bert-keyword-extractor
| Property | Value |
|---|---|
| License | Apache 2.0 |
| Framework | PyTorch 1.11.0 |
| Base Model | BERT-base-cased |
| Best F1 Score | 87.17% |
What is bert-keyword-extractor?
The bert-keyword-extractor is a sophisticated natural language processing model built on the BERT architecture, specifically designed for extracting keywords from English text. This model represents a fine-tuned version of bert-base-cased, achieving impressive performance metrics with 85.65% precision and 88.74% recall.
Implementation Details
The model was trained using a carefully optimized process with the following specifications: 8 epochs of training, utilizing the Adam optimizer with a learning rate of 2e-05, and implementing native AMP mixed precision training. The training process involved batch sizes of 16 for both training and evaluation phases.
- Training utilized linear learning rate scheduling
- Achieved final validation loss of 0.1341
- Demonstrates excellent accuracy at 97.38%
- Implements token classification for keyword extraction
Core Capabilities
- Precise keyword extraction from English text
- Token-level classification for accurate entity identification
- Optimized for both precision and recall balance
- Suitable for production deployments with inference endpoints
Frequently Asked Questions
Q: What makes this model unique?
This model stands out for its high accuracy (97.38%) and balanced performance metrics, making it particularly reliable for keyword extraction tasks. The implementation of native AMP mixed precision training also ensures efficient processing.
Q: What are the recommended use cases?
The model is ideal for applications requiring automated keyword extraction from English text, such as content categorization, document indexing, and automated tagging systems. It's particularly suitable for production environments due to its inference endpoint support.
