qa-persian-mdeberta-v3-base-squad2

This model is a fine-tuned version of makhataei/qa-persian-mdeberta-v3-base-squad2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 5.1273

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: 1e-07
• 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
4.8591	1.0	17	5.1273
4.9328	2.0	34	5.1273
4.9113	3.0	51	5.1273
4.8829	4.0	68	5.1273
4.9652	5.0	85	5.1273
4.8964	6.0	102	5.1273
4.8316	7.0	119	5.1273
4.8343	8.0	136	5.1273
4.8819	9.0	153	5.1273
4.8986	10.0	170	5.1273
4.9171	11.0	187	5.1273
4.9171	12.0	204	5.1273
4.8551	13.0	221	5.1273
4.8842	14.0	238	5.1273
4.8491	15.0	255	5.1273
4.8761	16.0	272	5.1273
4.8561	17.0	289	5.1273
4.8983	18.0	306	5.1273
4.844	19.0	323	5.1273
4.8985	20.0	340	5.1273
4.8629	21.0	357	5.1273
4.9092	22.0	374	5.1273
4.8956	23.0	391	5.1273
4.8141	24.0	408	5.1273
4.9507	25.0	425	5.1273
4.9157	26.0	442	5.1273
4.8573	27.0	459	5.1273
4.8307	28.0	476	5.1273
4.8523	29.0	493	5.1273
4.8635	30.0	510	5.1273
4.9477	31.0	527	5.1273
4.9748	32.0	544	5.1273
4.919	33.0	561	5.1273
4.9646	34.0	578	5.1273
4.906	35.0	595	5.1273
4.8959	36.0	612	5.1273
4.9289	37.0	629	5.1273
4.9716	38.0	646	5.1273
4.9264	39.0	663	5.1273
4.9511	40.0	680	5.1273
4.9559	41.0	697	5.1273
4.8925	42.0	714	5.1273
5.0009	43.0	731	5.1273
5.0171	44.0	748	5.1273
4.9922	45.0	765	5.1273
4.9442	46.0	782	5.1273
4.9718	47.0	799	5.1273
4.9723	48.0	816	5.1273
4.9641	49.0	833	5.1273
4.9623	50.0	850	5.1273
4.9661	51.0	867	5.1273
4.9751	52.0	884	5.1273
5.0271	53.0	901	5.1273
5.0186	54.0	918	5.1273
4.9842	55.0	935	5.1273
4.9846	56.0	952	5.1273
5.0147	57.0	969	5.1273
5.0593	58.0	986	5.1273
4.9787	59.0	1003	5.1273
5.0569	60.0	1020	5.1273
5.0237	61.0	1037	5.1273
4.9641	62.0	1054	5.1273
5.048	63.0	1071	5.1273
4.9656	64.0	1088	5.1273
5.0632	65.0	1105	5.1273
5.0678	66.0	1122	5.1273
5.0426	67.0	1139	5.1273
5.0061	68.0	1156	5.1273
4.9887	69.0	1173	5.1273
5.0531	70.0	1190	5.1273
5.1073	71.0	1207	5.1273
5.016	72.0	1224	5.1273
5.0383	73.0	1241	5.1273
5.032	74.0	1258	5.1273
5.0459	75.0	1275	5.1273
5.0734	76.0	1292	5.1273
5.0059	77.0	1309	5.1273
5.027	78.0	1326	5.1273
5.0383	79.0	1343	5.1273
5.104	80.0	1360	5.1273
5.0209	81.0	1377	5.1273
5.0443	82.0	1394	5.1273
4.9923	83.0	1411	5.1273
5.0462	84.0	1428	5.1273
5.0416	85.0	1445	5.1273
5.0593	86.0	1462	5.1273
5.1125	87.0	1479	5.1273
5.0125	88.0	1496	5.1273
5.0925	89.0	1513	5.1273
5.0681	90.0	1530	5.1273
5.0962	91.0	1547	5.1273
5.0843	92.0	1564	5.1273
5.0987	93.0	1581	5.1273
5.0251	94.0	1598	5.1273
5.0904	95.0	1615	5.1273
5.1356	96.0	1632	5.1273
5.1103	97.0	1649	5.1273
5.0244	98.0	1666	5.1273
5.0914	99.0	1683	5.1273
5.1083	100.0	1700	5.1273

Framework versions

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