Rename the "Intro" notebooks to call out specific functionality it supports (PDF to Embedings)

[Bytes-Explorer]

No description provided.

[shahrokhDaijavad]

@sujee Please suggest a couple of names other than intro for this, and after we agree on it, you can submit a PR for the name change. I know you use this example in workshops as an "introductory" notebook, and @Bytes-Explorer 's suggestion of using PDF2Embeddings (its functionality) is a little too "rigid" as a name for an example, so something along the lines of "Run_your_first-pipeline_pdf2embeddings" (seems too long, doesn't it?) is more appropriate.

[sujee]

@Bytes-Explorer @shahrokhDaijavad

how is something along the lines of pdf processing part 1

Totally open to suggestions :)

I plan to add other examples along the lines of

• PDF processing with OCR
• PDF processing with tables
• PDF processing removing PII
• etc

[shahrokhDaijavad]

@sujee I am ok with "PDF processing Part 1", especially if you are planning to add subsequent examples with OCR, Tables, ... and using the PII transform. Of course, the example does a lot more by showing the effectiveness of exact and fuzzy dedup along the way, but we cannot spell out everything in the name.
@Bytes-Explorer what do you think?

[Bytes-Explorer]

It will be nice to understand the functionality from the name. How about PDF processing for RAG? I would also suggest that there should be a readme at the top folder that tells a user what can they learn from every example

[sujee]

Very good! how about something like..

• pdf_processing_1_for_RAG
• pdf_processing_2_handling_PII
• pdf_processing_3_handling_duplicates
• etc

[shahrokhDaijavad]

Thanks, @Bytes-Explorer and @sujee. Let's go with pdf_processing_1_for_RAG
@sujee Please submit a PR with this new name (and the README explanation of what we learn from this example). We will then use the same PR to incrementally update to release 0.2.2.dev2 AFTER Michele solves the problem with the new Docling.

[shahrokhDaijavad]

@sujee Please see my last comment in #763 and based on that submit a PR for renaming this example and potentially taking care of pip installing humanfriendly and the parameter for Ray settings on Colab. Thanks.

[shahrokhDaijavad]

@sujee I just remembered that this example will also change its flow from "chunking" documents and then "deduplicating" chunks to "deduplicating" documents first and "chunking" next, so the PR should be submitted after that change.

[sujee]

@sujee I just remembered that this example will also change its flow from "chunking" documents and then "deduplicating" chunks to "deduplicating" documents first and "chunking" next, so the PR should be submitted after that change.

yes. a few changes are going to go into this example. I need to verify a couple of issues I raised to get this functionality (#605)

[shahrokhDaijavad]

Right, @sujee The blockers you had in #605, #756 and #767 should all be resolved now. Please test.