vision-evaluation 0.1.7

Evaluation code for vision tasks.

Introduction

This repo contains evaluation metric codes used in Microsoft Cognitive Services Computer Vision for tasks such as classification and object detection.

Functionalities

This repo currently offers evaluation metrics for two vision tasks:

• Image classification:
  • TopKAccuracyEvaluator: computes the top-k accuracy for multi-class classification problem. A prediction is considered correct, if the ground truth label is within the labels with top k confidences.
  • ThresholdAccuracyEvaluator: computes the threshold based accuracy (mainly for multi-label classification problem), i.e., accuracy of the predictions with confidence over a certain threshold.
  • AveragePrecisionEvaluator: computes the average precision, i.e., precision averaged across different confidence thresholds.
  • PrecisionEvaluator: computes precision
  • RecallEvaluator: computes recall
  • F1ScoreEvaluator: computes f1-score (recall and precision will be reported as well)
  • EceLossEvaluator: computes the ECE loss, i.e., the expected calibration error, given the model confidence and true labels for a set of data points.
• Object detection:
  • MeanAveragePrecisionEvaluatorForSingleIOU, evaluators.MeanAveragePrecisionEvaluatorForMultipleIOUs: computes the mean average precision (mAP), i.e. mean average precision across different classes, under single or multiple IoU(s).

While different machine learning problems/applications prefer different metrics, below are some general recommendations:

• Multiclass classification: Top-1 accuracy and Top-5 accuracy
• Multilabel classification: Average precision, Precision/recall/precision@k/threshold, where k and threshold can be very problem-specific
• Object detection: mAP@IoU=30 and mAP@IoU=50