mrm8488 / ama Goto Github PK

Hi, I am trying to install and modify your package manuelromeo but the instalations falis

Hello there. I have been following your work and at the onset I would like to thank you for the great work you are doing. I have one question regarding a pet project of mine. I would like to create a Turkish Bio BERT for question answering purposes. I have some translated medical corpus and some annotated QA pairs on the same. I have two approaches in my mind.

1. Use the BioBERT( which has been fine-tuned on SQUAD) and further perform fine-tuning with the annotated pairs. I was thinking of using Haystack for this. However, I do not think this is a good approach.
2. My second approach is to take a multilingual BERT - perform MLM on the corpus - finetune it on the XQUAD and then test with the annotated QA pairs.

I would really appreciate your opinion on this or any alternative that you think might be useful.
Thank you once again and looking forward to your comment as well as your work.

@atulmy

Can you provide an example using $lookup in Mongoose for following case:

User model

User {
  _id,
  name.
  ...
}

Participant Model

Participant {
  userId,
  ...
}

I'm trying:

Participant
  .find()
  .populate({ 
    path: 'userId', 
    options: { sort: { name : 1 } } 
  })

Hey all,

we are going to develop a Dating App, with peer to peer function, profiles, little news feed and so on. Maybe we need some times a Payment System. I am going to realise this with express and vue.js but i am not sure which DB we should run. I would pref MongoDB because i have some experience with it, but my friend mean Sequelize with a MySQL Database should be better.

Can you help out?

Thanks!