Adding projects created for content about hybrid search for an ecommerce #343
Conversation
|
Found 1 changed notebook. Review the changes at https://app.gitnotebooks.com/elastic/elasticsearch-labs/pull/343 |
supporting-blog-content/hybrid-search-for-an-e-commerce-product-catalogue/README.md
Outdated
Show resolved
Hide resolved
...g-blog-content/hybrid-search-for-an-e-commerce-product-catalogue/app-product-store/README.md
Show resolved
Hide resolved
| const defaultImage = "https://via.placeholder.com/100"; | ||
|
|
||
| const handleClick = () => { | ||
| //alert(`Produto: ${product.name} - Termo de pesquisa: ${searchTerm}`); |
There was a problem hiding this comment.
| //alert(`Produto: ${product.name} - Termo de pesquisa: ${searchTerm}`); |
Can we remove the commented out alert since it's not used.
| query: searchTerm, | ||
| selectedCategories: selectedFacets.categories, | ||
| selectedProductTypes: selectedFacets.productTypes, | ||
| selectedbrands: selectedFacets.brands, |
There was a problem hiding this comment.
| selectedbrands: selectedFacets.brands, | |
| selectedBrands: selectedFacets.brands, |
| query: searchTerm, | ||
| selectedCategories: selectedFacets.categories, | ||
| selectedProductTypes: selectedFacets.productTypes, | ||
| selectedbrands: selectedFacets.brands, |
There was a problem hiding this comment.
| selectedbrands: selectedFacets.brands, | |
| selectedBrands: selectedFacets.brands, |
|
|
||
| ## Referências | ||
|
|
||
| - [Documentação do Docker](https://docs.docker.com/) |
There was a problem hiding this comment.
| - [Documentação do Docker](https://docs.docker.com/) | |
| - [Docker Documentation](https://docs.docker.com/) |
| ## Referências | ||
|
|
||
| - [Documentação do Docker](https://docs.docker.com/) | ||
| - [Documentação do Docker Compose](https://docs.docker.com/compose/) |
There was a problem hiding this comment.
| - [Documentação do Docker Compose](https://docs.docker.com/compose/) | |
| - [Docker Compose Documentation](https://docs.docker.com/compose/) |
|
|
||
|
|
||
| if __name__ == '__main__': | ||
| # print(get_client_es().info()) |
There was a problem hiding this comment.
| # print(get_client_es().info()) |
|
|
||
| vector = get_text_vector([term])[0] | ||
|
|
||
| # Query híbrida com RRF (Reciprocal Rank Fusion) |
There was a problem hiding this comment.
| # Query híbrida com RRF (Reciprocal Rank Fusion) | |
| # Hybrid query with RRF (Reciprocal Rank Fusion) |
|
|
||
|
|
||
| def build_hybrid_query(term=None, categories=None, product_types=None, brands=None, hybrid=False): | ||
| # Query padrão |
There was a problem hiding this comment.
| # Query padrão | |
| # Standard query |
There was a problem hiding this comment.
The changes look good! I've spotted a few Portuguese comments that have been missed. I've added comments to the relevant files where I've seen them.
@andreluiz1987 do you mind checking the project once more to make sure those are translated?
| } | ||
| }; | ||
|
|
||
| // Função para buscar os facets |
There was a problem hiding this comment.
Can you translate this comment please?
| @@ -0,0 +1,34 @@ | |||
| # Arquivos de configuração local | |||
There was a problem hiding this comment.
Can you translate the comments to English for consistency please?
There was a problem hiding this comment.
I translated the other files from Portuguese to English.
|
@andreluiz1987 I think your changes have some issues. Can you check that you've run the pre-commit steps and raise a PR for those? python -m venv .venv |
|
This PR has mulitple issues and at least one issue is common to other PR as well. In addition to the wrong path for https://github.com/elastic/elasticsearch-labs/blob/main/supporting-blog-content/hybrid-search-for-an-e-commerce-product-catalogue/product-store-search/ingestion/ingestion.py#L49, I have to upgrade sentence-transformers upgrade to 3.3.0, otherwise it won't work with hugging face. |
| # Hybrid query with RRF (Reciprocal Rank Fusion) | ||
| query = { | ||
| "retriever": { | ||
| "rrf": { |
There was a problem hiding this comment.
@andreluiz1987 thanks for your post and contribution! it was great!
What are the consideration for chosing rrf over a simple should?
Do you get the same score range with rrf? (knn return _score between 0-1.0 and organic_query doesn't have an upper _score bound)
I asked in the forum but I wondered if you have the same issue here.
This PR contains the frontend and api projects built for the article How to use Hybrid search for an e-commerce product catalogue