2 * 1024 input vector

nlp
1

hello guys!

I want to design a binary classifier,
I have a dataset with size 5000 and 3 columns (
‘text_a’ a vector with 1024 numbers(this is the output of Roberta),
‘text_b’ a vector with 1024 numbers,
‘label’ 0/1)

in each row, if the text_a and text_b be similar the label would be 1 otherwise it would be 0.

I want to design a classifier that gives 2 inputs, each input with size 1024 but I can’t.

could you help me how should be the first layer? or the whole model?

2

Not clearly understood the problem statement. Please write something extra or in different language for me to help you out.

1 Like
3

thanks for your reply,

I am a beginner in Paytorch. And I only work with networks with one input.
Here we have two inputs for a classifier and I do not know what the network architecture and input layer should be like?

4

Okay. So lets do one thing. Since I am not able to grasp your task, so you can share your dataset with me via email. I will give it time and update you sometime tomorrow.

email: tejanmehndiratta15@gmail.com

5

ok, thank you very much

6

And lets chat on private msg regarding this.

1 Like