Modelpublic

MayBashendy/Arabic_FineTuningAraBERT_AugV0_k1_task5_organization_fold0

sourceHugging Faceupdated 2y agoView on Hugging Face

0likes11downloads

Model Card

ArabicFineTuningAraBERTAugV0k1task5organizationfold0

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: 0.5705
—Qwk: 0.7661
—Mse: 0.5705
—Rmse: 0.7553

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
—trainbatchsize: 8
—evalbatchsize: 8
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	0.1538	2	1.8911	0.1557	1.8911	1.3752
No log	0.3077	4	1.1544	0.2282	1.1544	1.0744
No log	0.4615	6	0.9616	0.1958	0.9616	0.9806
No log	0.6154	8	0.7681	0.3050	0.7681	0.8764
No log	0.7692	10	0.8793	0.2578	0.8793	0.9377
No log	0.9231	12	0.9346	0.4138	0.9346	0.9667
No log	1.0769	14	0.9222	0.5251	0.9222	0.9603
No log	1.2308	16	0.7779	0.4133	0.7779	0.8820
No log	1.3846	18	0.6276	0.5486	0.6276	0.7922
No log	1.5385	20	0.8435	0.3433	0.8435	0.9184
No log	1.6923	22	0.9130	0.3548	0.9130	0.9555
No log	1.8462	24	0.7264	0.6084	0.7264	0.8523
No log	2.0	26	0.5036	0.6707	0.5036	0.7096
No log	2.1538	28	0.5710	0.6721	0.5710	0.7556
No log	2.3077	30	0.7267	0.6010	0.7267	0.8524
No log	2.4615	32	0.7347	0.5195	0.7347	0.8571
No log	2.6154	34	0.7381	0.3972	0.7381	0.8591
No log	2.7692	36	0.6200	0.4649	0.6200	0.7874
No log	2.9231	38	0.5352	0.6726	0.5352	0.7316
No log	3.0769	40	0.5973	0.6509	0.5973	0.7729
No log	3.2308	42	0.6528	0.6411	0.6528	0.8080
No log	3.3846	44	0.6204	0.6411	0.6204	0.7877
No log	3.5385	46	0.5662	0.6977	0.5662	0.7525
No log	3.6923	48	0.5360	0.6868	0.5360	0.7321
No log	3.8462	50	0.5181	0.6868	0.5181	0.7198
No log	4.0	52	0.5466	0.6868	0.5466	0.7394
No log	4.1538	54	0.5423	0.7111	0.5423	0.7364
No log	4.3077	56	0.5460	0.7111	0.5460	0.7389
No log	4.4615	58	0.6107	0.7198	0.6107	0.7815
No log	4.6154	60	0.6135	0.7198	0.6135	0.7833
No log	4.7692	62	0.5292	0.7229	0.5292	0.7275
No log	4.9231	64	0.5232	0.7556	0.5232	0.7233
No log	5.0769	66	0.5019	0.7229	0.5019	0.7085
No log	5.2308	68	0.4606	0.7477	0.4606	0.6787
No log	5.3846	70	0.4648	0.7477	0.4648	0.6818
No log	5.5385	72	0.4967	0.7477	0.4967	0.7048
No log	5.6923	74	0.5472	0.6767	0.5472	0.7397
No log	5.8462	76	0.6211	0.7024	0.6211	0.7881
No log	6.0	78	0.6376	0.6806	0.6376	0.7985
No log	6.1538	80	0.5599	0.7040	0.5599	0.7482
No log	6.3077	82	0.5125	0.7269	0.5125	0.7159
No log	6.4615	84	0.4493	0.7269	0.4493	0.6703
No log	6.6154	86	0.4231	0.7269	0.4231	0.6505
No log	6.7692	88	0.4487	0.7269	0.4487	0.6698
No log	6.9231	90	0.4822	0.7269	0.4822	0.6944
No log	7.0769	92	0.5263	0.7269	0.5263	0.7254
No log	7.2308	94	0.5363	0.7269	0.5363	0.7323
No log	7.3846	96	0.5482	0.7269	0.5482	0.7404
No log	7.5385	98	0.6058	0.7843	0.6058	0.7783
No log	7.6923	100	0.6135	0.8102	0.6135	0.7832
No log	7.8462	102	0.5836	0.7843	0.5836	0.7639
No log	8.0	104	0.5647	0.7661	0.5647	0.7514
No log	8.1538	106	0.5660	0.7661	0.5660	0.7523
No log	8.3077	108	0.5570	0.7661	0.5570	0.7463
No log	8.4615	110	0.5803	0.7661	0.5803	0.7618
No log	8.6154	112	0.6152	0.7170	0.6152	0.7844
No log	8.7692	114	0.6225	0.7170	0.6225	0.7890
No log	8.9231	116	0.6243	0.7170	0.6243	0.7901
No log	9.0769	118	0.6087	0.7661	0.6087	0.7802
No log	9.2308	120	0.5899	0.7661	0.5899	0.7681
No log	9.3846	122	0.5790	0.7661	0.5790	0.7609
No log	9.5385	124	0.5734	0.7661	0.5734	0.7572
No log	9.6923	126	0.5707	0.7661	0.5707	0.7555
No log	9.8462	128	0.5699	0.7661	0.5699	0.7549
No log	10.0	130	0.5705	0.7661	0.5705	0.7553

Framework versions

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