Skip to content
This repository has been archived by the owner on Jun 10, 2021. It is now read-only.

translation result does not make sense ??? #440

Open
sondos90 opened this issue Nov 23, 2017 · 2 comments
Open

translation result does not make sense ??? #440

sondos90 opened this issue Nov 23, 2017 · 2 comments

Comments

@sondos90
Copy link

sondos90 commented Nov 23, 2017

i tried this model on its data , the result of translation was not accurate , why? "the prediction score were negative - what does that mean??-"

and what if i tried this model on a data of a complex language type like Arabic for example

which i had tried it the result were catastrophic ... keeps give me one word as a translation for all sentences
what is the explanation of this case and how can i handle this problem and make it work on my data ....???

@guillaumekln
Copy link
Collaborator

The prediction score is the negative log likelihood of the sentence.

On how much data did you train the model and how did you prepare them?

@Anishtiwary
Copy link

I trained model its working super in training data but for test data it is giving same output for everything what to do and i also train till 100000 steps but no improvements

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Development

No branches or pull requests

3 participants