columbiadvmm / spread-out_local_feature_descriptor Goto Github PK

The test accuracy is always change. Did you get the average precision of the test data ?

I mean, If we train 10 epoch, we can get a precision. If we train 20 epoches, we can get another result. How did you solve this problem ?

def ErrorRateAt95Recall(labels, scores):
    recall_point = 0.95
    # Sort label-score tuples by the score in descending order.
    sorted_scores = zip(labels, scores)
    sorted_scores.sort(key=operator.itemgetter(1), reverse=False)

I meet an error when excute to the lines "sorted_score.sort().."
The error information is

'zip' object has not attribute 'sort'

Could you share me a solution ?

Hi, thank you very much for sharing the code.
I have checked the code based on hardnet that you provide, but I can't find the part of the implementation of ''Spread-out_Local_Feature_Descriptor'' in it. It seems that the code is just hardnet, could you tell me where is the part of spread-out descriptor in the hardnet code.
Thank you very much.

thank you for providing the download linking of the ubc dataset. I find there is a different version of ubc dataset (http://matthewalunbrown.com/patchdata/patchdata.html) and patches with different 3D point ID's in the file "info.txt" that have different reference image ID's in the file "interest.txt" may possibly correspond to the same 3D point in this dataset(http://matthewalunbrown.com/patchdata/patchdata.html). I wonder if it is true in the ubc dataset downloaded from your linking