search

lccasagrande / Deep-Knowledge-Tracing

An implementation of the Deep Knowledge Tracing (DKT) using Tensorflow 2.0
MIT License
95 stars 38 forks source link

I get some question, why? #7

Open sunhao97 opened 4 years ago

sunhao97 commented 4 years ago

evaluate() got an unexpected keyword argument 'x' , l don't know how to solve it.

lccasagrande commented 3 years ago

The evaluate() from keras is not the same as the one used in this model. The evaluate() function implemented in this model, does not accept the argument "x". Check this.

You can modify this class method or you can follow the approach I've used (TF Dataset). Check this.

sunhao97 commented 3 years ago

ok, I have solved the problem in your way. Thank you for your help.

edloginova commented 3 years ago

Hello! Could you please elaborate what changes are needed to get rid of this error? I tried to run the .ipynb and run_dkt.py from the examples/, but got the same message about evaluate() method. Since the data is already passed as TF Dataset and the evaluate method of the class is supposed to be used, I am not sure what is the problem. Thank you!

sunhao97 commented 3 years ago

l change the evaluate() method's name in this model , you can change this name which you like ,and you can also change the version which can support this model.

sunhao97 commented 3 years ago

The main question is that the evaluate() method's name in this model is as same as which in the keras, because you did not set up the same version as the author.

thewhiteflower110 commented 3 years ago

I changed the version to tensorflow 2.0. This solved the issue.