fix format & revert parameteredIndex

src/lib.rs

@@ -9,11 +9,11 @@ mod error;
 mod inference;
 mod model;
 mod optimal_retention;
+mod parameter_clipper;
 mod pre_training;
 #[cfg(test)]
 mod test_helpers;
 mod training;
-mod parameter_clipper;
 
 pub use dataset::{FSRSItem, FSRSReview};
 pub use error::{FSRSError, Result};

src/optimal_retention.rs

@@ -7,7 +7,7 @@ use ndarray::{s, Array1, Array2, Ix0, Ix1, SliceInfoElem, Zip};
 use ndarray_rand::rand_distr::Distribution;
 use ndarray_rand::RandomExt;
 use rand::{
-    distributions::{Uniform, parameteredIndex},
+    distributions::{Uniform, WeightedIndex},
     rngs::StdRng,
     SeedableRng,
 };
@@ -148,10 +148,10 @@ pub fn simulate(
     let mut cost_per_day = Array1::zeros(learn_span);
 
     let first_rating_choices = [1, 2, 3, 4];
-    let first_rating_dist = parameteredIndex::new(first_rating_prob).unwrap();
+    let first_rating_dist = WeightedIndex::new(first_rating_prob).unwrap();
 
     let review_rating_choices = [2, 3, 4];
-    let review_rating_dist = parameteredIndex::new(review_rating_prob).unwrap();
+    let review_rating_dist = WeightedIndex::new(review_rating_prob).unwrap();
 
     let mut rng = StdRng::seed_from_u64(seed.unwrap_or(42));
 

src/training.rs

@@ -3,8 +3,8 @@ use crate::cosine_annealing::CosineAnnealingLR;
 use crate::dataset::{split_filter_data, FSRSBatcher, FSRSDataset, FSRSItem};
 use crate::error::Result;
 use crate::model::{Model, ModelConfig};
-use crate::pre_training::pretrain;
 use crate::parameter_clipper::parameter_clipper;
+use crate::pre_training::pretrain;
 use crate::{FSRSError, DEFAULT_PARAMETERS, FSRS};
 use burn::backend::Autodiff;