components llm_rag_image_embed_index - Azure/azureml-assets GitHub Wiki
Embeds input images and stores it in Azure Cognitive Search index with metadata using Florence embedding resource. MLIndex is stored to output_path.
Version: 0.0.25
Preview
View in Studio: https://ml.azure.com/registries/azureml/components/llm_rag_image_embed_index/version/0.0.25
| Name | Description | Type | Default | Optional | Enum |
|---|---|---|---|---|---|
| input_path | The input uri_folder containing a set of images and json metadata. | uri_folder | False | ||
| embedding_connection_id | The input connection id to the Azure cognitive vision resource for florence embedding. | string | False | ||
| acs_connection_id | The input connection id to the Azure Cognitive Search resource. | string | False | ||
| search_index_name | The index name for storing the images embeddings and metadata. | string |
| Name | Description | Type |
|---|---|---|
| output_path | Where to store the output MLIndex. | uri_folder |
azureml:llm-rag-embeddings@latest