diff --git a/docs/metrics/f1-score.md b/docs/metrics/f1-score.md index 5ae731c4a..a81ece056 100644 --- a/docs/metrics/f1-score.md +++ b/docs/metrics/f1-score.md @@ -1,5 +1,5 @@ --- -description: How to calculate and interpret precision to evaluate ML model performance +description: How to calculate and interpret F1-score to evaluate ML model performance --- # F1-score diff --git a/docs/metrics/index.md b/docs/metrics/index.md index 3b1b6bca9..f2b1a7ce1 100644 --- a/docs/metrics/index.md +++ b/docs/metrics/index.md @@ -1,7 +1,5 @@ --- icon: kolena/metrics-glossary-16 -hide: - - toc description: A glossary for machine learning metrics --- @@ -13,257 +11,57 @@ Each ML use case requires different metrics. Using the right metrics is critical comparing model performance. In each metrics guide, you can learn about the metric with examples, its limitations and biases, and its intended uses. -
- -- [Accuracy](accuracy.md) - - --- - - Accuracy measures how well a model predicts correctly. It's a good metric for assessing model performance in simple - cases with balanced data. - -- [Average Precision (AP)](average-precision.md) - - --- - - Average precision summarizes a precision-recall (PR) curve into a single threshold-independent value - representing model's performance across all thresholds. - -- [Averaging Methods: Macro, Micro, Weighted](averaging-methods.md) - - --- - - Different averaging methods for aggregating metrics for **multiclass** workflows, such as classification and - object detection. - -- [Coefficient of Determination (R²)](coefficient-of-determination.md) - - --- - - R² provides a scale that is generally from 0 to 1, where higher values indicate a better fit and imply that - the model can better explain the variation of the model inferences. - -- [Cohen's Kappa](cohens-kappa.md) - - --- - - Cohen's Kappa measures the agreement between model inferences and ground truths. It's a good performance metric - that considers class imbalance. - -- [Confusion Matrix](confusion-matrix.md) - - --- - - Confusion matrix is a structured plot describing classification model performance as a table that highlights counts - of objects with predicted classes (columns) against the actual classes (rows), indicating how confused a model is. - -- [F1-score](f1-score.md) - - --- - - F1-score is a metric that combines two competing metrics, [precision](precision.md) and - [recall](recall.md) with an equal weight. It symmetrically represents both precision and recall as one metric. - -- [False Positive Rate (FPR)](fpr.md) - - --- - - False positive rate (FPR) measures the proportion of negative ground truths that a model incorrectly predicts as - positive, ranging from 0 to 1. It is useful when the objective is to measure and reduce false positive inferences. - -- [Mean Absolute Error (MAE)](mean-absolute-error.md) - - --- - - MAE measures the average magnitude of errors in predictions, without considering their direction. It represents - the mean of the absolute differences between predicted and actual values across a dataset, treating each - discrepancy equally. - -- [Mean Squared Error (MSE)](mean-squared-error.md) - - --- - - MSE measures the average squared difference between the predicted values and the actual values, - emphasizing larger errors more significantly than smaller ones. It is mainly different from - [MAE](mean-absolute-error.md) in that larger errors are penalized more heavily. - -- [Precision](precision.md) - - --- - - Precision measures the proportion of positive inferences from a model that are correct. It is useful when the - objective is to measure and reduce false positive inferences. - -- [Precision-Recall (PR) Curve](pr-curve.md) - - --- - - Precision-recall curve is a plot that gauges machine learning model performance by using [precision](precision.md) - and [recall](recall.md). It is built with precision on the y-axis and recall on the x-axis computed across many - thresholds. - -- [Recall (TPR, Sensitivity)](recall.md) - - --- - - Recall, also known as true positive rate (TPR) and sensitivity, measures the proportion of all positive ground - truths that a model correctly predicts. It is useful when the objective is to measure and reduce false negative - ground truths, i.e. model misses. - -- [Receiver Operating Characteristic (ROC) Curve](roc-curve.md) - - --- - - A receiver operating characteristic (ROC) curve is a plot that is used to evaluate the performance of binary - classification models by using the [true positive rate (TPR)](./recall.md) and the - [false positive rate (FPR)](./fpr.md). - -- [Root Mean Squared Error (RMSE)](root-mean-squared-error.md) - - --- - - RMSE measures the square root of the average squared difference between the predicted values and the - actual values, emphasizing larger errors more significantly than smaller ones while maintaing the same - unit as the predicted and actual values which is what differentiates it from [MSE](mean-squared-error.md). - -- [Specificity (TNR)](specificity.md) - - --- - - Specificity, also known as true negative rate (TNR), measures the proportion of negative ground truths that a - model correctly predicts, ranging from 0 to 1. It is useful when the objective is to measure the model's ability to - correctly identify the negative class instances. - -- [TP / FP / FN / TN](tp-fp-fn-tn.md) - - --- - - The counts of TP, FP, FN and TN ground truths and inferences are essential for summarizing model performance. They - are the building blocks of many other metrics, including [accuracy](accuracy.md), [precision](precision.md), - and [recall](recall.md). - -
- -## Computer Vision - -
- -- [Geometry Matching](geometry-matching.md) - - --- - - Geometry matching is the process of matching inferences to ground truths for computer vision workflows with a - localization component. It is a core building block for metrics such as [TP, FP, and FN](tp-fp-fn-tn.md), and any - metrics built on top of these, like [precision](precision.md), [recall](recall.md), and - [F1-score](f1-score.md). - -- [Intersection over Union (IoU)](iou.md) - - --- - - IoU measures overlap between two geometries, segmentation masks, sets of labels, or time-series snippets. - Also known as Jaccard index in classification workflow. - -
- -## Large Language Models - -
- -- [Consistency Score](consistency-score.md) - - --- - - The consistency score is a sampling-based approach that computes a score by aggregating across `n` sample responses - obtained by using the same prompt `n` times. The model is considered hallucinating if there exists sample responses - that contradict one another or are not consistent. - -- [Contradiction Score](contradiction-score.md) - - --- - - NLI classification is an insightful tool to measure if candidate texts and reference texts are contradictory or - consistent, providing extra detail towards hallucination detection. - -- [HHEM Score](HHEM-score.md) - - --- - - [Hughes Hallucination Evaluation Model (HHEM)](https://huggingface.co/vectara/hallucination_evaluation_model) is an - open-source model that can be used to compute scores for hallucination detection. - -- [Prompt-based Hallucination Metric](prompt-based-hallucination-metric.md) - - --- - - Large Language Model (LLM) Prompt-based Metric involves using an LLM and various prompt engineering techniques - to perform evaluations. These methods include chain-of-thought, self-consistency, and others to determine whether or - not an inference text contains a hallucination. - -
- -## Natural Language Processing - -
- -- [BERTScore](bertscore.md) - - --- - - BERTScore is a metric used in NLP workflows to measure textual similarity between candidate texts and reference - texts. - -- [BLEU](bleu.md) - - --- - - BLEU is a metric commonly used in a variety of NLP workflows to evaluate the quality of candidate texts. BLEU can be - thought of as an analog to [precision](precision.md) for text comparisons. - -- [Diarization Error Rate](diarization-error-rate.md) - - --- - - Diarization Error Rate (DER) is an important metric used to evaluate speaker diarization systems. It quantifies the - overall performance of a speaker diarization system by measuring the ratio of the duration of errors to the total - ground truth speech duration. - -- [METEOR](meteor.md) - - --- - - METEOR is a widely recognized and vital metric used in NLP. It is used to measure the quality of candidate texts - against reference texts. Though it is an n-gram based metric, it goes beyond traditional methods by factoring in - elements such as precision, recall, and order to provide a comprehensive measure of text quality. - -- [Perplexity](perplexity.md) - - --- - - Perplexity is a metric commonly used in natural language processing to evaluate the quality of language models, - particularly in the context of text generation. Unlike metrics such as BLEU or BERT, perplexity doesn't directly - measure the quality of generated text by comparing it with reference texts. Instead, perplexity assesses the - "confidence" or "surprise" of a language model in predicting the next word in a sequence of words. - -- [ROUGE-N](rouge-n.md) - - --- - - ROUGE-N, a metric within the broader ROUGE metric collection, is a vital metric in the field of NLP. It assesses - the quality of a candidate text by measuring the overlap of n-grams between the candidate text and reference texts. - ROUGE-N can be thought of as an analog to [recall](recall.md) for text comparisons. - -- [Word, Character, and Match Error Rate](wer-cer-mer.md) - - --- - - Word Error Rate (WER), Character Error Rate (CER), and Match Error Rate (MER) are essential metrics used in the - evaluation of speech recognition and natural language processing systems. From a high level, they each quantify - the similarity between reference and candidate texts, with zero being a perfect score. - -
- -## Kolena Insights +### General Metrics + +| Metric | Description | +| --- | --- | +| [Accuracy](accuracy.md) | Accuracy measures how well a model predicts correctly. It's a good metric for assessing model performance in simple cases with balanced data. | +| [Average Precision (AP)](average-precision.md) | Average precision summarizes a precision-recall (PR) curve into a single threshold-independent value representing model's performance across all thresholds. | +| [Averaging Methods: Macro, Micro, Weighted](averaging-methods.md) | Different averaging methods for aggregating metrics for **multiclass** workflows, such as classification and object detection. | +| [Coefficient of Determination (R²)](coefficient-of-determination.md) | R² provides a scale that is generally from 0 to 1, where higher values indicate a better fit and imply that the model can better explain the variation of the model inferences. | +| [Cohen's Kappa](cohens-kappa.md) | Cohen's Kappa measures the agreement between model inferences and ground truths. It's a good performance metric that considers class imbalance. | +| [Confusion Matrix](confusion-matrix.md) | Confusion matrix is a structured plot describing classification model performance as a table that highlights counts of objects with predicted classes (columns) against the actual classes (rows), indicating how confused a model is. | +| [F1-score](f1-score.md) | F1-score is a metric that combines two competing metrics, [precision](precision.md) and [recall](recall.md) with an equal weight. It symmetrically represents both precision and recall as one metric. | +| [False Positive Rate (FPR)](fpr.md) | False positive rate (FPR) measures the proportion of negative ground truths that a model incorrectly predicts as positive, ranging from 0 to 1. It is useful when the objective is to measure and reduce false positive inferences. | +| [Mean Absolute Error (MAE)](mean-absolute-error.md) | MAE measures the average magnitude of errors in predictions, without considering their direction. It represents the mean of the absolute differences between predicted and actual values across a dataset, treating each discrepancy equally. | +| [Mean Squared Error (MSE)](mean-squared-error.md) | MSE measures the average squared difference between the predicted values and the actual values, emphasizing larger errors more significantly than smaller ones. It is mainly different from [MAE](mean-absolute-error.md) in that larger errors are penalized more heavily. | +| [Precision](precision.md) | Precision measures the proportion of positive inferences from a model that are correct. It is useful when the objective is to measure and reduce false positive inferences. | +| [Precision-Recall (PR) Curve](pr-curve.md) | Precision-recall curve is a plot that gauges machine learning model performance by using [precision](precision.md) and [recall](recall.md). It is built with precision on the y-axis and recall on the x-axis computed across many thresholds. | +| [Recall (TPR, Sensitivity)](recall.md) | Recall, also known as true positive rate (TPR) and sensitivity, measures the proportion of all positive ground truths that a model correctly predicts. It is useful when the objective is to measure and reduce false negative ground truths, i.e. model misses. | +| [Receiver Operating Characteristic (ROC) Curve](roc-curve.md) | A receiver operating characteristic (ROC) curve is a plot that is used to evaluate the performance of binary classification models by using the [true positive rate (TPR)](./recall.md) and the [false positive rate (FPR)](./fpr.md). | +| [Root Mean Squared Error (RMSE)](root-mean-squared-error.md) | RMSE measures the square root of the average squared difference between the predicted values and the actual values, emphasizing larger errors more significantly than smaller ones while maintaing the same unit as the predicted and actual values which is what differentiates it from [MSE](mean-squared-error.md). | +| [Specificity (TNR)](specificity.md) | Specificity, also known as true negative rate (TNR), measures the proportion of negative ground truths that a model correctly predicts, ranging from 0 to 1. It is useful when the objective is to measure the model's ability to correctly identify the negative class instances. | +| [TP / FP / FN / TN](tp-fp-fn-tn.md) | The counts of TP, FP, FN and TN ground truths and inferences are essential for summarizing model performance. They are the building blocks of many other metrics, including [accuracy](accuracy.md), [precision](precision.md), and [recall](recall.md). | + +### Computer Vision + +| Metric | Description | +| --- | --- | +| [Geometry Matching](geometry-matching.md) | Geometry matching is the process of matching inferences to ground truths for computer vision workflows with a localization component. It is a core building block for metrics such as [TP, FP, and FN](tp-fp-fn-tn.md), and any metrics built on top of these, like [precision](precision.md), [recall](recall.md), and [F1-score](f1-score.md). | +| [Intersection over Union (IoU)](iou.md) | IoU measures overlap between two geometries, segmentation masks, sets of labels, or time-series snippets. Also known as Jaccard index in classification workflow. | + +### Large Language Models + +| Metric | Description | +| --- | --- | +| [Consistency Score](consistency-score.md) | The consistency score is a sampling-based approach that computes a score by aggregating across `n` sample responses obtained by using the same prompt `n` times. The model is considered hallucinating if there exists sample responses that contradict one another or are not consistent. | +| [Contradiction Score](contradiction-score.md) | NLI classification is an insightful tool to measure if candidate texts and reference texts are contradictory or consistent, providing extra detail towards hallucination detection. | +| [HHEM Score](HHEM-score.md) | [Hughes Hallucination Evaluation Model (HHEM)](https://huggingface.co/vectara/hallucination_evaluation_model) is an open-source model that can be used to compute scores for hallucination detection. | +| [Prompt-based Hallucination Metric](prompt-based-hallucination-metric.md) | Large Language Model (LLM) Prompt-based Metric involves using an LLM and various prompt engineering techniques to perform evaluations. These methods include chain-of-thought, self-consistency, and others to determine whether or not an inference text contains a hallucination. | + +### Natural Language Processing + +| Metric | Description | +| --- | --- | +| [BERTScore](bertscore.md) | BERTScore is a metric used in NLP workflows to measure textual similarity between candidate texts and reference texts. | +| [BLEU](bleu.md) | BLEU is a metric commonly used in a variety of NLP workflows to evaluate the quality of candidate texts. BLEU can be thought of as an analog to [precision](precision.md) for text comparisons. | +| [Diarization Error Rate](diarization-error-rate.md) | Diarization Error Rate (DER) is an important metric used to evaluate speaker diarization systems. It quantifies the overall performance of a speaker diarization system by measuring the ratio of the duration of errors to the total ground truth speech duration. | +| [METEOR](meteor.md) | METEOR is a widely recognized and vital metric used in NLP. It is used to measure the quality of candidate texts against reference texts. Though it is an n-gram based metric, it goes beyond traditional methods by factoring in elements such as precision, recall, and order to provide a comprehensive measure of text quality. | +| [Perplexity](perplexity.md) | Perplexity is a metric commonly used in natural language processing to evaluate the quality of language models, particularly in the context of text generation. Unlike metrics such as BLEU or BERT, perplexity doesn't directly measure the quality of generated text by comparing it with reference texts. Instead, perplexity assesses the "confidence" or "surprise" of a language model in predicting the next word in a sequence of words. | +| [ROUGE-N](rouge-n.md) | ROUGE-N, a metric within the broader ROUGE metric collection, is a vital metric in the field of NLP. It assesses the quality of a candidate text by measuring the overlap of n-grams between the candidate text and reference texts. ROUGE-N can be thought of as an analog to [recall](recall.md) for text comparisons. | +| [Word, Character, and Match Error Rate](wer-cer-mer.md) | Word Error Rate (WER), Character Error Rate (CER), and Match Error Rate (MER) are essential metrics used in the evaluation of speech recognition and natural language processing systems. From a high level, they each quantify the similarity between reference and candidate texts, with zero being a perfect score. | + +### Kolena Insights
diff --git a/docs/metrics/specificity.md b/docs/metrics/specificity.md index 470724380..e02ae4838 100644 --- a/docs/metrics/specificity.md +++ b/docs/metrics/specificity.md @@ -1,5 +1,5 @@ --- -description: Understanding true negative rate (specificity) to evaluate ML model performance +description: How to compute and interpret true negative rate --- # Specificity (TNR) diff --git a/docs/metrics/tp-fp-fn-tn.md b/docs/metrics/tp-fp-fn-tn.md index 913a900f4..d7388dc32 100644 --- a/docs/metrics/tp-fp-fn-tn.md +++ b/docs/metrics/tp-fp-fn-tn.md @@ -1,12 +1,13 @@ --- -description: Learn about true positives, false positives, false negatives, and true negatives to evaluate ML model performance +description: Building blocks of ML model evaluation --- # TP / FP / FN / TN The counts of **true positive** (TP), **false positive** (FP), **false negative** (FN), and **true negative** (TN) ground truths and inferences are essential for summarizing model performance. These metrics are the building blocks of -many other metrics, including [accuracy](./accuracy.md), [precision](./precision.md), and [recall](./recall.md). +many other metrics, including [accuracy](./accuracy.md), [precision](./precision.md), [recall](./recall.md), +[F1-score](./f1-score.md), [false positive rate (FPR)](./fpr.md), and [specificity (TNR)](./specificity.md). | Metric | | Description | | --- | --- | --- |