1

I am new in Google prediction api. I created the bucket and uploaded the file to bucket.

I have the following the error in prediction.trainedmodels.get method.

But I tried training language_id.txt that file working fine.

what is the issue in my creation files.

{
   "kind": "prediction#training",
   "id": "sggurupreditmodel",
   "selfLink": "https://www.googleapis.com/prediction/v1.6/projects/623547173042/trainedmodels/s        ggurupreditmodel",
   "created": "2014-10-09T08:27:27.770Z",
   "trainingStatus": "ERROR: TOO FEW INSTANCES IN DATASET"
}


Contents of traning.csv file are:

'node.js'
'html"
'css'
'dhtml'
'php'
'jquery'
'xml'
'adobe'
'c++'
'c'
'vb'
'mysql'
'3D'
'typing'
'hr'
'copy right'
'crm'

Any suggestions, what is wrong here?

Ravi
  • 1,320
  • 12
  • 19
user3046205
  • 131
  • 1
  • 2
  • 13

1 Answers1

0

The csv formate is wrong.

instance:

"label1","feature1"
"label2","feature2"
"label3","feature3"

And, on the 2nd row about html, there is a double Quotation marks,it's maybe a error.

AdrieanKhisbe
  • 3,899
  • 8
  • 37
  • 45
lxc
  • 1
  • 2