datarust 0.6.5

Scikit-learn-style preprocessing and classical ML in Rust
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124

[package]
name = "datarust"
version = "0.6.5"
edition = "2021"
rust-version = "1.70"
description = "Scikit-learn-style preprocessing and classical ML in Rust"
license = "MIT"
repository = "https://github.com/genc-murat/datarust"
homepage = "https://datarust.dev"
readme = "README.md"
keywords = ["machine-learning", "scikit-learn", "preprocessing", "regression", "linear-model"]
categories = ["science", "mathematics", "algorithms"]
include = [
    "/src/**",
    "/tests/**",
    "/examples/**",
    "/benches/**",
    "/Cargo.toml",
    "/Cargo.lock",
    "/README.md",
    "/CHANGELOG.md",
    "/LICENSE",
    "/ROADMAP.md",
    "/ARCHITECTURE.md",
]

[lib]
name = "datarust"
path = "src/lib.rs"

[features]
default = []
# Enable JSON serialization / save-load support for fitted transformers.
serde = [
    "dep:serde",
    "dep:serde_json",
    "dep:proc-macro2",
    "dep:quote",
    "dep:ryu",
    "dep:syn",
    "dep:unicode-ident",
]
# Enable parallelism via rayon for large datasets.
rayon = ["dep:rayon", "dep:rayon-core"]
# Enable a tuned pure-Rust GEMM (via the `matrixmultiply` crate) for matrix
# products and covariance computation. Significantly speeds up PCA,
# TruncatedSVD and `Matrix::matmul` on large dense inputs. The default build
# remains zero-external-dependency; opt in with this feature.
matrixmultiply = ["dep:matrixmultiply"]
# Embed small classic datasets (Iris, Breast Cancer, Wine, Diabetes) for
# examples, tests, and onboarding. The data is compiled into the binary as
# `const` arrays, adding no runtime or external-file dependency.
datasets = []

[dependencies]
serde = { version = "=1.0.228", features = ["derive"], optional = true }
serde_json = { version = "=1.0.145", optional = true }
# Cargo 1.70 does not filter newer, MSRV-incompatible transitive releases while
# resolving a library for downstream users. These optional constraints keep the
# Serde derive/JSON graph compatible even when this crate's lockfile is ignored.
proc-macro2 = { version = "=1.0.106", optional = true }
quote = { version = "=1.0.44", optional = true }
ryu = { version = "=1.0.22", optional = true }
syn = { version = "=2.0.114", optional = true }
unicode-ident = { version = "=1.0.22", optional = true }
# Exact Rayon pins keep the advertised Rust 1.70 MSRV. Rayon 1.11+ and
# rayon-core 1.13+ require Rust 1.80.
rayon = { version = "=1.10.0", optional = true }
rayon-core = { version = "=1.12.1", optional = true }
matrixmultiply = { version = "=0.3.10", optional = true }

[dev-dependencies]
criterion = "=0.5.1"
proptest = "=1.5.0"
# Criterion accepts any Clap 4 release; keep its test/bench graph on the crate's
# Rust 1.70 MSRV as well.
clap = "=4.4.18"
# Proptest accepts newer tempfile releases whose manifests require Edition
# 2024; pin the last verified Rust 1.70-compatible line.
tempfile = "=3.15.0"
# Criterion's Ciborium dependency permits newer half releases with a higher
# MSRV; constrain the shared test graph to Rust 1.70.
half = "=2.4.1"

[[bench]]
name = "benchmarks"
harness = false

[[example]]
name = "basic_preprocessing"
path = "examples/basic_preprocessing.rs"

[[example]]
name = "pipeline_workflow"
path = "examples/pipeline_workflow.rs"

[[example]]
name = "target_encoding"
path = "examples/target_encoding.rs"

[[example]]
name = "regression_workflow"
path = "examples/regression_workflow.rs"

[[example]]
name = "classification_workflow"
path = "examples/classification_workflow.rs"

[[example]]
name = "regularization_comparison"
path = "examples/regularization_comparison.rs"

[[example]]
name = "model_persistence"
path = "examples/model_persistence.rs"
required-features = ["serde"]

[[example]]
name = "kmeans_clustering"
path = "examples/kmeans_clustering.rs"

[package.metadata.docs.rs]
features = ["serde", "rayon", "matrixmultiply", "datasets"]
rustdoc-args = ["--cfg", "docsrs"]