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
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163

use crate::{glm::link::Link, LinearError, TweedieRegressor};
use linfa::{Float, ParamGuard};
use serde::{Deserialize, Serialize};

/// Generalized Linear Model (GLM) with a Tweedie distribution
///
/// The Regressor can be used to model different GLMs depending on
/// [`power`](struct.TweedieRegressor.html#method.power),
/// which determines the underlying distribution.
///
/// | Power  | Distribution           |
/// | ------ | ---------------------- |
/// | 0      | Normal                 |
/// | 1      | Poisson                |
/// | (1, 2) | Compound Poisson Gamma |
/// | 2      | Gamma                  |
/// | 3      | Inverse Gaussian       |
///
/// NOTE: No distribution exists between 0 and 1
///
/// Learn more from sklearn's excellent [User Guide](https://scikit-learn.org/stable/modules/linear_model.html#generalized-linear-regression)
///
/// ## Examples
///
/// Here's an example on how to train a GLM on the `diabetes` dataset
/// ```rust
/// use linfa::traits::{Fit, Predict};
/// use linfa_linear::TweedieRegressor;
/// use linfa::prelude::SingleTargetRegression;
///
/// let dataset = linfa_datasets::diabetes();
/// let model = TweedieRegressor::params().fit(&dataset).unwrap();
/// let pred = model.predict(&dataset);
/// let r2 = pred.r2(&dataset).unwrap();
/// println!("r2 from prediction: {}", r2);
/// ```
#[derive(Debug, Serialize, Deserialize, Clone, PartialEq)]
pub struct TweedieRegressorValidParams<F> {
    alpha: F,
    fit_intercept: bool,
    power: F,
    link: Option<Link>,
    max_iter: usize,
    tol: F,
}

impl<F: Float> TweedieRegressorValidParams<F> {
    pub fn alpha(&self) -> F {
        self.alpha
    }

    pub fn fit_intercept(&self) -> bool {
        self.fit_intercept
    }

    pub fn power(&self) -> F {
        self.power
    }

    pub fn link(&self) -> Link {
        match self.link {
            Some(x) => x,
            None if self.power <= F::zero() => Link::Identity,
            None => Link::Log,
        }
    }

    pub fn max_iter(&self) -> usize {
        self.max_iter
    }

    pub fn tol(&self) -> F {
        self.tol
    }
}

#[derive(Debug, Clone, PartialEq)]
pub struct TweedieRegressorParams<F>(TweedieRegressorValidParams<F>);

impl<F: Float> Default for TweedieRegressorParams<F> {
    fn default() -> Self {
        Self::new()
    }
}

impl<F: Float> TweedieRegressor<F> {
    pub fn params() -> TweedieRegressorParams<F> {
        TweedieRegressorParams::new()
    }
}

impl<F: Float> TweedieRegressorParams<F> {
    pub fn new() -> Self {
        Self(TweedieRegressorValidParams {
            alpha: F::one(),
            fit_intercept: true,
            power: F::one(),
            link: None,
            max_iter: 100,
            tol: F::cast(1e-4),
        })
    }

    /// Constant that multiplies with the penalty term and thus determines the
    /// regularization strenght. `alpha` set to 0 is equivalent to unpenalized GLM.
    pub fn alpha(mut self, alpha: F) -> Self {
        self.0.alpha = alpha;
        self
    }

    /// Specifies whether a bias or intercept should be added to the model
    pub fn fit_intercept(mut self, fit_intercept: bool) -> Self {
        self.0.fit_intercept = fit_intercept;
        self
    }

    /// The power determines the underlying target distribution
    pub fn power(mut self, power: F) -> Self {
        self.0.power = power;
        self
    }

    /// The link function of the GLM, for mapping from linear predictor `x @ coeff + intercept` to
    /// the prediction. If no value is set, the link will be selected based on the following,
    /// - [`identity`](Link::Identity) for Normal distribution (`power` = 0)
    /// - [`log`](Link::Log) for Poisson, Gamma and Inverse Gaussian distributions (`power` >= 1)
    pub fn link(mut self, link: Link) -> Self {
        self.0.link = Some(link);
        self
    }

    /// Maximum number of iterations for the LBFGS solver
    pub fn max_iter(mut self, max_iter: usize) -> Self {
        self.0.max_iter = max_iter;
        self
    }

    /// Stopping criterion for the LBFGS solver
    pub fn tol(mut self, tol: F) -> Self {
        self.0.tol = tol;
        self
    }
}

impl<F: Float> ParamGuard for TweedieRegressorParams<F> {
    type Checked = TweedieRegressorValidParams<F>;
    type Error = LinearError<F>;

    fn check_ref(&self) -> Result<&Self::Checked, Self::Error> {
        if self.0.alpha.is_sign_negative() {
            Err(LinearError::InvalidPenalty(self.0.alpha))
        } else if self.0.power > F::zero() && self.0.power < F::one() {
            Err(LinearError::InvalidTweediePower(self.0.power))
        } else {
            Ok(&self.0)
        }
    }

    fn check(self) -> Result<Self::Checked, Self::Error> {
        self.check_ref()?;
        Ok(self.0)
    }
}