2308	68	0.6142	0.7336	0.6142	0.7837
No log	5.3846	70	0.6316	0.7206	0.6316	0.7947
No log	5.5385	72	0.6868	0.6522	0.6868	0.8287
No log	5.6923	74	0.6916	0.6522	0.6916	0.8316
No log	5.8462	76	0.6466	0.7025	0.6466	0.8041
No log	6.0	78	0.6069	0.7184	0.6069	0.7791
No log	6.1538	80	0.6068	0.7184	0.6068	0.7790
No log	6.3077	82	0.6161	0.7123	0.6161	0.7849
No log	6.4615	84	0.6503	0.6501	0.6503	0.8064
No log	6.6154	86	0.6488	0.6501	0.6488	0.8055
No log	6.7692	88	0.6133	0.7322	0.6133	0.7831
No log	6.9231	90	0.6196	0.7493	0.6196	0.7871
No log	7.0769	92	0.6246	0.7469	0.6246	0.7903
No log	7.2308	94	0.6128	0.7482	0.6128	0.7828
No log	7.3846	96	0.5996	0.7523	0.5996	0.7743
No log	7.5385	98	0.6055	0.7370	0.6055	0.7781
No log	7.6923	100	0.6069	0.7416	0.6069	0.7791
No log	7.8462	102	0.5985	0.7344	0.5985	0.7736
No log	8.0	104	0.5919	0.7383	0.5919	0.7693
No log	8.1538	106	0.5890	0.7375	0.5890	0.7674
No log	8.3077	108	0.5907	0.7474	0.5907	0.7685
No log	8.4615	110	0.5903	0.7554	0.5903	0.7683
No log	8.6154	112	0.5882	0.7554	0.5882	0.7670
No log	8.7692	114	0.5852	0.7251	0.5852	0.7650
No log	8.9231	116	0.5891	0.7407	0.5891	0.7675
No log	9.0769	118	0.5915	0.7399	0.5915	0.7691
No log	9.2308	120	0.5964	0.7272	0.5964	0.7723
No log	9.3846	122	0.5958	0.7360	0.5958	0.7719
No log	9.5385	124	0.5922	0.7399	0.5922	0.7696
No log	9.6923	126	0.5886	0.7367	0.5886	0.7672
No log	9.8462	128	0.5865	0.7383	0.5865	0.7659
No log	10.0	130	0.5863	0.7383	0.5863	0.7657

Framework versions

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