LeonardBongard/IMDBSentimentAnalysisLoRaBertUncased

bert Model Card

Model Details

• Model: bert-base-uncased
• Architecture: ['BertForSequenceClassification']
• Training Parameters:
  • Epochs: 3
  • Batch Size: 16
  • Learning Rate: 3.757374065772398e-05
  • Weight Decay: 0.15233987118417686
  • LoRA Parameters:
    • r: 13
    • alpha: 64
    • dropout: 0.1
• Evaluation Metrics:
  • Accuracy: 0.94116
  • F1 Score: 0.9411596941245539
• Dataset: IMDb
• Task: Sentiment Analysis

Intended Use and Limitations

This model is fine-tuned for text classification. Use with caution for out-of-scope tasks.

Training Data

The model was fine-tuned on the specified dataset.