0

I recently got this error while trying to put my data into tensorflow format:

Error in py_call_impl(callable, dots$args, dots$keywords) : 
ValueError: column_name: age vocabulary dtype must be string or integer. dtype: <dtype: 'float64'>.

Is there a command that will allow me to change the format of my data frame's column name from a float to a string?

Thank you!

aynber
  • 22,380
  • 8
  • 50
  • 63
Emma
  • 31
  • 5

1 Answers1

1

From the NLP (Natural Language Processing) package, you can use as.String(x) to coerce the column names to a string and then store them again as the column names of the dataframe. This can be done iteratively:

library(NLP)

float_colnames <- colnames(df)

string_colnames <- rep("", length(float_colnames))
for (i in 1:length(float_colnames)) {
  string_colnames[i] <- as.String(test[i])
}

colnames(df) <- string_colnames

Hope this helps!

Kel Varnsen
  • 314
  • 2
  • 8