google prediction api示例“hello world”中结果不一样？

Question

实例中输入法语 Muy Bueno ，就会显示French,Spanish,English猜中的几率。
{
"kind": "prediction#output",
"id": "languageidentifier",
"selfLink": "https://www.googleapis.com/prediction/v1.6/projects/12345678910/trainedmodels/languageidentifier/predict",
"outputLabel": "Spanish",
"outputMulti": [
{
"label": "French",
"score": "0.334130"
},
{
"label": "Spanish",
"score": "0.418339"
},
{
"label": "English",
"score": "0.247531"
}
]
}

但是我的结果却是：
    {
 "kind": "prediction#output",
 "id": "languageidentifier",
 "selfLink": "https://www.googleapis.com/prediction/v1.6/projects/891700192979/trainedmodels/languageidentifier/predict",
 "outputLabel": "English",
 "outputMulti": [
  {
   "label": "English",
   "score": "1.000000"
  },
  {
   "label": "French",
   "score": "0.000000"
  },
  {
   "label": "Spanish",
   "score": "0.000000"
  }
 ]
}

根本没有几率。是输入长一点的语句，就能判断正确，但是也是一锤子买卖，不是1.0000，就是0.00000。
我的训练集都是按照helloworld例子去做的，有什么问题吗？

Answer 1

Great ideas, i love your article. It gives me more ideas on home to organize my room with my shoes and stuff for my children.Keep up the good work

• minion rush