mmillet's picture
update model card README.md
7f7f008
|
raw
history blame
2.49 kB
metadata
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - precision
  - recall
model-index:
  - name: >-
      distilrubert-tiny-cased-conversational-v1_single_finetuned_on_cedr_augmented
    results: []

distilrubert-tiny-cased-conversational-v1_single_finetuned_on_cedr_augmented

This model is a fine-tuned version of DeepPavlov/distilrubert-tiny-cased-conversational-v1 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5908
  • Accuracy: 0.8653
  • F1: 0.8656
  • Precision: 0.8665
  • Recall: 0.8653

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-06
  • lr_scheduler_type: linear
  • num_epochs: 20

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Precision Recall
0.9172 1.0 69 0.5124 0.8246 0.8220 0.8271 0.8246
0.4709 2.0 138 0.4279 0.8528 0.8505 0.8588 0.8528
0.3194 3.0 207 0.3770 0.8737 0.8727 0.8740 0.8737
0.2459 4.0 276 0.3951 0.8685 0.8682 0.8692 0.8685
0.1824 5.0 345 0.4005 0.8831 0.8834 0.8841 0.8831
0.1515 6.0 414 0.4356 0.8800 0.8797 0.8801 0.8800
0.1274 7.0 483 0.4642 0.8727 0.8726 0.8731 0.8727
0.0833 8.0 552 0.5226 0.8633 0.8627 0.8631 0.8633
0.073 9.0 621 0.5327 0.8695 0.8686 0.8692 0.8695
0.0575 10.0 690 0.5908 0.8653 0.8656 0.8665 0.8653

Framework versions

  • Transformers 4.19.3
  • Pytorch 1.11.0+cu113
  • Datasets 2.2.2
  • Tokenizers 0.12.1