SentimentT2 / README.md
LogischeIP's picture
End of training
1c1470f verified
|
raw
history blame
2.34 kB
metadata
license: apache-2.0
base_model: distilbert-base-uncased
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
model-index:
  - name: SentimentT2
    results: []

SentimentT2

This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.8793
  • Accuracy: 0.7143
  • F1: 0.7099
  • Auc Roc: 0.8948
  • Log Loss: 1.8793

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: 2e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Auc Roc Log Loss
No log 1.0 8 1.7750 0.7429 0.7448 0.8959 1.7750
No log 2.0 16 1.9519 0.6857 0.6896 0.8992 1.9519
No log 3.0 24 1.8079 0.7286 0.7293 0.8990 1.8079
No log 4.0 32 1.8638 0.6714 0.6745 0.8975 1.8638
No log 5.0 40 1.9660 0.6714 0.6656 0.9049 1.9660
No log 6.0 48 1.9401 0.7143 0.7042 0.8977 1.9401
No log 7.0 56 1.9271 0.7 0.6934 0.8911 1.9271
No log 8.0 64 1.9025 0.7 0.6930 0.8896 1.9025
No log 9.0 72 1.8840 0.7143 0.7099 0.8942 1.8840
No log 10.0 80 1.8793 0.7143 0.7099 0.8948 1.8793

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.1.0+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.0