unimib-thesis-mad
/
reCTBN
1
0
Fork 0
Code Issues 9 Pull Requests 1 Releases Activity

Refactored `src/` and `tests/` files to be compliant to `rustfmt`

Browse Source
pull/55/head
Meliurwen 2 years ago
parent d6e93b351c
commit 780515707c
16 changed files with 247 additions and 252 deletions
Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL
Show Stats Download Patch File Download Diff File
  1. 82
      src/ctbn.rs
  2. 7
      src/lib.rs
  3. 17
      src/network.rs
  4. 51
      src/parameter_learning.rs
  5. 44
      src/params.rs
  6. 9
      src/structure_learning.rs
  7. 2
      src/structure_learning/constraint_based_algorithm.rs
  8. 24
      src/structure_learning/hypothesis_test.rs
  9. 6
      src/structure_learning/score_based_algorithm.rs
  10. 54
      src/structure_learning/score_function.rs
  11. 13
      src/tools.rs
  12. 3
      tests/ctbn.rs
  13. 20
      tests/parameter_learning.rs
  14. 3
      tests/params.rs
  15. 75
      tests/structure_learning.rs
  16. 13
      tests/utils.rs

82
src/ctbn.rs
Unescape View File

@ -1,10 +1,9 @@
use ndarray::prelude::*;
use crate::params::{StateType, Params, ParamsTrait};
use crate::network;
use std::collections::BTreeSet; use std::collections::BTreeSet;
use ndarray::prelude::*;
use crate::network;
use crate::params::{Params, ParamsTrait, StateType};
///CTBN network. It represents both the structure and the parameters of a CTBN. CtbnNetwork is ///CTBN network. It represents both the structure and the parameters of a CTBN. CtbnNetwork is
///composed by the following elements: ///composed by the following elements:
@ -54,30 +53,30 @@ use std::collections::BTreeSet;
/// ``` /// ```
pub struct CtbnNetwork { pub struct CtbnNetwork {
adj_matrix: Option<Array2<u16>>, adj_matrix: Option<Array2<u16>>,
nodes: Vec<Params> nodes: Vec<Params>,
} }
impl CtbnNetwork { impl CtbnNetwork {
pub fn new() -> CtbnNetwork { pub fn new() -> CtbnNetwork {
CtbnNetwork { CtbnNetwork {
adj_matrix: None, adj_matrix: None,
nodes: Vec::new() nodes: Vec::new(),
} }
} }
} }
impl network::Network for CtbnNetwork { impl network::Network for CtbnNetwork {
fn initialize_adj_matrix(&mut self) { fn initialize_adj_matrix(&mut self) {
self.adj_matrix = Some(Array2::<u16>::zeros((self.nodes.len(), self.nodes.len()).f())); self.adj_matrix = Some(Array2::<u16>::zeros(
(self.nodes.len(), self.nodes.len()).f(),
));
} }
fn add_node(&mut self, mut n: Params) -> Result<usize, network::NetworkError> { fn add_node(&mut self, mut n: Params) -> Result<usize, network::NetworkError> {
n.reset_params(); n.reset_params();
self.adj_matrix = Option::None; self.adj_matrix = Option::None;
self.nodes.push(n); self.nodes.push(n);
Ok(self.nodes.len() -1) Ok(self.nodes.len() - 1)
} }
fn add_edge(&mut self, parent: usize, child: usize) { fn add_edge(&mut self, parent: usize, child: usize) {
@ -91,7 +90,7 @@ impl network::Network for CtbnNetwork {
} }
} }
fn get_node_indices(&self) -> std::ops::Range<usize>{ fn get_node_indices(&self) -> std::ops::Range<usize> {
0..self.nodes.len() 0..self.nodes.len()
} }
@ -99,64 +98,65 @@ impl network::Network for CtbnNetwork {
self.nodes.len() self.nodes.len()
} }
fn get_node(&self, node_idx: usize) -> &Params{ fn get_node(&self, node_idx: usize) -> &Params {
&self.nodes[node_idx] &self.nodes[node_idx]
} }
fn get_node_mut(&mut self, node_idx: usize) -> &mut Params {
fn get_node_mut(&mut self, node_idx: usize) -> &mut Params{
&mut self.nodes[node_idx] &mut self.nodes[node_idx]
} }
fn get_param_index_network(&self, node: usize, current_state: &Vec<StateType>) -> usize {
fn get_param_index_network(&self, node: usize, current_state: &Vec<StateType>) -> usize{ self.adj_matrix
self.adj_matrix.as_ref().unwrap().column(node).iter().enumerate().fold((0, 1), |mut acc, x| { .as_ref()
.unwrap()
.column(node)
.iter()
.enumerate()
.fold((0, 1), |mut acc, x| {
if x.1 > &0 { if x.1 > &0 {
acc.0 += self.nodes[x.0].state_to_index(&current_state[x.0]) * acc.1; acc.0 += self.nodes[x.0].state_to_index(&current_state[x.0]) * acc.1;
acc.1 *= self.nodes[x.0].get_reserved_space_as_parent(); acc.1 *= self.nodes[x.0].get_reserved_space_as_parent();
} }
acc acc
}).0 })
.0
} }
fn get_param_index_from_custom_parent_set(
fn get_param_index_from_custom_parent_set(&self, current_state: &Vec<StateType>, parent_set: &BTreeSet<usize>) -> usize { &self,
parent_set.iter().fold((0, 1), |mut acc, x| { current_state: &Vec<StateType>,
parent_set: &BTreeSet<usize>,
) -> usize {
parent_set
.iter()
.fold((0, 1), |mut acc, x| {
acc.0 += self.nodes[*x].state_to_index(&current_state[*x]) * acc.1; acc.0 += self.nodes[*x].state_to_index(&current_state[*x]) * acc.1;
acc.1 *= self.nodes[*x].get_reserved_space_as_parent(); acc.1 *= self.nodes[*x].get_reserved_space_as_parent();
acc acc
}).0 })
.0
} }
fn get_parent_set(&self, node: usize) -> BTreeSet<usize> { fn get_parent_set(&self, node: usize) -> BTreeSet<usize> {
self.adj_matrix.as_ref() self.adj_matrix
.as_ref()
.unwrap() .unwrap()
.column(node) .column(node)
.iter() .iter()
.enumerate() .enumerate()
.filter_map(|(idx, x)| { .filter_map(|(idx, x)| if x > &0 { Some(idx) } else { None })
if x > &0 { .collect()
Some(idx)
} else {
None
}
}).collect()
} }
fn get_children_set(&self, node: usize) -> BTreeSet<usize>{ fn get_children_set(&self, node: usize) -> BTreeSet<usize> {
self.adj_matrix.as_ref() self.adj_matrix
.as_ref()
.unwrap() .unwrap()
.row(node) .row(node)
.iter() .iter()
.enumerate() .enumerate()
.filter_map(|(idx, x)| { .filter_map(|(idx, x)| if x > &0 { Some(idx) } else { None })
if x > &0 { .collect()
Some(idx)
} else {
None
}
}).collect()
} }
} }

7
src/lib.rs
Unescape View File

@ -2,10 +2,9 @@
#[cfg(test)] #[cfg(test)]
extern crate approx; extern crate approx;
pub mod params;
pub mod network;
pub mod ctbn; pub mod ctbn;
pub mod tools; pub mod network;
pub mod parameter_learning; pub mod parameter_learning;
pub mod params;
pub mod structure_learning; pub mod structure_learning;
pub mod tools;

17
src/network.rs
Unescape View File

@ -1,15 +1,16 @@
use std::collections::BTreeSet;
use thiserror::Error; use thiserror::Error;
use crate::params; use crate::params;
use std::collections::BTreeSet;
/// Error types for trait Network /// Error types for trait Network
#[derive(Error, Debug)] #[derive(Error, Debug)]
pub enum NetworkError { pub enum NetworkError {
#[error("Error during node insertion")] #[error("Error during node insertion")]
NodeInsertionError(String) NodeInsertionError(String),
} }
///Network ///Network
///The Network trait define the required methods for a structure used as pgm (such as ctbn). ///The Network trait define the required methods for a structure used as pgm (such as ctbn).
pub trait Network { pub trait Network {
@ -26,13 +27,17 @@ pub trait Network {
///Compute the index that must be used to access the parameters of a node given a specific ///Compute the index that must be used to access the parameters of a node given a specific
///configuration of the network. Usually, the only values really used in *current_state* are ///configuration of the network. Usually, the only values really used in *current_state* are
///the ones in the parent set of the *node*. ///the ones in the parent set of the *node*.
fn get_param_index_network(&self, node: usize, current_state: &Vec<params::StateType>) -> usize; fn get_param_index_network(&self, node: usize, current_state: &Vec<params::StateType>)
-> usize;
///Compute the index that must be used to access the parameters of a node given a specific ///Compute the index that must be used to access the parameters of a node given a specific
///configuration of the network and a generic parent_set. Usually, the only values really used ///configuration of the network and a generic parent_set. Usually, the only values really used
///in *current_state* are the ones in the parent set of the *node*. ///in *current_state* are the ones in the parent set of the *node*.
fn get_param_index_from_custom_parent_set(&self, current_state: &Vec<params::StateType>, parent_set: &BTreeSet<usize>) -> usize; fn get_param_index_from_custom_parent_set(
&self,
current_state: &Vec<params::StateType>,
parent_set: &BTreeSet<usize>,
) -> usize;
fn get_parent_set(&self, node: usize) -> BTreeSet<usize>; fn get_parent_set(&self, node: usize) -> BTreeSet<usize>;
fn get_children_set(&self, node: usize) -> BTreeSet<usize>; fn get_children_set(&self, node: usize) -> BTreeSet<usize>;
} }

51
src/parameter_learning.rs
Unescape View File

@ -1,11 +1,12 @@
use crate::network;
use crate::params::*;
use crate::tools;
use ndarray::prelude::*;
use std::collections::BTreeSet; use std::collections::BTreeSet;
pub trait ParameterLearning{ use ndarray::prelude::*;
fn fit<T:network::Network>(
use crate::params::*;
use crate::{network, tools};
pub trait ParameterLearning {
fn fit<T: network::Network>(
&self, &self,
net: &T, net: &T,
dataset: &tools::Dataset, dataset: &tools::Dataset,
@ -14,24 +15,19 @@ pub trait ParameterLearning{
) -> Params; ) -> Params;
} }
pub fn sufficient_statistics<T:network::Network>( pub fn sufficient_statistics<T: network::Network>(
net: &T, net: &T,
dataset: &tools::Dataset, dataset: &tools::Dataset,
node: usize, node: usize,
parent_set: &BTreeSet<usize> parent_set: &BTreeSet<usize>,
) -> (Array3<usize>, Array2<f64>) { ) -> (Array3<usize>, Array2<f64>) {
//Get the number of values assumable by the node //Get the number of values assumable by the node
let node_domain = net let node_domain = net.get_node(node.clone()).get_reserved_space_as_parent();
.get_node(node.clone())
.get_reserved_space_as_parent();
//Get the number of values assumable by each parent of the node //Get the number of values assumable by each parent of the node
let parentset_domain: Vec<usize> = parent_set let parentset_domain: Vec<usize> = parent_set
.iter() .iter()
.map(|x| { .map(|x| net.get_node(x.clone()).get_reserved_space_as_parent())
net.get_node(x.clone())
.get_reserved_space_as_parent()
})
.collect(); .collect();
//Vector used to convert a specific configuration of the parent_set to the corresponding index //Vector used to convert a specific configuration of the parent_set to the corresponding index
@ -70,13 +66,11 @@ pub fn sufficient_statistics<T:network::Network>(
} }
return (M, T); return (M, T);
} }
pub struct MLE {} pub struct MLE {}
impl ParameterLearning for MLE { impl ParameterLearning for MLE {
fn fit<T: network::Network>( fn fit<T: network::Network>(
&self, &self,
net: &T, net: &T,
@ -84,7 +78,6 @@ impl ParameterLearning for MLE {
node: usize, node: usize,
parent_set: Option<BTreeSet<usize>>, parent_set: Option<BTreeSet<usize>>,
) -> Params { ) -> Params {
//Use parent_set from parameter if present. Otherwise use parent_set from network. //Use parent_set from parameter if present. Otherwise use parent_set from network.
let parent_set = match parent_set { let parent_set = match parent_set {
Some(p) => p, Some(p) => p,
@ -96,7 +89,7 @@ impl ParameterLearning for MLE {
let mut CIM: Array3<f64> = Array::zeros((M.shape()[0], M.shape()[1], M.shape()[2])); let mut CIM: Array3<f64> = Array::zeros((M.shape()[0], M.shape()[1], M.shape()[2]));
CIM.axis_iter_mut(Axis(2)) CIM.axis_iter_mut(Axis(2))
.zip(M.mapv(|x| x as f64).axis_iter(Axis(2))) .zip(M.mapv(|x| x as f64).axis_iter(Axis(2)))
.for_each(|(mut C, m)| C.assign(&(&m/&T))); .for_each(|(mut C, m)| C.assign(&(&m / &T)));
//Set the diagonal of the inner matrices to the the row sum multiplied by -1 //Set the diagonal of the inner matrices to the the row sum multiplied by -1
let tmp_diag_sum: Array2<f64> = CIM.sum_axis(Axis(2)).mapv(|x| x * -1.0); let tmp_diag_sum: Array2<f64> = CIM.sum_axis(Axis(2)).mapv(|x| x * -1.0);
@ -106,8 +99,6 @@ impl ParameterLearning for MLE {
C.diag_mut().assign(&diag); C.diag_mut().assign(&diag);
}); });
let mut n: Params = net.get_node(node).clone(); let mut n: Params = net.get_node(node).clone();
match n { match n {
@ -115,8 +106,6 @@ impl ParameterLearning for MLE {
dsct.set_cim_unchecked(CIM); dsct.set_cim_unchecked(CIM);
dsct.set_transitions(M); dsct.set_transitions(M);
dsct.set_residence_time(T); dsct.set_residence_time(T);
} }
}; };
return n; return n;
@ -125,7 +114,7 @@ impl ParameterLearning for MLE {
pub struct BayesianApproach { pub struct BayesianApproach {
pub alpha: usize, pub alpha: usize,
pub tau: f64 pub tau: f64,
} }
impl ParameterLearning for BayesianApproach { impl ParameterLearning for BayesianApproach {
@ -151,7 +140,7 @@ impl ParameterLearning for BayesianApproach {
let mut CIM: Array3<f64> = Array::zeros((M.shape()[0], M.shape()[1], M.shape()[2])); let mut CIM: Array3<f64> = Array::zeros((M.shape()[0], M.shape()[1], M.shape()[2]));
CIM.axis_iter_mut(Axis(2)) CIM.axis_iter_mut(Axis(2))
.zip(M.mapv(|x| x as f64).axis_iter(Axis(2))) .zip(M.mapv(|x| x as f64).axis_iter(Axis(2)))
.for_each(|(mut C, m)| C.assign(&(&m.mapv(|y| y + alpha)/&T.mapv(|y| y + tau)))); .for_each(|(mut C, m)| C.assign(&(&m.mapv(|y| y + alpha) / &T.mapv(|y| y + tau))));
//Set the diagonal of the inner matrices to the the row sum multiplied by -1 //Set the diagonal of the inner matrices to the the row sum multiplied by -1
let tmp_diag_sum: Array2<f64> = CIM.sum_axis(Axis(2)).mapv(|x| x * -1.0); let tmp_diag_sum: Array2<f64> = CIM.sum_axis(Axis(2)).mapv(|x| x * -1.0);
@ -161,8 +150,6 @@ impl ParameterLearning for BayesianApproach {
C.diag_mut().assign(&diag); C.diag_mut().assign(&diag);
}); });
let mut n: Params = net.get_node(node).clone(); let mut n: Params = net.get_node(node).clone();
match n { match n {
@ -170,27 +157,25 @@ impl ParameterLearning for BayesianApproach {
dsct.set_cim_unchecked(CIM); dsct.set_cim_unchecked(CIM);
dsct.set_transitions(M); dsct.set_transitions(M);
dsct.set_residence_time(T); dsct.set_residence_time(T);
} }
}; };
return n; return n;
} }
} }
pub struct Cache<P: ParameterLearning> { pub struct Cache<P: ParameterLearning> {
parameter_learning: P, parameter_learning: P,
dataset: tools::Dataset, dataset: tools::Dataset,
} }
impl<P: ParameterLearning> Cache<P> { impl<P: ParameterLearning> Cache<P> {
pub fn fit<T:network::Network>( pub fn fit<T: network::Network>(
&mut self, &mut self,
net: &T, net: &T,
node: usize, node: usize,
parent_set: Option<BTreeSet<usize>>, parent_set: Option<BTreeSet<usize>>,
) -> Params { ) -> Params {
self.parameter_learning.fit(net, &self.dataset, node, parent_set) self.parameter_learning
.fit(net, &self.dataset, node, parent_set)
} }
} }

44
src/params.rs
Unescape View File

@ -1,9 +1,10 @@
use std::collections::BTreeSet;
use enum_dispatch::enum_dispatch; use enum_dispatch::enum_dispatch;
use ndarray::prelude::*; use ndarray::prelude::*;
use rand::Rng; use rand::Rng;
use std::collections::{BTreeSet};
use thiserror::Error;
use rand_chacha::ChaCha8Rng; use rand_chacha::ChaCha8Rng;
use thiserror::Error;
/// Error types for trait Params /// Error types for trait Params
#[derive(Error, Debug, PartialEq)] #[derive(Error, Debug, PartialEq)]
@ -35,11 +36,21 @@ pub trait ParamsTrait {
/// Randomly generate a residence time for the given node taking into account the node state /// Randomly generate a residence time for the given node taking into account the node state
/// and its parent set. /// and its parent set.
fn get_random_residence_time(&self, state: usize, u: usize, rng: &mut ChaCha8Rng) -> Result<f64, ParamsError>; fn get_random_residence_time(
&self,
state: usize,
u: usize,
rng: &mut ChaCha8Rng,
) -> Result<f64, ParamsError>;
/// Randomly generate a possible state for the given node taking into account the node state /// Randomly generate a possible state for the given node taking into account the node state
/// and its parent set. /// and its parent set.
fn get_random_state(&self, state: usize, u: usize, rng: &mut ChaCha8Rng) -> Result<StateType, ParamsError>; fn get_random_state(
&self,
state: usize,
u: usize,
rng: &mut ChaCha8Rng,
) -> Result<StateType, ParamsError>;
/// Used by childern of the node described by this parameters to reserve spaces in their CIMs. /// Used by childern of the node described by this parameters to reserve spaces in their CIMs.
fn get_reserved_space_as_parent(&self) -> usize; fn get_reserved_space_as_parent(&self) -> usize;
@ -102,7 +113,7 @@ impl DiscreteStatesContinousTimeParams {
///This function check if the cim is valid using the validate_params method. ///This function check if the cim is valid using the validate_params method.
///- **Valid cim inserted**: it substitute the CIM in self.cim and return Ok(()) ///- **Valid cim inserted**: it substitute the CIM in self.cim and return Ok(())
///- **Invalid cim inserted**: it replace the self.cim value with None and it retu ParamsError ///- **Invalid cim inserted**: it replace the self.cim value with None and it retu ParamsError
pub fn set_cim(&mut self, cim: Array3<f64>) -> Result<(), ParamsError>{ pub fn set_cim(&mut self, cim: Array3<f64>) -> Result<(), ParamsError> {
self.cim = Some(cim); self.cim = Some(cim);
match self.validate_params() { match self.validate_params() {
Ok(()) => Ok(()), Ok(()) => Ok(()),
@ -113,7 +124,6 @@ impl DiscreteStatesContinousTimeParams {
} }
} }
///Unchecked version of the setter function for CIM. ///Unchecked version of the setter function for CIM.
pub fn set_cim_unchecked(&mut self, cim: Array3<f64>) { pub fn set_cim_unchecked(&mut self, cim: Array3<f64>) {
self.cim = Some(cim); self.cim = Some(cim);
@ -124,7 +134,6 @@ impl DiscreteStatesContinousTimeParams {
&self.transitions &self.transitions
} }
///Setter function for transitions ///Setter function for transitions
pub fn set_transitions(&mut self, transitions: Array3<usize>) { pub fn set_transitions(&mut self, transitions: Array3<usize>) {
self.transitions = Some(transitions); self.transitions = Some(transitions);
@ -135,12 +144,10 @@ impl DiscreteStatesContinousTimeParams {
&self.residence_time &self.residence_time
} }
///Setter function for residence_time ///Setter function for residence_time
pub fn set_residence_time(&mut self, residence_time: Array2<f64>) { pub fn set_residence_time(&mut self, residence_time: Array2<f64>) {
self.residence_time = Some(residence_time); self.residence_time = Some(residence_time);
} }
} }
impl ParamsTrait for DiscreteStatesContinousTimeParams { impl ParamsTrait for DiscreteStatesContinousTimeParams {
@ -154,7 +161,12 @@ impl ParamsTrait for DiscreteStatesContinousTimeParams {
StateType::Discrete(rng.gen_range(0..(self.domain.len()))) StateType::Discrete(rng.gen_range(0..(self.domain.len())))
} }
fn get_random_residence_time(&self, state: usize, u: usize, rng: &mut ChaCha8Rng) -> Result<f64, ParamsError> { fn get_random_residence_time(
&self,
state: usize,
u: usize,
rng: &mut ChaCha8Rng,
) -> Result<f64, ParamsError> {
// Generate a random residence time given the current state of the node and its parent set. // Generate a random residence time given the current state of the node and its parent set.
// The method used is described in: // The method used is described in:
// https://en.wikipedia.org/wiki/Exponential_distribution#Generating_exponential_variates // https://en.wikipedia.org/wiki/Exponential_distribution#Generating_exponential_variates
@ -170,7 +182,12 @@ impl ParamsTrait for DiscreteStatesContinousTimeParams {
} }
} }
fn get_random_state(&self, state: usize, u: usize, rng: &mut ChaCha8Rng) -> Result<StateType, ParamsError> { fn get_random_state(
&self,
state: usize,
u: usize,
rng: &mut ChaCha8Rng,
) -> Result<StateType, ParamsError> {
// Generate a random transition given the current state of the node and its parent set. // Generate a random transition given the current state of the node and its parent set.
// The method used is described in: // The method used is described in:
// https://en.wikipedia.org/wiki/Multinomial_distribution#Sampling_from_a_multinomial_distribution // https://en.wikipedia.org/wiki/Multinomial_distribution#Sampling_from_a_multinomial_distribution
@ -246,7 +263,9 @@ impl ParamsTrait for DiscreteStatesContinousTimeParams {
} }
// Check if each row sum up to 0 // Check if each row sum up to 0
if cim.sum_axis(Axis(2)).iter() if cim
.sum_axis(Axis(2))
.iter()
.any(|x| f64::abs(x.clone()) > f64::EPSILON * 3.0) .any(|x| f64::abs(x.clone()) > f64::EPSILON * 3.0)
{ {
return Err(ParamsError::InvalidCIM(String::from( return Err(ParamsError::InvalidCIM(String::from(
@ -260,5 +279,4 @@ impl ParamsTrait for DiscreteStatesContinousTimeParams {
fn get_label(&self) -> &String { fn get_label(&self) -> &String {
&self.label &self.label
} }
} }

9
src/structure_learning.rs
Unescape View File

@ -1,12 +1,11 @@
pub mod score_function;
pub mod score_based_algorithm;
pub mod constraint_based_algorithm; pub mod constraint_based_algorithm;
pub mod hypothesis_test; pub mod hypothesis_test;
use crate::network; pub mod score_based_algorithm;
use crate::tools; pub mod score_function;
use crate::{network, tools};
pub trait StructureLearningAlgorithm { pub trait StructureLearningAlgorithm {
fn fit_transform<T, >(&self, net: T, dataset: &tools::Dataset) -> T fn fit_transform<T>(&self, net: T, dataset: &tools::Dataset) -> T
where where
T: network::Network; T: network::Network;
} }

2
src/structure_learning/constraint_based_algorithm.rs
Unescape View File

@ -1,5 +1,3 @@
//pub struct CTPC { //pub struct CTPC {
// //
//} //}

24
src/structure_learning/hypothesis_test.rs
Unescape View File

@ -1,11 +1,10 @@
use ndarray::Array3; use std::collections::BTreeSet;
use ndarray::Axis;
use ndarray::{Array3, Axis};
use statrs::distribution::{ChiSquared, ContinuousCDF}; use statrs::distribution::{ChiSquared, ContinuousCDF};
use crate::network;
use crate::parameter_learning;
use crate::params::*; use crate::params::*;
use std::collections::BTreeSet; use crate::{network, parameter_learning};
pub trait HypothesisTest { pub trait HypothesisTest {
fn call<T, P>( fn call<T, P>(
@ -110,15 +109,15 @@ impl HypothesisTest for ChiSquare {
// Prendo dalla cache l'apprendimento dei parametri, che sarebbe una CIM // Prendo dalla cache l'apprendimento dei parametri, che sarebbe una CIM
// di dimensione nxn // di dimensione nxn
// (CIM, M, T) // (CIM, M, T)
let P_small = match cache.fit(net, child_node, Some(separation_set.clone())){ let P_small = match cache.fit(net, child_node, Some(separation_set.clone())) {
Params::DiscreteStatesContinousTime(node) => node Params::DiscreteStatesContinousTime(node) => node,
}; };
// //
let mut extended_separation_set = separation_set.clone(); let mut extended_separation_set = separation_set.clone();
extended_separation_set.insert(parent_node); extended_separation_set.insert(parent_node);
let P_big = match cache.fit(net, child_node, Some(extended_separation_set.clone())){ let P_big = match cache.fit(net, child_node, Some(extended_separation_set.clone())) {
Params::DiscreteStatesContinousTime(node) => node Params::DiscreteStatesContinousTime(node) => node,
}; };
// Commentare qui // Commentare qui
let partial_cardinality_product: usize = extended_separation_set let partial_cardinality_product: usize = extended_separation_set
@ -132,7 +131,12 @@ impl HypothesisTest for ChiSquare {
/ (partial_cardinality_product / (partial_cardinality_product
* net.get_node(parent_node).get_reserved_space_as_parent())) * net.get_node(parent_node).get_reserved_space_as_parent()))
* partial_cardinality_product; * partial_cardinality_product;
if !self.compare_matrices(idx_M_small, P_small.get_transitions().as_ref().unwrap(), idx_M_big, P_big.get_transitions().as_ref().unwrap()) { if !self.compare_matrices(
idx_M_small,
P_small.get_transitions().as_ref().unwrap(),
idx_M_big,
P_big.get_transitions().as_ref().unwrap(),
) {
return false; return false;
} }
} }

6
src/structure_learning/score_based_algorithm.rs
Unescape View File

@ -1,8 +1,8 @@
use crate::network; use std::collections::BTreeSet;
use crate::structure_learning::score_function::ScoreFunction; use crate::structure_learning::score_function::ScoreFunction;
use crate::structure_learning::StructureLearningAlgorithm; use crate::structure_learning::StructureLearningAlgorithm;
use crate::tools; use crate::{network, tools};
use std::collections::BTreeSet;
pub struct HillClimbing<S: ScoreFunction> { pub struct HillClimbing<S: ScoreFunction> {
score_function: S, score_function: S,

54
src/structure_learning/score_function.rs
Unescape View File

@ -1,10 +1,9 @@
use crate::network; use std::collections::BTreeSet;
use crate::parameter_learning;
use crate::params;
use crate::tools;
use ndarray::prelude::*; use ndarray::prelude::*;
use statrs::function::gamma; use statrs::function::gamma;
use std::collections::BTreeSet;
use crate::{network, parameter_learning, params, tools};
pub trait ScoreFunction { pub trait ScoreFunction {
fn call<T>( fn call<T>(
@ -25,7 +24,6 @@ pub struct LogLikelihood {
impl LogLikelihood { impl LogLikelihood {
pub fn new(alpha: usize, tau: f64) -> LogLikelihood { pub fn new(alpha: usize, tau: f64) -> LogLikelihood {
//Tau must be >=0.0 //Tau must be >=0.0
if tau < 0.0 { if tau < 0.0 {
panic!("tau must be >=0.0"); panic!("tau must be >=0.0");
@ -44,7 +42,7 @@ impl LogLikelihood {
T: network::Network, T: network::Network,
{ {
//Identify the type of node used //Identify the type of node used
match &net.get_node(node){ match &net.get_node(node) {
params::Params::DiscreteStatesContinousTime(_params) => { params::Params::DiscreteStatesContinousTime(_params) => {
//Compute the sufficient statistics M (number of transistions) and T (residence //Compute the sufficient statistics M (number of transistions) and T (residence
//time) //time)
@ -57,33 +55,38 @@ impl LogLikelihood {
let tau = self.tau / M.shape()[0] as f64; let tau = self.tau / M.shape()[0] as f64;
//Compute the log likelihood for q //Compute the log likelihood for q
let log_ll_q:f64 = M let log_ll_q: f64 = M
.sum_axis(Axis(2)) .sum_axis(Axis(2))
.iter() .iter()
.zip(T.iter()) .zip(T.iter())
.map(|(m, t)| { .map(|(m, t)| {
gamma::ln_gamma(alpha + *m as f64 + 1.0) gamma::ln_gamma(alpha + *m as f64 + 1.0) + (alpha + 1.0) * f64::ln(tau)
+ (alpha + 1.0) * f64::ln(tau)
- gamma::ln_gamma(alpha + 1.0) - gamma::ln_gamma(alpha + 1.0)
- (alpha + *m as f64 + 1.0) * f64::ln(tau + t) - (alpha + *m as f64 + 1.0) * f64::ln(tau + t)
}) })
.sum(); .sum();
//Compute the log likelihood for theta //Compute the log likelihood for theta
let log_ll_theta: f64 = M.outer_iter() let log_ll_theta: f64 = M
.map(|x| x.outer_iter() .outer_iter()
.map(|y| gamma::ln_gamma(alpha) .map(|x| {
- gamma::ln_gamma(alpha + y.sum() as f64) x.outer_iter()
+ y.iter().map(|z| .map(|y| {
gamma::ln_gamma(alpha) - gamma::ln_gamma(alpha + y.sum() as f64)
+ y.iter()
.map(|z| {
gamma::ln_gamma(alpha + *z as f64) gamma::ln_gamma(alpha + *z as f64)
- gamma::ln_gamma(alpha)).sum::<f64>()).sum::<f64>()).sum(); - gamma::ln_gamma(alpha)
})
.sum::<f64>()
})
.sum::<f64>()
})
.sum();
(log_ll_theta + log_ll_q, M) (log_ll_theta + log_ll_q, M)
} }
} }
} }
} }
impl ScoreFunction for LogLikelihood { impl ScoreFunction for LogLikelihood {
@ -102,13 +105,13 @@ impl ScoreFunction for LogLikelihood {
} }
pub struct BIC { pub struct BIC {
ll: LogLikelihood ll: LogLikelihood,
} }
impl BIC { impl BIC {
pub fn new(alpha: usize, tau: f64) -> BIC { pub fn new(alpha: usize, tau: f64) -> BIC {
BIC { BIC {
ll: LogLikelihood::new(alpha, tau) ll: LogLikelihood::new(alpha, tau),
} }
} }
} }
@ -122,14 +125,19 @@ impl ScoreFunction for BIC {
dataset: &tools::Dataset, dataset: &tools::Dataset,
) -> f64 ) -> f64
where where
T: network::Network { T: network::Network,
{
//Compute the log-likelihood //Compute the log-likelihood
let (ll, M) = self.ll.compute_score(net, node, parent_set, dataset); let (ll, M) = self.ll.compute_score(net, node, parent_set, dataset);
//Compute the number of parameters //Compute the number of parameters
let n_parameters = M.shape()[0] * M.shape()[1] * (M.shape()[2] - 1); let n_parameters = M.shape()[0] * M.shape()[1] * (M.shape()[2] - 1);
//TODO: Optimize this //TODO: Optimize this
//Compute the sample size //Compute the sample size
let sample_size: usize = dataset.get_trajectories().iter().map(|x| x.get_time().len() - 1).sum(); let sample_size: usize = dataset
.get_trajectories()
.iter()
.map(|x| x.get_time().len() - 1)
.sum();
//Compute BIC //Compute BIC
ll - f64::ln(sample_size as f64) / 2.0 * n_parameters as f64 ll - f64::ln(sample_size as f64) / 2.0 * n_parameters as f64
} }

13
src/tools.rs
Unescape View File

@ -1,10 +1,10 @@
use crate::network;
use crate::params;
use crate::params::ParamsTrait;
use ndarray::prelude::*; use ndarray::prelude::*;
use rand_chacha::rand_core::SeedableRng; use rand_chacha::rand_core::SeedableRng;
use rand_chacha::ChaCha8Rng; use rand_chacha::ChaCha8Rng;
use crate::params::ParamsTrait;
use crate::{network, params};
pub struct Trajectory { pub struct Trajectory {
time: Array1<f64>, time: Array1<f64>,
events: Array2<usize>, events: Array2<usize>,
@ -35,7 +35,6 @@ pub struct Dataset {
impl Dataset { impl Dataset {
pub fn new(trajectories: Vec<Trajectory>) -> Dataset { pub fn new(trajectories: Vec<Trajectory>) -> Dataset {
//All the trajectories in the same dataset must represent the same process. For this reason //All the trajectories in the same dataset must represent the same process. For this reason
//each trajectory must represent the same number of variables. //each trajectory must represent the same number of variables.
if trajectories if trajectories
@ -58,7 +57,6 @@ pub fn trajectory_generator<T: network::Network>(
t_end: f64, t_end: f64,
seed: Option<u64>, seed: Option<u64>,
) -> Dataset { ) -> Dataset {
//Tmp growing vector containing generated trajectories. //Tmp growing vector containing generated trajectories.
let mut trajectories: Vec<Trajectory> = Vec::new(); let mut trajectories: Vec<Trajectory> = Vec::new();
@ -67,7 +65,7 @@ pub fn trajectory_generator<T: network::Network>(
//If a seed is present use it to initialize the random generator. //If a seed is present use it to initialize the random generator.
Some(seed) => SeedableRng::seed_from_u64(seed), Some(seed) => SeedableRng::seed_from_u64(seed),
//Otherwise create a new random generator using the method `from_entropy` //Otherwise create a new random generator using the method `from_entropy`
None => SeedableRng::from_entropy() None => SeedableRng::from_entropy(),
}; };
//Each iteration generate one trajectory //Each iteration generate one trajectory
@ -78,7 +76,8 @@ pub fn trajectory_generator<T: network::Network>(
let mut time: Vec<f64> = Vec::new(); let mut time: Vec<f64> = Vec::new();
//Configuration of the process variables at time t initialized with an uniform //Configuration of the process variables at time t initialized with an uniform
//distribution. //distribution.
let mut current_state: Vec<params::StateType> = net.get_node_indices() let mut current_state: Vec<params::StateType> = net
.get_node_indices()
.map(|x| net.get_node(x).get_random_state_uniform(&mut rng)) .map(|x| net.get_node(x).get_random_state_uniform(&mut rng))
.collect(); .collect();
//History of all the configurations of the process variables. //History of all the configurations of the process variables.

3
tests/ctbn.rs
Unescape View File

@ -1,8 +1,9 @@
mod utils; mod utils;
use std::collections::BTreeSet;
use reCTBN::ctbn::*; use reCTBN::ctbn::*;
use reCTBN::network::Network; use reCTBN::network::Network;
use reCTBN::params::{self, ParamsTrait}; use reCTBN::params::{self, ParamsTrait};
use std::collections::BTreeSet;
use utils::generate_discrete_time_continous_node; use utils::generate_discrete_time_continous_node;
#[test] #[test]

20
tests/parameter_learning.rs
Unescape View File

@ -1,13 +1,13 @@
#![allow(non_snake_case)] #![allow(non_snake_case)]
mod utils; mod utils;
use utils::*;
use ndarray::arr3; use ndarray::arr3;
use reCTBN::ctbn::*; use reCTBN::ctbn::*;
use reCTBN::network::Network; use reCTBN::network::Network;
use reCTBN::parameter_learning::*; use reCTBN::parameter_learning::*;
use reCTBN::{params, tools::*}; use reCTBN::params;
use reCTBN::tools::*;
use utils::*;
extern crate approx; extern crate approx;
@ -41,7 +41,7 @@ fn learn_binary_cim<T: ParameterLearning>(pl: T) {
let data = trajectory_generator(&net, 100, 100.0, Some(6347747169756259)); let data = trajectory_generator(&net, 100, 100.0, Some(6347747169756259));
let p = match pl.fit(&net, &data, 1, None) { let p = match pl.fit(&net, &data, 1, None) {
params::Params::DiscreteStatesContinousTime(p) => p params::Params::DiscreteStatesContinousTime(p) => p,
}; };
assert_eq!(p.get_cim().as_ref().unwrap().shape(), [2, 2, 2]); assert_eq!(p.get_cim().as_ref().unwrap().shape(), [2, 2, 2]);
assert!(p.get_cim().as_ref().unwrap().abs_diff_eq( assert!(p.get_cim().as_ref().unwrap().abs_diff_eq(
@ -99,8 +99,8 @@ fn learn_ternary_cim<T: ParameterLearning>(pl: T) {
} }
let data = trajectory_generator(&net, 100, 200.0, Some(6347747169756259)); let data = trajectory_generator(&net, 100, 200.0, Some(6347747169756259));
let p = match pl.fit(&net, &data, 1, None){ let p = match pl.fit(&net, &data, 1, None) {
params::Params::DiscreteStatesContinousTime(p) => p params::Params::DiscreteStatesContinousTime(p) => p,
}; };
assert_eq!(p.get_cim().as_ref().unwrap().shape(), [3, 3, 3]); assert_eq!(p.get_cim().as_ref().unwrap().shape(), [3, 3, 3]);
assert!(p.get_cim().as_ref().unwrap().abs_diff_eq( assert!(p.get_cim().as_ref().unwrap().abs_diff_eq(
@ -162,8 +162,8 @@ fn learn_ternary_cim_no_parents<T: ParameterLearning>(pl: T) {
} }
let data = trajectory_generator(&net, 100, 200.0, Some(6347747169756259)); let data = trajectory_generator(&net, 100, 200.0, Some(6347747169756259));
let p = match pl.fit(&net, &data, 0, None){ let p = match pl.fit(&net, &data, 0, None) {
params::Params::DiscreteStatesContinousTime(p) => p params::Params::DiscreteStatesContinousTime(p) => p,
}; };
assert_eq!(p.get_cim().as_ref().unwrap().shape(), [1, 3, 3]); assert_eq!(p.get_cim().as_ref().unwrap().shape(), [1, 3, 3]);
assert!(p.get_cim().as_ref().unwrap().abs_diff_eq( assert!(p.get_cim().as_ref().unwrap().abs_diff_eq(
@ -291,8 +291,8 @@ fn learn_mixed_discrete_cim<T: ParameterLearning>(pl: T) {
} }
let data = trajectory_generator(&net, 300, 300.0, Some(6347747169756259)); let data = trajectory_generator(&net, 300, 300.0, Some(6347747169756259));
let p = match pl.fit(&net, &data, 2, None){ let p = match pl.fit(&net, &data, 2, None) {
params::Params::DiscreteStatesContinousTime(p) => p params::Params::DiscreteStatesContinousTime(p) => p,
}; };
assert_eq!(p.get_cim().as_ref().unwrap().shape(), [9, 4, 4]); assert_eq!(p.get_cim().as_ref().unwrap().shape(), [9, 4, 4]);
assert!(p.get_cim().as_ref().unwrap().abs_diff_eq( assert!(p.get_cim().as_ref().unwrap().abs_diff_eq(

3
tests/params.rs
Unescape View File

@ -1,5 +1,6 @@
use ndarray::prelude::*; use ndarray::prelude::*;
use rand_chacha::{rand_core::SeedableRng, ChaCha8Rng}; use rand_chacha::rand_core::SeedableRng;
use rand_chacha::ChaCha8Rng;
use reCTBN::params::{ParamsTrait, *}; use reCTBN::params::{ParamsTrait, *};
mod utils; mod utils;

75
tests/structure_learning.rs
Unescape View File

@ -1,17 +1,18 @@
#![allow(non_snake_case)] #![allow(non_snake_case)]
mod utils; mod utils;
use utils::*; use std::collections::BTreeSet;
use ndarray::{arr1, arr2, arr3}; use ndarray::{arr1, arr2, arr3};
use reCTBN::ctbn::*; use reCTBN::ctbn::*;
use reCTBN::network::Network; use reCTBN::network::Network;
use reCTBN::params; use reCTBN::params;
use reCTBN::structure_learning::score_function::*;
use reCTBN::structure_learning::{score_based_algorithm::*, StructureLearningAlgorithm};
use reCTBN::structure_learning::hypothesis_test::*; use reCTBN::structure_learning::hypothesis_test::*;
use reCTBN::structure_learning::score_based_algorithm::*;
use reCTBN::structure_learning::score_function::*;
use reCTBN::structure_learning::StructureLearningAlgorithm;
use reCTBN::tools::*; use reCTBN::tools::*;
use std::collections::BTreeSet; use utils::*;
#[macro_use] #[macro_use]
extern crate approx; extern crate approx;
@ -320,73 +321,43 @@ pub fn learn_mixed_discrete_net_3_nodes_hill_climbing_bic_1_parent_constraint()
} }
#[test] #[test]
pub fn chi_square_compare_matrices () { pub fn chi_square_compare_matrices() {
let i: usize = 1; let i: usize = 1;
let M1 = arr3(&[ let M1 = arr3(&[
[[ 0, 2, 3], [[0, 2, 3], [4, 0, 6], [7, 8, 0]],
[ 4, 0, 6], [[0, 12, 90], [3, 0, 40], [6, 40, 0]],
[ 7, 8, 0]], [[0, 2, 3], [4, 0, 6], [44, 66, 0]],
[[0, 12, 90],
[ 3, 0, 40],
[ 6, 40, 0]],
[[ 0, 2, 3],
[ 4, 0, 6],
[ 44, 66, 0]]
]); ]);
let j: usize = 0; let j: usize = 0;
let M2 = arr3(&[ let M2 = arr3(&[[[0, 200, 300], [400, 0, 600], [700, 800, 0]]]);
[[ 0, 200, 300],
[ 400, 0, 600],
[ 700, 800, 0]]
]);
let chi_sq = ChiSquare::new(0.1); let chi_sq = ChiSquare::new(0.1);
assert!(!chi_sq.compare_matrices( i, &M1, j, &M2)); assert!(!chi_sq.compare_matrices(i, &M1, j, &M2));
} }
#[test] #[test]
pub fn chi_square_compare_matrices_2 () { pub fn chi_square_compare_matrices_2() {
let i: usize = 1; let i: usize = 1;
let M1 = arr3(&[ let M1 = arr3(&[
[[ 0, 2, 3], [[0, 2, 3], [4, 0, 6], [7, 8, 0]],
[ 4, 0, 6], [[0, 20, 30], [40, 0, 60], [70, 80, 0]],
[ 7, 8, 0]], [[0, 2, 3], [4, 0, 6], [44, 66, 0]],
[[0, 20, 30],
[ 40, 0, 60],
[ 70, 80, 0]],
[[ 0, 2, 3],
[ 4, 0, 6],
[ 44, 66, 0]]
]); ]);
let j: usize = 0; let j: usize = 0;
let M2 = arr3(&[ let M2 = arr3(&[[[0, 200, 300], [400, 0, 600], [700, 800, 0]]]);
[[ 0, 200, 300],
[ 400, 0, 600],
[ 700, 800, 0]]
]);
let chi_sq = ChiSquare::new(0.1); let chi_sq = ChiSquare::new(0.1);
assert!(chi_sq.compare_matrices( i, &M1, j, &M2)); assert!(chi_sq.compare_matrices(i, &M1, j, &M2));
} }
#[test] #[test]
pub fn chi_square_compare_matrices_3 () { pub fn chi_square_compare_matrices_3() {
let i: usize = 1; let i: usize = 1;
let M1 = arr3(&[ let M1 = arr3(&[
[[ 0, 2, 3], [[0, 2, 3], [4, 0, 6], [7, 8, 0]],
[ 4, 0, 6], [[0, 21, 31], [41, 0, 59], [71, 79, 0]],
[ 7, 8, 0]], [[0, 2, 3], [4, 0, 6], [44, 66, 0]],
[[0, 21, 31],
[ 41, 0, 59],
[ 71, 79, 0]],
[[ 0, 2, 3],
[ 4, 0, 6],
[ 44, 66, 0]]
]); ]);
let j: usize = 0; let j: usize = 0;
let M2 = arr3(&[ let M2 = arr3(&[[[0, 200, 300], [400, 0, 600], [700, 800, 0]]]);
[[ 0, 200, 300],
[ 400, 0, 600],
[ 700, 800, 0]]
]);
let chi_sq = ChiSquare::new(0.1); let chi_sq = ChiSquare::new(0.1);
assert!(chi_sq.compare_matrices( i, &M1, j, &M2)); assert!(chi_sq.compare_matrices(i, &M1, j, &M2));
} }

13
tests/utils.rs
Unescape View File

@ -1,12 +1,19 @@
use reCTBN::params;
use std::collections::BTreeSet; use std::collections::BTreeSet;
use reCTBN::params;
#[allow(dead_code)] #[allow(dead_code)]
pub fn generate_discrete_time_continous_node(label: String, cardinality: usize) -> params::Params { pub fn generate_discrete_time_continous_node(label: String, cardinality: usize) -> params::Params {
params::Params::DiscreteStatesContinousTime(generate_discrete_time_continous_params(label, cardinality)) params::Params::DiscreteStatesContinousTime(generate_discrete_time_continous_params(
label,
cardinality,
))
} }
pub fn generate_discrete_time_continous_params(label: String, cardinality: usize) -> params::DiscreteStatesContinousTimeParams{ pub fn generate_discrete_time_continous_params(
label: String,
cardinality: usize,
) -> params::DiscreteStatesContinousTimeParams {
let domain: BTreeSet<String> = (0..cardinality).map(|x| x.to_string()).collect(); let domain: BTreeSet<String> = (0..cardinality).map(|x| x.to_string()).collect();
params::DiscreteStatesContinousTimeParams::new(label, domain) params::DiscreteStatesContinousTimeParams::new(label, domain)
} }

Write Preview
Loading…
Cancel
Save