avinasht's picture
Acc0.8520599250936329, F10.8500507249833522 , Augmented with bert-base-uncased.csv, finetuned on google/electra-base-discriminator
c21d11a verified
metadata
license: apache-2.0
base_model: google/electra-base-discriminator
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - precision
  - recall
model-index:
  - name: electra-base-discriminator_bert-base-uncased
    results: []

electra-base-discriminator_bert-base-uncased

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

  • Loss: 0.5398
  • Accuracy: 0.8705
  • F1: 0.8691
  • Precision: 0.8729
  • Recall: 0.8705

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: 0.0001
  • train_batch_size: 64
  • eval_batch_size: 64
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 1000
  • num_epochs: 25

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Precision Recall
0.9217 1.0 91 0.8648 0.6459 0.5998 0.6333 0.6459
0.5726 2.0 182 0.5369 0.8066 0.8064 0.8238 0.8066
0.3522 3.0 273 0.4095 0.8440 0.8415 0.8477 0.8440
0.2589 4.0 364 0.5367 0.8097 0.8069 0.8258 0.8097
0.2718 5.0 455 0.4216 0.8612 0.8621 0.8670 0.8612
0.164 6.0 546 0.5346 0.8612 0.8602 0.8616 0.8612
0.1075 7.0 637 0.5398 0.8705 0.8691 0.8729 0.8705
0.1461 8.0 728 0.6163 0.8362 0.8368 0.8442 0.8362
0.132 9.0 819 0.4933 0.8674 0.8675 0.8701 0.8674
0.1359 10.0 910 0.7141 0.8424 0.8416 0.8489 0.8424
0.0971 11.0 1001 0.5662 0.8596 0.8578 0.8623 0.8596
0.1148 12.0 1092 0.5685 0.8612 0.8609 0.8610 0.8612

Framework versions

  • Transformers 4.37.0
  • Pytorch 2.1.2
  • Datasets 2.1.0
  • Tokenizers 0.15.1