ArabicNewSplits6_WithDuplicationsForScore5_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: 0.8650
• Qwk: 0.6919
• Mse: 0.8650
• Rmse: 0.9300

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	2.0346	0.0362	2.0346	1.4264
No log	0.2857	4	1.5391	0.1368	1.5391	1.2406
No log	0.4286	6	1.5160	0.1757	1.5160	1.2313
No log	0.5714	8	1.5114	0.3157	1.5114	1.2294
No log	0.7143	10	1.6108	0.3555	1.6108	1.2692
No log	0.8571	12	1.6283	0.3651	1.6283	1.2761
No log	1.0	14	1.7242	0.3671	1.7242	1.3131
No log	1.1429	16	1.5555	0.3906	1.5555	1.2472
No log	1.2857	18	1.4191	0.4412	1.4191	1.1913
No log	1.4286	20	1.2309	0.4319	1.2309	1.1094
No log	1.5714	22	1.1050	0.4615	1.1050	1.0512
No log	1.7143	24	1.0800	0.5323	1.0800	1.0392
No log	1.8571	26	1.0299	0.5578	1.0299	1.0148
No log	2.0	28	1.0225	0.5368	1.0225	1.0112
No log	2.1429	30	0.9996	0.5088	0.9996	0.9998
No log	2.2857	32	0.9550	0.5342	0.9550	0.9772
No log	2.4286	34	0.9319	0.6098	0.9319	0.9653
No log	2.5714	36	0.9668	0.5837	0.9668	0.9833
No log	2.7143	38	1.0704	0.6009	1.0704	1.0346
No log	2.8571	40	1.1417	0.5519	1.1417	1.0685
No log	3.0	42	1.0984	0.5814	1.0984	1.0480
No log	3.1429	44	1.0081	0.5746	1.0081	1.0040
No log	3.2857	46	0.9283	0.5491	0.9283	0.9635
No log	3.4286	48	0.9378	0.5628	0.9378	0.9684
No log	3.5714	50	0.9137	0.5770	0.9137	0.9559
No log	3.7143	52	0.9853	0.5956	0.9853	0.9926
No log	3.8571	54	1.1063	0.5648	1.1063	1.0518
No log	4.0	56	0.9839	0.6097	0.9839	0.9919
No log	4.1429	58	0.9951	0.6402	0.9951	0.9975
No log	4.2857	60	1.0231	0.6480	1.0231	1.0115
No log	4.4286	62	1.1371	0.5979	1.1371	1.0664
No log	4.5714	64	1.3151	0.5509	1.3151	1.1468
No log	4.7143	66	1.3119	0.5601	1.3119	1.1454
No log	4.8571	68	1.1709	0.6021	1.1709	1.0821
No log	5.0	70	1.0425	0.6381	1.0425	1.0210
No log	5.1429	72	0.9332	0.6646	0.9332	0.9660
No log	5.2857	74	0.8774	0.6686	0.8774	0.9367
No log	5.4286	76	1.0476	0.6181	1.0476	1.0235
No log	5.5714	78	1.3138	0.5745	1.3138	1.1462
No log	5.7143	80	1.3322	0.5717	1.3322	1.1542
No log	5.8571	82	1.1701	0.5991	1.1701	1.0817
No log	6.0	84	1.0119	0.6352	1.0119	1.0059
No log	6.1429	86	1.0154	0.6282	1.0154	1.0077
No log	6.2857	88	0.9122	0.6948	0.9122	0.9551
No log	6.4286	90	0.8392	0.6940	0.8392	0.9161
No log	6.5714	92	0.8675	0.6922	0.8675	0.9314
No log	6.7143	94	0.9784	0.6870	0.9784	0.9892
No log	6.8571	96	1.1947	0.6303	1.1947	1.0930
No log	7.0	98	1.3446	0.6099	1.3446	1.1596
No log	7.1429	100	1.2833	0.6099	1.2833	1.1328
No log	7.2857	102	1.1037	0.6381	1.1037	1.0506
No log	7.4286	104	0.9545	0.7039	0.9545	0.9770
No log	7.5714	106	0.8584	0.6974	0.8584	0.9265
No log	7.7143	108	0.8527	0.6974	0.8527	0.9234
No log	7.8571	110	0.8452	0.6974	0.8452	0.9193
No log	8.0	112	0.8963	0.6939	0.8963	0.9467
No log	8.1429	114	0.9788	0.6983	0.9788	0.9894
No log	8.2857	116	1.0606	0.6575	1.0606	1.0299
No log	8.4286	118	1.1104	0.6333	1.1104	1.0537
No log	8.5714	120	1.1105	0.6333	1.1105	1.0538
No log	8.7143	122	1.1374	0.6279	1.1374	1.0665
No log	8.8571	124	1.1385	0.6333	1.1385	1.0670
No log	9.0	126	1.0880	0.6426	1.0880	1.0431
No log	9.1429	128	1.0170	0.6734	1.0170	1.0085
No log	9.2857	130	0.9619	0.6962	0.9619	0.9808
No log	9.4286	132	0.9065	0.6917	0.9065	0.9521
No log	9.5714	134	0.8730	0.6830	0.8730	0.9343
No log	9.7143	136	0.8644	0.6919	0.8644	0.9297
No log	9.8571	138	0.8654	0.6919	0.8654	0.9302
No log	10.0	140	0.8650	0.6919	0.8650	0.9300

Framework versions

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