从 AutoML PredictResponse 对象返回分数

Question

I'm trying to pull the score values out of my google automl PredictResponse Object. the object returned from the API looks as follows:

payload {
  classification {
    score: 0.989063024521
  }
  display_name: "No"
}
payload {
  classification {
    score: 0.0109369996935
  }
  display_name: "Yes"
}

I've gotten close using this:

result.payload[1]

Which returns:

classification {
  score: 0.0109369996935
}
display_name: "Yes"

But I still can't get to just the score for a yes or a no.

I've tried using simplejson, but the PredictResponse Object doesn't play nice with that either. Is there a function within the Object I can use to get to the Score for yes and the score for no? Appreciate the help!

Answer 1

result.payload[1].classification.score

Answer 2

You may use the this code

d = {} for i in range(len(response.payload)): d[response.payload[i].display_name] = response.payload[i].classification.score

It will provide you a dictionary for all the categories (Yes and No in your case)

Answer 3

 predicted_data = []
 for i in range(len(response.payload)):
   predicted_data.append({
        "key": response.payload[i].display_name,
        "value": response.payload[i].classification.score
    })

key-value list of payload objects

Answer 4

You can extract the display_name field using the following:

result.payload[N].display_name

From the Vision and Natural Language AutoML documentation, you can use the loop below to return the display_name of your predictions.

for result in response.payload:
    print("Predicted class name: {}".format(result.display_name))

Answer 5

这是您问题的答案：

result.payload[0].display_name