Mirror of forum.nim-lang.org

7627 :: Data type mismatch issue

• livingstone1337 (orginal) [2021-03-15T03:27:55+01:00] view original

  Hi could you give me some advice on this code I wrote. I got a type mismatch error related to the for statement. I tried converting it to int but that doesn't work.

  proc dataset_minmax(dataset : seq[float]): float =

  var minmax : seq[float]

  for i in 0..dataset[0].len:

  var col_values = row[i] for row in dataset var value_min = min(col_values) var value_max = max(col_values) minmax.append([value_min, value_max])

---

• inventormatt (orginal) [2021-03-15T03:47:09+01:00] view original

  are you trying to get the min and max value of a 1-dimensional seq? if so you can just do

  proc dataset_minmax(dataset : seq[float]): (float, float) =
    return (dataset.min, dataset.max)
  
  Run

  also as a sidenote you can encase your code inside of triple backticks to format it like code in the forum

---

• livingstone1337 (orginal) [2021-03-15T03:54:00+01:00] view original

  cool thanks I'll try it out. Also thanks for the tip about posting code

---

• DeletedUser (orginal) [2021-03-15T06:57:56+01:00] view original

  You are calling dataset[0].len. dataset[0] is a float and you cannot call len on it. I'm guessing from this code that you want dataset: seq[seq[float]] and not dataset: seq[float]. Also you should either do 0..<dataset[0].len or 0 .. dataset[0].len - 1 or 0..dataset[0].high.

  row[i] for row in dataset is not valid Nim syntax. You have to do:

  var col_values = newSeq[float](dataset.len)
  for j in 0..<dataset.len:
    col_values[j] = dataset[j][i]
  Run

  or some kind of list comprehension macro like sugar.collect.

  There is no append procedure, the right name is add. And the type of [dataset.min, dataset.max] is array[2, float], adding this to minmax will make it become [min0, max0, min1, max1...]. So you might want to change minmax to seq[seq[float]] and do minmax.add(@[dataset.min, dataset.max]) or, even better, to seq[(float, float)] and minmax.add((dataset.min, dataset.max)).

---