Model Evaluation

Classification metrics
Accuracy: computes the number of correct predictions divided by the total number of samples.
​Accuracy=number correct predictionsnumber of samplesAccuracy = \frac{number \space correct \space predictions}{number \space of \space samples}Accuracy=number of samplesnumber correct predictions​
​
Sensitivity: also known as recall, is computed as the fraction of true positives that are correctly identified.
​Sensitivity=number of true positivesnumber of true positives+number of false negativesSensitivity = \frac{number \space of \space true \space positives}{number \space of \space true \space positives + number \space of \space false \space negatives}Sensitivity=number of true positives+number of false negativesnumber of true positives​
​
Precision: computed as the fraction of retrieved instances that are relevant.
​Precision=number of true positivesnumber of true positives+number of false positivesPrecision = \frac{number \space of \space true \space positives}{number \space of \space true \space positives + number \space of \space false \space positives}Precision=number of true positives+number of false positivesnumber of true positives​
​
Specificity: computed as the fraction of true negatives that are correctly identified.
​Specificity=number of true negativesnumber of true negatives+number of false positivesSpecificity = \frac{number \space of \space true \space negatives}{number \space of \space true \space negatives + number \space of \space false \space positives}Specificity=number of true negatives+number of false positivesnumber of true negatives​
​
​
🚧 This section is still under construction!

Train and Evaluate - Previous

Training Neural Networks

Next - Productize It

ML Life Cycle

Last modified 1yr ago

Copy link