Expand description
§use-ml
Facade crate for the focused machine-learning primitive crates in RustUse.
§Experimental
use-ml is experimental while the workspace remains below 0.3.0.
§Example
use use_ml::{MlDatasetName, MlFeatureName, MlModelName, TensorShape};
let dataset = MlDatasetName::new("iris")?;
let feature = MlFeatureName::new("sepal_width")?;
let model = MlModelName::new("baseline-classifier")?;
let shape = TensorShape::new([150, 4])?;
assert_eq!(dataset.as_str(), "iris");
assert_eq!(feature.as_str(), "sepal_width");
assert_eq!(model.as_str(), "baseline-classifier");
assert_eq!(shape.rank(), 2);§Scope
- Re-export the focused
use-ml-*primitive crates. - Keep implementation logic inside focused child crates.
- Provide one dependency for machine-learning metadata primitives.
§Relationship to use-ai
use-ml models machine-learning primitives: datasets, features, labels,
tensors, model artifacts, training, inference, evaluation, metrics, pipelines,
embeddings, experiments, and model documentation.
use-ai models AI interaction primitives: prompts, messages, roles, context
windows, tool calls, agents, RAG, reasoning, memory, guardrails, AI model
interfaces, and AI-specific evaluation.
These sets are siblings. They should interoperate conceptually but avoid dependency cycles.
§Non-goals
- Training, inference, serving, tensor math, vector search, registry behavior, or experiment tracking.
- Prompt, chat, agent, RAG, guardrail, or AI-provider interface modeling.
§License
Licensed under either Apache-2.0 or MIT.
Structs§
- Embedding
Dimension - Embedding
Model Name - Embedding
Vector Id - Embedding
Vector Shape - MlArtifact
Uri - MlBatch
Size - MlBenchmark
Name - MlClass
Id - MlClass
Name - MlConfidence
Score - MlConfusion
Matrix Shape - MlDataset
Card Ref - MlDataset
Id - MlDataset
License - MlDataset
Name - MlDataset
Schema Ref - MlDataset
Version - MlEpoch
Count - MlEval
Slice Name - MlEvaluation
RunId - MlExample
Id - MlExperiment
Id - MlExperiment
Name - MlFeature
Id - MlFeature
Name - MlHyperparameter
Name - MlHyperparameter
Value - MlInference
Request Id - MlLabel
Id - MlLabel
Name - MlLearning
Rate - MlMetric
Name - MlMetric
Value - MlModel
Card - MlModel
Card Dataset Ref - MlModel
Card Evaluation Summary - MlModel
Card Limitation - MlModel
Card Name - MlModel
Card Owner - MlModel
Id - MlModel
License - MlModel
Name - MlModel
Provider - MlModel
Version - MlParameter
Name - MlParameter
Value - MlPipeline
Id - MlPipeline
Name - MlPipeline
RunId - MlPipeline
Step Name - MlPrediction
Id - MlRunId
- MlRun
Tag - MlServing
Endpoint Name - MlThreshold
- MlTraining
JobName - MlTraining
RunId - Tensor
Axis - Tensor
Dim - Tensor
Rank - Tensor
Shape
Enums§
- Embedding
Distance Metric - Embedding
Error - Embedding
Index Kind - Embedding
Modality - Embedding
Normalization Kind - Embedding
Search Kind - Embedding
Vector Format - MlAnnotation
Kind - MlBatching
Kind - MlCheckpoint
Kind - MlClassification
Metric - MlClustering
Metric - MlDataset
Error - MlDataset
Kind - MlDataset
Provenance - MlDataset
Split - MlEval
Slice Kind - MlEvaluation
Error - MlEvaluation
Kind - MlEvaluation
Status - MlEvaluation
Target - MlExample
Kind - MlExperiment
Error - MlExperiment
Stage - MlFeature
Drift Status - MlFeature
Encoding Kind - MlFeature
Error - MlFeature
Kind - MlFeature
Missing Value Policy - MlFeature
Role - MlFeature
Scaling Kind - MlFeature
Source - MlFeature
Transform Kind - MlGeneration
Metric - MlInference
Error - MlInference
Mode - MlInference
Status - MlInput
Kind - MlLabel
Cardinality - MlLabel
Error - MlLabel
Kind - MlLabel
Quality - MlLabel
Source - MlLatency
Bucket - MlLoss
Kind - MlMetric
Aggregation - MlMetric
Direction - MlMetric
Error - MlMetric
Kind - MlModel
Architecture Kind - MlModel
Artifact Kind - MlModel
Card Audience - MlModel
Card Error - MlModel
Card Intended Use - MlModel
Card Risk - MlModel
Card Section - MlModel
Error - MlModel
Format - MlModel
Kind - MlModel
Stage - MlModel
Task - MlOptimizer
Kind - MlOutput
Kind - MlPipeline
Artifact Kind - MlPipeline
Dependency Kind - MlPipeline
Error - MlPipeline
Schedule Kind - MlPipeline
Status - MlPipeline
Step Kind - MlPipeline
Trigger Kind - MlRanking
Metric - MlRegression
Metric - MlRun
Status - MlServing
Kind - MlTarget
Kind - MlTracking
Backend Kind - MlTraining
Error - MlTraining
Phase - MlTraining
Status - MlValidation
Strategy - TensorD
Type - Tensor
Device Kind - Tensor
Layout - Tensor
Memory Format - Tensor
Shape Error