Harshkmr's picture
Model save
0f5ffb9 verified
metadata
license: mit
base_model: microsoft/deberta-v3-base
tags:
  - generated_from_trainer
model-index:
  - name: deberta-v3-base_finetuned_bluegennx_run2.21_3e
    results: []

deberta-v3-base_finetuned_bluegennx_run2.21_3e

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

  • Loss: 0.0168
  • Overall Precision: 0.9773
  • Overall Recall: 0.9878
  • Overall F1: 0.9825
  • Overall Accuracy: 0.9959
  • Aadhar Card F1: 0.9866
  • Age F1: 0.9707
  • City F1: 0.9868
  • Country F1: 0.9865
  • Creditcardcvv F1: 0.9888
  • Creditcardnumber F1: 0.9587
  • Date F1: 0.9643
  • Dateofbirth F1: 0.9165
  • Email F1: 0.9894
  • Expirydate F1: 0.9921
  • Organization F1: 0.9917
  • Pan Card F1: 0.9856
  • Person F1: 0.9883
  • Phonenumber F1: 0.9868
  • Pincode F1: 0.9936
  • Secondaryaddress F1: 0.9861
  • State F1: 0.9901
  • Time F1: 0.9821
  • Url F1: 0.9949

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: 5e-05
  • train_batch_size: 4
  • eval_batch_size: 4
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine_with_restarts
  • lr_scheduler_warmup_ratio: 0.2
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss Overall Precision Overall Recall Overall F1 Overall Accuracy Aadhar Card F1 Age F1 City F1 Country F1 Creditcardcvv F1 Creditcardnumber F1 Date F1 Dateofbirth F1 Email F1 Expirydate F1 Organization F1 Pan Card F1 Person F1 Phonenumber F1 Pincode F1 Secondaryaddress F1 State F1 Time F1 Url F1
0.0308 1.0 16005 0.0359 0.9500 0.9755 0.9626 0.9920 0.9350 0.9315 0.9658 0.9706 0.9631 0.9282 0.9251 0.8430 0.9719 0.9842 0.9866 0.9696 0.9772 0.9503 0.9835 0.9680 0.9815 0.9739 0.9845
0.0202 2.0 32010 0.0195 0.9737 0.9836 0.9786 0.9950 0.9756 0.9586 0.9790 0.9826 0.9866 0.9497 0.9593 0.9060 0.9893 0.9872 0.9902 0.9716 0.9875 0.9830 0.9939 0.9831 0.9871 0.9799 0.9927
0.0107 3.0 48015 0.0168 0.9773 0.9878 0.9825 0.9959 0.9866 0.9707 0.9868 0.9865 0.9888 0.9587 0.9643 0.9165 0.9894 0.9921 0.9917 0.9856 0.9883 0.9868 0.9936 0.9861 0.9901 0.9821 0.9949

Framework versions

  • Transformers 4.41.2
  • Pytorch 2.1.2
  • Datasets 2.19.2
  • Tokenizers 0.19.1