How to resolve this specific kind of rbind(deparse.level, ...): replacement has length zero error?

Question

Problem

I am doing a rbind and I am experiencing the following error as shown below:

Error in rbind(deparse.level, ...) : replacement has length zero

How may I resolve this error?

Details

The code is inside a loop. I am going through a dataframe and if a certain row has a null value for steps, I replace it with something and move on. That was working great until I got to row 289 which basically as far as I know looks exactly just like those before it (except the interval is 0 instead of 2355, but when I changed it to 2356 just to check if that was it, still get the same error, so this is some sort of formatting thing).

The code is as follows:

rbind(df, row)

I have a data frame, lets call it df for short that looks like something like this.

    steps         interval      dates
283 2.6037736     2330          2012-10-01
284 4.6981132     2335          2012-10-01
285 3.3018868     2340          2012-10-01
286 0.6415094     2345          2012-10-01
287 0.2264151     2350          2012-10-01
288 1.0754717     2355          2012-10-01

The row is as follows:

        steps    interval      dates
289     0        0             2012-10-02

Here are some details about df and row as follows:

data.frame':    288 obs. of  3 variables:
 $ steps   :'data.frame':   288 obs. of  1 variable:
  ..$ steps: num  1.717 0.3396 0.1321 0.1509 0.0755 ...
 $ interval: int  0 5 10 15 20 25 30 35 40 45 ...
 $ dates   : Date, format: "2012-10-01" "2012-10-01" "2012-10-01" "2012-10-01" ...

and

'data.frame':   1 obs. of  3 variables:
 $ steps   : int 0
 $ interval: int 0
 $ dates   : Date, format: "2012-10-02"

Update

I think the problem is that my steps is a dataframe within a dataframe. It should be just a column of numeric values. Here is some code that specifies the intended definition for df. Please see:

df <- data.frame(steps = numeric, date = date, interval = integer)
> str(df)
'data.frame':   0 obs. of  3 variables:
 $ steps   :function (...)  
 $ date    :function (...)  
 $ interval:function (...)  

The following cannot be easily duplicated, I apologize in advanced, but I can describe what the code is doing. I have a dataset that has NA in steps and I use row$steps <- subset(avgStepsByInterval, interval == row$interval, steps) to replace that NA with a number. For example, the idea is that:

     steps    interval      dates
289     NA        0             2012-10-02

Becomes

     steps    interval      dates
289     0        0             2012-10-02

The loop I am using is as follows:

for(i in 1:size)
{
  index<-i
  row <- data[i,]
  isNa <- is.na(row$steps)
  if(isNa)
  {
    row$steps <- subset(avgStepsByInterval, interval == row$interval, steps)
  }
  df<-rbind(df, row)
}

Don't worry about trying to duplicate the above code, I don't think that is important, but I am hoping it can help clue folks into something obviously wrong that I might be doing. If not, let me know and I'll see if I can create some replicable code (which will take some time). It might be blowing up because steps is somehow a dataframe within a dataframe in df. Which is not right. Our row definition is correct.

Attempts

I have tried the following approaches to no avail:

• Format row with a do.call method
• Cast row to data.frame (which it already is)
• Manipulating values in steps and interval in row.

I'm thinking there is obviously something wrong with my formatting but I can't fix a problem if I don't know what is wrong. The above is all I was able to gather from debugging the code. I'm frankly completely at a loss and would appreciate someone with more experience than me to indicate to me what is going wrong and how I can best address it? Any assistance, questions, or comments would be most welcome.

Answer 1

Simple Answer

Do the following: row$steps <- subset(avgStepsByInterval, interval == row$interval, steps)

Explanation

The problem was that steps was a dataframe object. Said in another way, we were trying to bind a row that had the desired dataframe definition to a dataframe object df that didn't have the proper definitions.

This is because row$steps <- subset(avgStepsByInterval, interval == row$interval, steps) is returning a dataframe object. So we simply need to rewrite it as follows:

row$steps <- as.numeric(subset(avgStepsByInterval, interval == row$interval, steps))

This means the definition for df and row will match.

Summary

If you are searching on stackoverflow and you are seeing this error. Be sure that the objects you are trying to rbind have a matching and intended structure.