Train content - Ongoing

[aashutosh-samagra]

Content sheet - https://docs.google.com/spreadsheets/d/12NPm3r4odguaI_3dqEiyJKSzhIsCZXp30BFFrMss1UY/edit?pli=1#gid=1447499902

• Text Context;
• Image Content - Context tagging to the heading is done manually
• @Amruth-Vamshi to add chunks
• @aashutosh-samagra to verify

[aashutosh-samagra]

Web PDFs - https://github.com/rachitavya/crops_webpages_scrapper/tree/main/output

[ChakshuGautam]

• Hindi PDF => Extract text => Translate =>

[aashutosh-samagra]

To be trained on priority -PM-Kisan, PMFBY, FPO FAQs

[ChakshuGautam]

@Amruth-Vamshi to upload the 1000 new chunks.

[rachitavya]

Hey @RyanWalker277, can you pls upload this data with embeddings to the bot content.

Combined chunks P1.csv

[RyanWalker277]

Hey @RyanWalker277, can you pls upload this data with embeddings to the bot content.

Combined chunks P1.csv

Have pinged @Amruth-Vamshi, will coordinate with him for uploading the chunks and let you know once done

[RyanWalker277]

Hey @rachitavya, summaryEmbedding is missing from the CSV. The CSV should be in the format

content,heading,summary,startPage,endPage,contentEmbedding,headingEmbedding,summaryEmbedding

[Gautam-Rajeev]

Copy paste content embedding to summary embedding