deberta-semeval25_EN08_fold3

This model is a fine-tuned version of microsoft/deberta-v3-base on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 9.1157
  • Precision Samples: 0.1421
  • Recall Samples: 0.5756
  • F1 Samples: 0.2085
  • Precision Macro: 0.7606
  • Recall Macro: 0.3423
  • F1 Macro: 0.2156
  • Precision Micro: 0.1314
  • Recall Micro: 0.4788
  • F1 Micro: 0.2062
  • Precision Weighted: 0.4667
  • Recall Weighted: 0.4788
  • F1 Weighted: 0.1520

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: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Precision Samples Recall Samples F1 Samples Precision Macro Recall Macro F1 Macro Precision Micro Recall Micro F1 Micro Precision Weighted Recall Weighted F1 Weighted
11.095 1.0 19 10.6400 1.0 0.0 0.0 1.0 0.1556 0.1556 1.0 0.0 0.0 1.0 0.0 0.0
10.2278 2.0 38 10.2333 0.1374 0.3013 0.1697 0.9615 0.1926 0.1657 0.1442 0.1785 0.1595 0.8356 0.1785 0.0527
9.5925 3.0 57 9.9327 0.1375 0.3699 0.1830 0.9031 0.2134 0.1735 0.1336 0.2408 0.1719 0.6911 0.2408 0.0774
8.9765 4.0 76 9.6535 0.1451 0.4784 0.2044 0.8754 0.2701 0.1879 0.1362 0.3711 0.1992 0.6423 0.3711 0.1069
9.0712 5.0 95 9.4718 0.1306 0.5154 0.1935 0.8449 0.2948 0.1925 0.1264 0.4079 0.1930 0.5821 0.4079 0.1145
8.8157 6.0 114 9.2988 0.1297 0.5550 0.1938 0.8061 0.3186 0.2014 0.1247 0.4561 0.1959 0.5153 0.4561 0.1305
9.2663 7.0 133 9.3128 0.1386 0.5483 0.2044 0.7855 0.3211 0.2127 0.1323 0.4504 0.2045 0.4887 0.4504 0.1487
8.1988 8.0 152 9.1760 0.1436 0.5540 0.2084 0.7656 0.3217 0.2097 0.1331 0.4561 0.2060 0.4778 0.4561 0.1483
8.7719 9.0 171 9.1413 0.1400 0.5696 0.2045 0.7801 0.3316 0.2110 0.1295 0.4674 0.2028 0.4820 0.4674 0.1482
8.6954 10.0 190 9.1157 0.1421 0.5756 0.2085 0.7606 0.3423 0.2156 0.1314 0.4788 0.2062 0.4667 0.4788 0.1520

Framework versions

  • Transformers 4.46.0
  • Pytorch 2.3.1
  • Datasets 2.21.0
  • Tokenizers 0.20.1
Downloads last month
13
Safetensors
Model size
184M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for g-assismoraes/deberta-semeval25_EN08_fold3

Finetuned
(268)
this model