✅ Training Complete!

Model "{{ model_name }}" trained successfully

Your custom embedding model is ready to use. It has learned domain-specific terminology and synonyms from your training corpus.

📊 Model Statistics

Embedding Dimension

Training Chunks

Total Words

Unique Terms

📁 Model Location

Model directory: {{ model_dir }}

Files created:

model.onnx - ONNX model file (~8 MB)
config.json - Model metadata

🚀 How to Use Your Model

Python API

# Load your custom model
from justembed.embedder import CustomEmbedder
import numpy as np

embedder = CustomEmbedder("{{ model_name }}")

# Embed documents
docs = ["your document text here"]
doc_embeddings = embedder.embed(docs)

# Embed query
query = "your search query"
query_embedding = embedder.embed_query(query)

# Compute similarity
similarities = [
    np.dot(query_embedding, doc_emb)
    for doc_emb in doc_embeddings
]
        

⚙️ Model Configuration

{
  "model_name": "{{ config.model_name }}",
  "created_at": "{{ config.created_at }}",
  "embedding_dim": {{ config.embedding_dim }},
  "max_features": {{ config.max_features }},
  "training_params": {
    "hidden_layers": {{ config.training_params.hidden_layers }},
    "activation": "{{ config.training_params.activation }}",
    "solver": "{{ config.training_params.solver }}"
  },
  "version": "{{ config.version }}"
}
        

🎯 Next Steps

Test your model: Use the Python API to test synonym learning
Compare with E5: See how your custom model performs vs the default E5 model
Train more models: Create models for different domains
Integration: KB integration coming soon!

Train Another Model Back to Home