Hugging Face's logo

ahotrod
/
deberta-v3-large-finetuned-squadv2

Question Answering
Transformers
PyTorch
deberta-v2
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
1
deberta-v3-large-finetuned-squadv2 / README.md
ahotrod's picture
ahotrod
ahotrod/deberta-v3-large-finetuned-squadv2
cca586b
preview code
|
raw
history blame
6.57 kB
metadata
license: mit
base_model: microsoft/deberta-v3-large
tags:
  - generated_from_trainer
datasets:
  - squad_v2
model-index:
  - name: deberta-v3-large-finetuned-squadv2
    results: []

deberta-v3-large-finetuned-squadv2

This model is a fine-tuned version of microsoft/deberta-v3-large on the squad_v2 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5579

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-05
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 64
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-06
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 1000
  • training_steps: 5200

Training results

Training Loss Epoch Step Validation Loss
0.5293 1.57 3200 0.5739
0.5106 1.58 3220 0.5783
0.5338 1.59 3240 0.5718
0.5128 1.6 3260 0.5827
0.5205 1.61 3280 0.6045
0.5114 1.62 3300 0.5880
0.5072 1.63 3320 0.5788
0.5512 1.64 3340 0.5863
0.4723 1.65 3360 0.5898
0.5011 1.66 3380 0.5917
0.5419 1.67 3400 0.6027
0.5425 1.68 3420 0.5699
0.5703 1.69 3440 0.5897
0.4646 1.7 3460 0.5917
0.4652 1.71 3480 0.5745
0.5323 1.72 3500 0.5860
0.5129 1.73 3520 0.5656
0.5441 1.74 3540 0.5642
0.5624 1.75 3560 0.5873
0.4645 1.76 3580 0.5891
0.5577 1.77 3600 0.5816
0.5199 1.78 3620 0.5579
0.5061 1.79 3640 0.5837
0.484 1.79 3660 0.5721
0.5095 1.8 3680 0.5821
0.5342 1.81 3700 0.5602
0.5435 1.82 3720 0.5911
0.5288 1.83 3740 0.5647
0.5476 1.84 3760 0.5733
0.5199 1.85 3780 0.5675
0.5067 1.86 3800 0.5839
0.5418 1.87 3820 0.5757
0.4965 1.88 3840 0.5764
0.5273 1.89 3860 0.5906
0.5808 1.9 3880 0.5762
0.5161 1.91 3900 0.5612
0.4863 1.92 3920 0.5804
0.4827 1.93 3940 0.5841
0.4643 1.94 3960 0.5822
0.5029 1.95 3980 0.6052
0.509 1.96 4000 0.5800
0.5382 1.97 4020 0.5645
0.469 1.98 4040 0.5685
0.5032 1.99 4060 0.5779
0.5171 2.0 4080 0.5686
0.3938 2.01 4100 0.5889
0.4321 2.02 4120 0.6039
0.4185 2.03 4140 0.5996
0.4782 2.04 4160 0.5800
0.424 2.05 4180 0.6374
0.3766 2.06 4200 0.6096
0.415 2.07 4220 0.6221
0.4352 2.08 4240 0.6150
0.4336 2.09 4260 0.6055
0.4289 2.1 4280 0.6138
0.4433 2.11 4300 0.5946
0.4478 2.12 4320 0.6118
0.4787 2.13 4340 0.5969
0.4432 2.14 4360 0.6048
0.4319 2.15 4380 0.5948
0.3939 2.16 4400 0.6116
0.3921 2.17 4420 0.6082
0.4381 2.18 4440 0.6282
0.4461 2.19 4460 0.6084
0.4012 2.2 4480 0.6092
0.3849 2.21 4500 0.6152
0.4178 2.22 4520 0.6004
0.4163 2.23 4540 0.6059
0.4006 2.24 4560 0.6115
0.4225 2.25 4580 0.6130
0.4008 2.26 4600 0.6095
0.4706 2.27 4620 0.6136
0.3902 2.28 4640 0.6103
0.4048 2.29 4660 0.6085
0.4411 2.3 4680 0.6139
0.403 2.31 4700 0.6047
0.4799 2.31 4720 0.6043
0.4316 2.32 4740 0.5960
0.4198 2.33 4760 0.6031
0.4254 2.34 4780 0.6033
0.387 2.35 4800 0.6120
0.3882 2.36 4820 0.6128
0.4307 2.37 4840 0.6150
0.434 2.38 4860 0.6077
0.4225 2.39 4880 0.6071
0.4134 2.4 4900 0.6036
0.3846 2.41 4920 0.6124
0.3943 2.42 4940 0.6291
0.4455 2.43 4960 0.6185
0.4104 2.44 4980 0.6064
0.4158 2.45 5000 0.6095
0.4135 2.46 5020 0.6155
0.3789 2.47 5040 0.6209
0.418 2.48 5060 0.6106
0.3931 2.49 5080 0.6047
0.4289 2.5 5100 0.6055
0.4051 2.51 5120 0.6084
0.4217 2.52 5140 0.6118
0.3843 2.53 5160 0.6139
0.4435 2.54 5180 0.6126
0.4274 2.55 5200 0.6120

Framework versions

  • Transformers 4.35.0.dev0
  • Pytorch 2.1.0+cu121
  • Datasets 2.14.5
  • Tokenizers 0.14.0