# How can I evaluate my sequence prediction model?

I want to evaluate the performance of my prediction model , which is an VED (Variational Encoder Decoder) used for sequences prediction (it predicts the next sequence knowing the actual) I want to evaluate the model by calculating the similarity between the input sequence (x_train) and the predicted one.

I found that there is a score called BELU can be used for this, but I'm not sure. If someone has an idea, please help me .

• Sequences could be seen as vectors. See here for measures used for calculating the similarity of vectors. Also, BLEU could be used as mentioned here. – Shubham Panchal May 18 at 5:40
• thank you, but the links are for the same page, I think the second one is not the right link – Kikio May 18 at 13:40