Show HN: Adaptive-classifier – text classification with continuous learning
Hi HN! I've built a Python library that lets you create text classifiers that can continuously learn and adapt to new classes without retraining from scratch.
What makes this different from typical text classifiers:
- Dynamically add new classes at any time without full retraining - Combines neural networks with prototype learning for better few-shot performance - Uses Elastic Weight Consolidation to prevent catastrophic forgetting - Works with any HuggingFace transformer base model - Memory-efficient through prototype-based storage
You can try it out in under a minute:
pip install adaptive-classifier
from adaptive_classifier import AdaptiveClassifier
# Initialize with any HuggingFace model classifier = AdaptiveClassifier("bert-base-uncased")
# Add initial examples texts = ["Great product!", "Terrible experience", "Average performance"] labels = ["positive", "negative", "neutral"] classifier.add_examples(texts, labels)
# Make predictions print(classifier.predict("This is amazing!")) # [('positive', 0.85), ('neutral', 0.10), ('negative', 0.05)]
# Add a completely new class later classifier.add_examples( ["Error 404", "System crashed"], ["technical", "technical"] )
The library came out of my work on building a model router for optillm where approaches and model types keep changing. Traditional classifiers require full retraining when adding new classes, which becomes impractical with large datasets.
Source code: https://github.com/codelion/adaptive-classifier
I'd love to hear your thoughts and feedback! Happy to answer any questions about the implementation details, use cases, or future plans.
Some technical details about how it works:
The core architecture combines a transformer model for embeddings with a prototype memory system and an adaptive neural head.
When adding new classes, it uses Elastic Weight Consolidation (EWC) to preserve performance on existing classes while learning new ones. This prevents the common problem of catastrophic forgetting.
The prototype memory system maintains class prototypes that get updated efficiently as new examples are added, making it memory-efficient even with large datasets.
All state (prototypes, examples, neural weights) can be saved and loaded, making it easy to deploy and update models in production.
The library is built on PyTorch and integrates with the HuggingFace ecosystem. It's tested with Python 3.8+ and requires minimal dependencies.
Let me know if you'd like me to explain any part in more detail!