Manirathinam21/DistilBert_SMSSpam_classifier
This model is a fine-tuned version of distilbert-base-uncased on an SMSSpam Detection dataset. It achieves the following results on the evaluation set:
- Train Loss: 0.0114
- Train Accuracy: 0.9962
- Epoch: 2
Target Labels
label: a classification label, with possible values including
- Ham : 0
- Spam : 1
Model description
Tokenizer used is DistilBertTokenizerFast with return_tensors='tf' parameter in tokenizer because building model in a tensorflow framework
Model: TFDistilBertForSequenceClassification
Optimizer: Adam with learning rate=5e-5
Loss: SparseCategoricalCrossentropy
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
After Tokenized, Encoded datasets are converted to Dataset Objects by using tf.data.Dataset.from_tensor_slices((dict(train_encoding), train_y))
This step is done to inject a dataset into TFModel in a specific TF format
Training hyperparameters
The following hyperparameters were used during training:
- optimizer: {'name': 'Adam', 'learning_rate': 5e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
- training_precision: float32
Training results
Train Loss | Train Accuracy | Epoch |
---|---|---|
0.0754 | 0.9803 | 0 |
0.0252 | 0.9935 | 1 |
0.0114 | 0.9962 | 2 |
Framework versions
- Transformers 4.21.1
- TensorFlow 2.8.2
- Tokenizers 0.12.1
- Downloads last month
- 10