0

I have a model for binary classification that includes 2 linear layers with RELU activation function and Sigmoid in the last layer. The input features are FastText word embedding, frequency, and statistical signals.

This model has a 93% f1-score and I want to add an explanation to this model but I don't know how can I start.

My question is which models or papers good for these complex input features?

I appreciate any advice to achieve this goal.

hanugm
  • 3,571
  • 3
  • 18
  • 50
Abolfazl Mohajeri
  • 109
  • 3
  • Hello. What do you mean by "i want to add explanation to this model"? You need to clarify what exactly do you expect the model to do. Moreover, I would recommend that you take a look at [this post](https://ai.stackexchange.com/q/12870/2444) before proceeding. Once you have done that, please, edit the post to ask a more specific question. – nbro Nov 04 '21 at 12:36
  • @nbro tnx for your response, yes i read about this articles & areas but in my model, input features are in high dimensions (word embedding) and these articles not good in this situation my question is: which Explainable AI techniques good for such models that i said. – Abolfazl Mohajeri Nov 04 '21 at 15:21

0 Answers0