search

oswaldoludwig / Seq2seq-Chatbot-for-Keras

This repository contains a new generative model of chatbot based on seq2seq modeling.
Apache License 2.0
332 stars 98 forks source link

TypeError: Value passed to parameter 'shape' has DataType float32 not in list of allowed values: int32, int64 #17

Closed gauravgola96 closed 6 years ago

gauravgola96 commented 6 years ago

I am using python 3. error in this line --> 155 out = Dense(dictionary_size/2, activation="relu", name='relu-activation')(merge_layer)

oswaldoludwig commented 6 years ago

Hi, this project uses Python 2.7.6, Ubuntu 14.04.3 LTS, Theano 0.9.0, and Keras 2.0.4. Your problem seems to be related to the version of Keras or the backend.

gauravgola96 commented 6 years ago

int(dictionary_size/2) this worked for me.

oswaldoludwig commented 6 years ago

This makes sense, especially if the adopted dictionary size is not an even number, e.g. 10001, 7003, ...

gauravgola96 commented 6 years ago

can you provide me the data for question answer .

oswaldoludwig commented 6 years ago

If you mean the training data that I used, it is here: https://www.dropbox.com/sh/o0rze9dulwmon8b/AAA6g6QoKM8hBEHGst6W4JGDa?dl=0

gauravgola96 commented 6 years ago

this what is happening ... any solution for this

Starting the model...

CHAT:

computer: hi ! please type your name.

user: hh computer: hi , hh ! My name is john.

user: benjamin computer: yeah , but my favorite one is hard to find . (with probability of 0.000973) user: will benjamin computer: yeah , but my favorite one is hard to find . (with probability of 0.000973) user: exit computer: yeah , but my favorite one is hard to find . (with probability of 0.000973)

oswaldoludwig commented 6 years ago

It seems like you keep entering your name when you should start a conversation.

gauravgola96 commented 6 years ago

Starting the model...

I guess No!! 

CHAT:

computer: hi ! please type your name.

user: how are you? computer: hi , how are you? ! My name is john