# Advanced Usage

Embedding functions are rarely used alone.

Its used in two way

* In **ETL** and Vectordb to convert all the raw data extracted by the ETL into embeddings to be stored in the Vectordb. It also helps in converting the query to embeddings.
* In **Retrieval** it converts the user query into an embedding to search against the other data in the vectordb index

### Usage

**Imports:**

```python
from genai_stack.etl.langchain import LangchainETL
from genai_stack.stack.stack import Stack
from genai_stack.vectordb.chromadb import ChromaDB
from genai_stack.etl.utils import get_config_from_source_kwargs
from genai_stack.embedding.utils import get_default_embeddings
```

**Configuration:**

```json
config = {
    "name": "HuggingFaceEmbeddings",
    "fields": {
        "model_name": "sentence-transformers/all-mpnet-base-v2",
        "model_kwargs": {"device": "cpu"},
        "encode_kwargs": {"normalize_embeddings": False},
    }
}
```

#### Using with ETL

Once you have defined your configuration as a Python dictionary, you can use it with the `LangchainEmbedding.from_kwargs()` method:

```python
embeddings = LangchainETL.from_kwargs(**config)

etl = LangchainETL.from_config(get_config_from_source_kwargs("pdf", "path/to/pdf"))

# Connect the ETL, Embedding and Vectordb component using Stack
stack = Stack(model=None, embedding=get_default_embeddings(), etl=etl, vectordb=ChromaDB.from_kwargs())

etl.run()
```

**Using with retriever**

```python
# Initialise all your components
etl = LangchainETL.from_kwargs(name="CSVLoader", fields={"file_path": "addresses.csv"})
embedding = LangchainEmbedding.from_kwargs(**config)
chromadb = ChromaDB.from_kwargs()
llm = OpenAIGpt35Model.from_kwargs(parameters={"openai_api_key": "<OPENAI-API-KEY>"})
prompt_engine = PromptEngine.from_kwargs(should_validate=False)
retriever = LangChainRetriever.from_kwargs()
memory = ConversationBufferMemory.from_kwargs()

# Initialise your stack by connecting the components end-to-end
stack = Stack(
    etl=etl,
    embedding=embedding,
    vectordb=chromadb,
    model=llm,
    prompt_engine=prompt_engine,
    retriever=retriever,
    memory=memory
)

# Query to get RAG based results
response = retriever.retrieve("Where does John live?")


```


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://genaistack.aiplanet.com/components/embedding/advanced-usage.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.