ollama does not exist in the registry

[qifuxiao]

when I want to use ollama engine instead of default_embeding_engine. I write config.yml:`models:

• type: main
  engine: ollama
  model: llama3.3:latest
  parameters:
  base_url: http://10.180.158.23:11454
• type: embeddings
  engine: ollama
  model: linux6200/bge-reranker-v2-m3:latest
  parameters:
  base_url: http://10.180.158.23:11454`
  I run code, get: File "/home/alexqi/pyvenv/guard/lib/python3.11/site-packages/nemoguardrails/embeddings/basic.py", line 154, in _get_embeddings self._init_model() File "/home/alexqi/pyvenv/guard/lib/python3.11/site-packages/nemoguardrails/embeddings/basic.py", line 137, in _init_model self._model = init_embedding_model( ^^^^^^^^^^^^^^^^^^^^^ File "/home/alexqi/pyvenv/guard/lib/python3.11/site-packages/nemoguardrails/embeddings/providers/__init__.py", line 98, in init_embedding_model provider_class = EmbeddingProviderRegistry().get(embedding_engine) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/alexqi/pyvenv/guard/lib/python3.11/site-packages/nemoguardrails/registry.py", line 63, in get raise KeyError(f"{name} does not exist in the registry") KeyError: 'ollama does not exist in the registry' Parameter max_tokens does not exist for Ollama

Then,I debug the code ,I see EmbeddingProviderRegistry(items=dict_keys(['FastEmbed', 'openai', 'SentenceTransformers', 'nim', 'nvidia_ai_endpoints'])) How can I add Ollama for EmbeddingProviderRegistry.
I would be most grateful if you could help me solve this confusion

Python Version: Python 3.11.11
requirements: nemoguardrails 0.12.0

[qifuxiao]

I solved it.
I added a providers ,Created a OllamaEmbeddingModel,Used it

[Pouyanpi]

Hi @qifuxiao,

Great to hear! I’m glad you were able to implement and register it.

If you’d like to share it with the community, feel free to open a PR, totally up to you.