agro_all-MiniLM-L6-v2_concat_gcn_h128_o64_triplet_e256_knowledge
This is a knowledge-enhanced sentence transformer model created with on2vec.
Model Details
- Base Model: sentence-transformers/all-MiniLM-L6-v2
- Architecture: Knowledge-Enhanced Transformer (experimental)
- Knowledge Dim: 256
- Max Concepts: 3
- Created with: on2vec knowledge-enhanced architecture
Usage
⚠️ Note: This is an experimental knowledge-enhanced model that requires special handling.
# This model cannot be loaded with standard SentenceTransformer.load()
# Contact the model creator for usage instructions
Architecture
This model uses a fundamentally different approach than standard fusion models:
- Token embeddings are enhanced with ontology knowledge during forward pass
- End-to-end training in unified representation space
- No separate lookup/fusion step
Generated by on2vec knowledge-enhanced transformer.
from sentence_transformers import SentenceTransformer model = SentenceTransformer("david4096/agro-all-MiniLM-L6-v2_concat_gcn_h128_o64_triplet_e256_knowledge-3") sentences = [ "That is a happy person", "That is a happy dog", "That is a very happy person", "Today is a sunny day" ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4]