metadata

license: cc-by-nc-4.0
tags:
  - generated_from_keras_callback
model-index:
  - name: snar7/ooo_phrase
    results: []
language:
  - en
pipeline_tag: question-answering
widget:
  - text: What is the out office duration ?
    context: >-
      Good morning, everyone! I'll be on vacation starting today until Friday,
      so please reach out to my colleagues for assistance.
    example_title: Question Answering

snar7/ooo_phrase

This model is a fine-tuned version of bert-large-uncased-whole-word-masking-finetuned-squad on a private dataset of out-of-office emails tagged with the exact phrase which contains the out-of-office context. It achieves the following results on the evaluation set:

Eval Loss (during training): 0.2761, Epochs : 3
Jaccard Score on a test set of tagged out-of-office phrases: ~ 94%

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:

optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 1140, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
training_precision: mixed_float16

Training results

Train Loss	Epoch
0.5315	1
0.3629	2
0.2761	3

Framework versions

Transformers 4.29.1
TensorFlow 2.11.0
Datasets 2.12.0
Tokenizers 0.13.2