Module metric
Expand description
The metric module.
Modules§
- state
- State module.
- store
- Module responsible to save and exposes data collected during training.
- vision
- Metrics module for vision tasks.
Structs§
- Accuracy
Input - The accuracy metric input type.
- Accuracy
Metric - The accuracy metric.
- Auroc
Input - The AUROC metric input type.
- Auroc
Metric - The Area Under the Receiver Operating Characteristic Curve (AUROC, also referred to as ROC AUC) for binary classification.
- CerInput
- The character error rate metric input type.
- Char
Error Rate - Character error rate (CER) is defined as the edit distance (e.g. Levenshtein distance) between the predicted and reference character sequences, divided by the total number of characters in the reference. This metric is commonly used in tasks such as speech recognition, OCR, or text generation to quantify how closely the predicted output matches the ground truth at a character level.
- Confusion
Stats Input - Input for confusion statistics error types.
- CpuMemory
- Memory information
- CpuTemperature
- CPU Temperature in celsius degrees
- CpuUse
- General CPU Usage metric
- Cuda
Metric - Track basic cuda infos.
- FBeta
Score Metric - The F-beta score metric.
- Hamming
Score - The hamming score, sometimes referred to as multi-label or label-based accuracy.
- Hamming
Score Input - The hamming score input type.
- Iteration
Speed Metric - The loss metric.
- Learning
Rate Metric - Track the learning rate across iterations.
- Loss
Input - The loss metric input type.
- Loss
Metric - The loss metric.
- Metric
Definition - Definition of a metric.
- Metric
Entry - Data type that contains the current state of a metric at a given time.
- Metric
Id - Metric id that can be used to compare metrics and retrieve entries of the same metric. For now we take the name as id to make sure that the same metric has the same id across different runs.
- Metric
Metadata - Metric metadata that can be used when computing metrics.
- Numeric
Attributes - Attributes that describe intrinsic properties of a numeric metric.
- Perplexity
Input - The perplexity metric input type.
- Perplexity
Metric - The perplexity metric.
- Precision
Metric - The Precision Metric
- Recall
Metric - The Recall Metric
- Serialized
Entry - Serialized form of a metric entry.
- TopK
Accuracy Input - The top-k accuracy metric input type.
- TopK
Accuracy Metric - The Top-K accuracy metric.
- WerInput
- The word error rate metric input type.
- Word
Error Rate - The word error rate (WER) metric, similar to the CER, is defined as the edit distance (e.g. Levenshtein distance) between the predicted and reference word sequences, divided by the total number of words in the reference. Here, the “units” within the sequences are words.
Enums§
- Class
Reduction - The reduction strategy for classification metrics.
- Metric
Attributes - Metric attributes define the properties intrinsic to different types of metric.
- Numeric
Entry - Numeric metric entry.
Traits§
- Adaptor
- Adaptor are used to transform types so that they can be used by metrics.
- Item
Lazy - Items that are lazy are not ready to be processed by metrics.
- Metric
- Metric trait.
- Numeric
- Declare a metric to be numeric.
Functions§
- format_
float - Format a float with the given precision. Will use scientific notation if necessary.
Type Aliases§
- Metric
Name - Type used to store metric names efficiently.