qa-fa-mdeberta-v3-base

This model is a fine-tuned version of makhataei/qa-fa-mdeberta-v3-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 5.5578

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: 7.8125e-10
• train_batch_size: 14
• eval_batch_size: 14
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss
5.547	1.0	18	5.5578
5.5724	2.0	36	5.5578
5.558	3.0	54	5.5578
5.5752	4.0	72	5.5578
5.5684	5.0	90	5.5578
5.5479	6.0	108	5.5578
5.5724	7.0	126	5.5578
5.5792	8.0	144	5.5578
5.5603	9.0	162	5.5578
5.5868	10.0	180	5.5578
5.5626	11.0	198	5.5578
5.5889	12.0	216	5.5578
5.5413	13.0	234	5.5578
5.5526	14.0	252	5.5578
5.5584	15.0	270	5.5578
5.5539	16.0	288	5.5578
5.5728	17.0	306	5.5578
5.5584	18.0	324	5.5578
5.5555	19.0	342	5.5578
5.5809	20.0	360	5.5578
5.577	21.0	378	5.5578
5.5784	22.0	396	5.5578
5.5537	23.0	414	5.5578
5.6048	24.0	432	5.5578
5.5687	25.0	450	5.5578
5.5683	26.0	468	5.5578
5.5949	27.0	486	5.5578
5.5585	28.0	504	5.5578
5.5477	29.0	522	5.5578
5.5668	30.0	540	5.5578
5.5919	31.0	558	5.5578
5.5527	32.0	576	5.5578
5.5661	33.0	594	5.5578
5.589	34.0	612	5.5578
5.579	35.0	630	5.5578
5.5495	36.0	648	5.5578
5.5671	37.0	666	5.5578
5.5379	38.0	684	5.5578
5.54	39.0	702	5.5578
5.559	40.0	720	5.5578
5.5825	41.0	738	5.5578
5.5422	42.0	756	5.5578
5.5507	43.0	774	5.5578
5.5464	44.0	792	5.5578
5.5746	45.0	810	5.5578
5.5704	46.0	828	5.5578
5.559	47.0	846	5.5578
5.5813	48.0	864	5.5578
5.5634	49.0	882	5.5578
5.5797	50.0	900	5.5578
5.545	51.0	918	5.5578
5.5357	52.0	936	5.5578
5.6026	53.0	954	5.5578
5.5914	54.0	972	5.5578
5.5708	55.0	990	5.5578
5.5938	56.0	1008	5.5578
5.5768	57.0	1026	5.5578
5.5647	58.0	1044	5.5578
5.5822	59.0	1062	5.5578
5.5632	60.0	1080	5.5578
5.5508	61.0	1098	5.5578
5.559	62.0	1116	5.5578
5.5485	63.0	1134	5.5578
5.5532	64.0	1152	5.5578
5.5877	65.0	1170	5.5578
5.5546	66.0	1188	5.5578
5.5623	67.0	1206	5.5578
5.5603	68.0	1224	5.5578
5.5697	69.0	1242	5.5578
5.5674	70.0	1260	5.5578
5.5506	71.0	1278	5.5578
5.5451	72.0	1296	5.5578
5.5678	73.0	1314	5.5578
5.5547	74.0	1332	5.5578
5.5799	75.0	1350	5.5578
5.5647	76.0	1368	5.5578
5.5858	77.0	1386	5.5578
5.6046	78.0	1404	5.5578
5.5658	79.0	1422	5.5578
5.5844	80.0	1440	5.5578
5.583	81.0	1458	5.5578
5.5796	82.0	1476	5.5578
5.5706	83.0	1494	5.5578
5.576	84.0	1512	5.5578
5.5662	85.0	1530	5.5578
5.5903	86.0	1548	5.5578
5.5475	87.0	1566	5.5578
5.5882	88.0	1584	5.5578
5.5492	89.0	1602	5.5578
5.5985	90.0	1620	5.5578
5.5673	91.0	1638	5.5578
5.554	92.0	1656	5.5578
5.5894	93.0	1674	5.5578
5.5466	94.0	1692	5.5578
5.56	95.0	1710	5.5578
5.5847	96.0	1728	5.5578
5.5732	97.0	1746	5.5578
5.5662	98.0	1764	5.5578
5.5647	99.0	1782	5.5578
5.5472	100.0	1800	5.5578

Framework versions

• Transformers 4.35.2
• Pytorch 2.0.1+cu117
• Datasets 2.15.0
• Tokenizers 0.15.0