The RAG Experiment Accelerator is a versatile tool designed to expedite and facilitate the process of conducting experiments and evaluations using Azure Cognitive Search and RAG pattern.
Currently, RAGE does not provide Page number and file names in its final output to the user. The final output should contain the page number and file name from which the chunk are been taken from. This is applicable for docx and pdf files. This should work across all chunking methods and should be configurable.
Currently, RAGE does not provide Page number and file names in its final output to the user. The final output should contain the page number and file name from which the chunk are been taken from. This is applicable for docx and pdf files. This should work across all chunking methods and should be configurable.