search

sophiaalthammer / parm

This repository contains the code for the paper 'PARM: Paragraph Aggregation Retrieval Model for Dense Document-to-Document Retrieval' published at ECIR'22.
MIT License
40 stars 4 forks source link

How to inference on our own data? #1

Open avijit9 opened 2 years ago

avijit9 commented 2 years ago

Hi,

Thanks for open-sourcing this great repository!

I am trying to do inference on my own dataset. However, I don't have labeled answers for the queries. So, I can't really evaluate the method and I would like to see the results qualitatively. It would be great if you can provide me with some pointers on this.

Thanks in advance!

sophiaalthammer commented 2 years ago

Hi Avijit,

thanks for your interest in my repository! What kind of data are you using? I think it depends on the task, which dense retrieval model I would use for creating the dense embeddings, so if you let me know more about the task.

Other than that you can get the output ids of the retrieval model and implement that you look at them, for qualitative analysis I dont have a tool unfortunately!

Best, Sophia

Am 12.02.2022 um 05:42 schrieb Avijit Dasgupta @.***>:

 Hi,

Thanks for open-sourcing this great repository!

I am trying to do inference on my own dataset. However, I don't have labeled answers for the queries. So, I can't really evaluate the method and I would like to see the results qualitatively. It would be great if you can provide me with some pointers on this.

Thanks in advance!

— Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android. You are receiving this because you are subscribed to this thread.

avijit9 commented 2 years ago

Hi Sophia,

Thanks for your prompt reply.

In my task, I have a paragraph and I would like to search the most similar paragraph from a set of other paragraphs. It would be great if you can point out the exact function or script I should be looking into.

Thanks again.