makhataei's picture
End of training
5dcac3a verified
|
raw
history blame
6.45 kB
metadata
license: mit
base_model: makhataei/qa-persian-mdeberta-v3-base-squad2
tags:
  - generated_from_trainer
model-index:
  - name: qa-persian-mdeberta-v3-base-squad2
    results: []

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: 0.0005
  • 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
5.111 1.0 17 5.1273
5.1193 2.0 34 5.1273
5.1078 3.0 51 5.1273
5.1228 4.0 68 5.1273
5.1136 5.0 85 5.1273
5.0927 6.0 102 5.1273
5.0957 7.0 119 5.1273
5.095 8.0 136 5.1273
5.0955 9.0 153 5.1273
5.0885 10.0 170 5.1273
5.0845 11.0 187 5.1273
5.0957 12.0 204 5.1273
5.1226 13.0 221 5.1273
5.1309 14.0 238 5.1273
5.0945 15.0 255 5.1273
5.1088 16.0 272 5.1273
5.1217 17.0 289 5.1273
5.106 18.0 306 5.1273
5.1087 19.0 323 5.1273
5.0938 20.0 340 5.1273
5.0991 21.0 357 5.1273
5.1055 22.0 374 5.1273
5.111 23.0 391 5.1273
5.0891 24.0 408 5.1273
5.0975 25.0 425 5.1273
5.1118 26.0 442 5.1273
5.1243 27.0 459 5.1273
5.0983 28.0 476 5.1273
5.0957 29.0 493 5.1273
5.0902 30.0 510 5.1273
5.1136 31.0 527 5.1273
5.1265 32.0 544 5.1273
5.0934 33.0 561 5.1273
5.1181 34.0 578 5.1273
5.1073 35.0 595 5.1273
5.1229 36.0 612 5.1273
5.1245 37.0 629 5.1273
5.1194 38.0 646 5.1273
5.1061 39.0 663 5.1273
5.1074 40.0 680 5.1273
5.1171 41.0 697 5.1273
5.0937 42.0 714 5.1273
5.1348 43.0 731 5.1273
5.1332 44.0 748 5.1273
5.1284 45.0 765 5.1273
5.1244 46.0 782 5.1273
5.1383 47.0 799 5.1273
5.1139 48.0 816 5.1273
5.1148 49.0 833 5.1273
5.1059 50.0 850 5.1273
5.1214 51.0 867 5.1273
5.1363 52.0 884 5.1273
5.1318 53.0 901 5.1273
5.1278 54.0 918 5.1273
5.1175 55.0 935 5.1273
5.1292 56.0 952 5.1273
5.1272 57.0 969 5.1273
5.1151 58.0 986 5.1273
5.1071 59.0 1003 5.1273
5.1274 60.0 1020 5.1273
5.1197 61.0 1037 5.1273
5.127 62.0 1054 5.1273
5.1143 63.0 1071 5.1273
5.1158 64.0 1088 5.1273
5.1444 65.0 1105 5.1273
5.1225 66.0 1122 5.1273
5.1298 67.0 1139 5.1273
5.1308 68.0 1156 5.1273
5.0992 69.0 1173 5.1273
5.1287 70.0 1190 5.1273
5.1615 71.0 1207 5.1273
5.1434 72.0 1224 5.1273
5.1388 73.0 1241 5.1273
5.1334 74.0 1258 5.1273
5.1329 75.0 1275 5.1273
5.1276 76.0 1292 5.1273
5.1257 77.0 1309 5.1273
5.1174 78.0 1326 5.1273
5.1257 79.0 1343 5.1273
5.141 80.0 1360 5.1273
5.1339 81.0 1377 5.1273
5.1455 82.0 1394 5.1273
5.0912 83.0 1411 5.1273
5.1391 84.0 1428 5.1273
5.1367 85.0 1445 5.1273
5.1402 86.0 1462 5.1273
5.1413 87.0 1479 5.1273
5.1176 88.0 1496 5.1273
5.1492 89.0 1513 5.1273
5.1104 90.0 1530 5.1273
5.1255 91.0 1547 5.1273
5.1406 92.0 1564 5.1273
5.1245 93.0 1581 5.1273
5.1398 94.0 1598 5.1273
5.1293 95.0 1615 5.1273
5.1343 96.0 1632 5.1273
5.1222 97.0 1649 5.1273
5.1238 98.0 1666 5.1273
5.1221 99.0 1683 5.1273
5.1336 100.0 1700 5.1273

Framework versions

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