CODEWITHSUNDEEP
pythonrtensorflowrstudio
Aman Mathur
Aman Mathur

Reputation: 719

Error in tensorflow for R example

I am new to Tensorflow. I heard really great things about the tensorflow interface to R. I am however finding it difficult to run example listed here: https://tensorflow.rstudio.com/tensorflow/articles/examples/linear_regression_multiple.html

The specific problem is in the following lines

while (TRUE) {
+     session$run(optimizer, feed_dict = feed_dict)
+     current_cost <- session$run(cost, feed_dict = feed_dict)
+     if (last_cost - current_cost < epsilon) break
+     last_cost <- current_cost
+ }

This throws the following error : 

Error in py_call_impl(callable, dots$args, dots$keywords) : 
  TypeError: Fetch argument <tensorflow.python.training.gradient_descent.GradientDescentOptimizer object at 0x126eaaf50> of <tensorflow.python.training.gradient_descent.GradientDescentOptimizer object at 0x126eaaf50> has invalid type <class 'tensorflow.python.training.gradient_descent.GradientDescentOptimizer'>, must be a string or Tensor. (Can not convert a GradientDescentOptimizer into a Tensor or Operation.)

The only thing I have changed from the above example is :

tf$global_variables_initializer

to

tf$initialize_all_variables

Anyone faced similar error?

Upvotes: 0

Views: 236

Answers (1)

Carlos Santillan
Carlos Santillan

Reputation: 1087

The initialize_all_variables API has been deprecated in tensorflow (and was removed on 2017-03-02) , the correct API to use is global_variables_initializer

https://www.tensorflow.org/api_docs/python/tf/initialize_all_variables

Upvotes: 2

Related Questions