✨ NamedTuple for cache and less documentation redundancy

src/encoders/contrast_encoder/contrast_encoder.jl

@@ -134,10 +134,9 @@ function contrast_encoder_fit(
         X, features; ignore = ignore, ordered_factor = ordered_factor,
         feature_mapper = feature_mapper,
     )
-
-    cache = Dict(
-        :vector_given_value_given_feature  => vector_given_value_given_feature,
-        :encoded_features => encoded_features,
+    cache = (
+        vector_given_value_given_feature = vector_given_value_given_feature,
+        encoded_features = encoded_features,
     )
 
     return cache
@@ -157,7 +156,7 @@ Use a fitted contrast encoder to encode the levels of selected categorical varia
 
   - `X_tr`: The table with selected features after the selected features are encoded by contrast encoding.
 """
-function contrast_encoder_transform(X, cache::Dict)
-    vector_given_value_given_feature = cache[:vector_given_value_given_feature]
+function contrast_encoder_transform(X, cache::NamedTuple)
+    vector_given_value_given_feature = cache.vector_given_value_given_feature
     return generic_transform(X, vector_given_value_given_feature, single_feat = false; use_levelnames = true)
 end

src/encoders/contrast_encoder/interface_mlj.jl

@@ -36,19 +36,18 @@ function MMI.fit(transformer::ContrastEncoder, verbosity::Int, X)
         buildmatrix = transformer.buildmatrix,
         ordered_factor = transformer.ordered_factor,
     )
-    fitresult = generic_cache[:vector_given_value_given_feature]
+    fitresult = generic_cache.vector_given_value_given_feature
 
-    report = (encoded_features = generic_cache[:encoded_features],)        # report only has list of encoded features
+    report = (encoded_features = generic_cache.encoded_features,)        # report only has list of encoded features
     cache = nothing
     return fitresult, cache, report
 end;
 
 
 # 6. Transform method
 function MMI.transform(transformer::ContrastEncoder, fitresult, Xnew)
-    generic_cache = Dict(
-        :vector_given_value_given_feature =>
-            fitresult,
+    generic_cache = (
+        vector_given_value_given_feature = fitresult,
     )
     Xnew_transf = contrast_encoder_transform(Xnew, generic_cache)
     return Xnew_transf

src/encoders/frequency_encoding/frequency_encoding.jl

@@ -39,11 +39,11 @@ function frequency_encoder_fit(
     # 2. Pass it to generic_fit
     statistic_given_feat_val, encoded_features = generic_fit(
         X, features; ignore = ignore, ordered_factor = ordered_factor,
-        feature_mapper = feature_mapper,
-    )
-    cache = Dict(
-        :statistic_given_feat_val => statistic_given_feat_val,
-        :encoded_features => encoded_features,
+        feature_mapper = feature_mapper,)
+        
+    cache = (
+      statistic_given_feat_val = statistic_given_feat_val,
+      encoded_features = encoded_features,
     )
     return cache
 end
@@ -62,7 +62,7 @@ Encode the levels of a categorical variable in a given table with their (normali
 
   - `X_tr`: The table with selected features after the selected features are encoded by frequency encoding.
 """
-function frequency_encoder_transform(X, cache::Dict)
-    statistic_given_feat_val = cache[:statistic_given_feat_val]
+function frequency_encoder_transform(X, cache::NamedTuple)
+    statistic_given_feat_val = cache.statistic_given_feat_val
     return generic_transform(X, statistic_given_feat_val)
 end

src/encoders/frequency_encoding/interface_mlj.jl

@@ -36,19 +36,18 @@ function MMI.fit(transformer::FrequencyEncoder, verbosity::Int, X)
         normalize = transformer.normalize,
         output_type = transformer.output_type,
     )
-    fitresult = generic_cache[:statistic_given_feat_val]
+    fitresult = generic_cache.statistic_given_feat_val
 
-    report = (encoded_features = generic_cache[:encoded_features],)        # report only has list of encoded features
+    report = (encoded_features = generic_cache.encoded_features,)        # report only has list of encoded features
     cache = nothing
     return fitresult, cache, report
 end;
 
 
 # 6. Transform method
 function MMI.transform(transformer::FrequencyEncoder, fitresult, Xnew)
-    generic_cache = Dict(
-        :statistic_given_feat_val =>
-            fitresult,
+    generic_cache = (
+        statistic_given_feat_val = fitresult,
     )
     Xnew_transf = frequency_encoder_transform(Xnew, generic_cache)
     return Xnew_transf

src/encoders/missingness_encoding/interface_mlj.jl

@@ -39,19 +39,18 @@ function MMI.fit(transformer::MissingnessEncoder, verbosity::Int, X)
         ordered_factor = transformer.ordered_factor,
         label_for_missing = transformer.label_for_missing,
     )
-    fitresult = generic_cache[:label_for_missing_given_feature]
+    fitresult = generic_cache.label_for_missing_given_feature
 
-    report = (encoded_features = generic_cache[:encoded_features],)        # report only has list of encoded features
+    report = (encoded_features = generic_cache.encoded_features,)        # report only has list of encoded features
     cache = nothing
     return fitresult, cache, report
 end;
 
 
 # 6. Transform method
 function MMI.transform(transformer::MissingnessEncoder, fitresult, Xnew)
-    generic_cache = Dict(
-        :label_for_missing_given_feature =>
-            fitresult,
+    generic_cache = (
+        label_for_missing_given_feature = fitresult,
     )
     Xnew_transf = missingness_encoder_transform(Xnew, generic_cache)
     return Xnew_transf

src/encoders/missingness_encoding/missingness_encoding.jl

@@ -94,9 +94,9 @@ function missingness_encoder_fit(
         X, features; ignore = ignore, ordered_factor = ordered_factor,
         feature_mapper = feature_mapper,
     )
-    cache = Dict(
-        :label_for_missing_given_feature => label_for_missing_given_feature,
-        :encoded_features => encoded_features,
+    cache = (
+        label_for_missing_given_feature = label_for_missing_given_feature,
+        encoded_features = encoded_features,
     )
     return cache
 end
@@ -116,8 +116,8 @@ Apply a fitted missingness encoder to a table given the output of `missingness_e
 
   - `X_tr`: The table with selected features after the selected features are transformed by missingness encoder
 """
-function missingness_encoder_transform(X, cache::Dict)
-    label_for_missing_given_feature = cache[:label_for_missing_given_feature]
+function missingness_encoder_transform(X, cache::NamedTuple)
+    label_for_missing_given_feature = cache.label_for_missing_given_feature
     return generic_transform(
         X,
         label_for_missing_given_feature;

src/encoders/ordinal_encoding/interface_mlj.jl

@@ -34,18 +34,16 @@ function MMI.fit(transformer::OrdinalEncoder, verbosity::Int, X)
         output_type = transformer.output_type,
     )
     fitresult =
-        generic_cache[:index_given_feat_level]
-    report = (encoded_features = generic_cache[:encoded_features],)        # report only has list of encoded features
+        generic_cache.index_given_feat_level
+    report = (encoded_features = generic_cache.encoded_features,)        # report only has list of encoded features
     cache = nothing
     return fitresult, cache, report
 end;
 
 
 # 6. Transform method
 function MMI.transform(transformer::OrdinalEncoder, fitresult, Xnew)
-    generic_cache = Dict(
-        :index_given_feat_level => fitresult,
-    )
+    generic_cache = (index_given_feat_level = fitresult,)
     Xnew_transf = ordinal_encoder_transform(Xnew, generic_cache)
     return Xnew_transf
 end

src/encoders/ordinal_encoding/ordinal_encoding.jl

@@ -11,6 +11,7 @@ Fit an encoder to encode the levels of categorical variables in a given table as
   - `ignore=true`: Whether to exclude or includes the features given in `features`
   - `ordered_factor=false`: Whether to encode `OrderedFactor` or ignore them
   - `dtype`: The numerical concrete type of the encoded features. Default is `Float32`.
+
 # Returns (in a dict)
 
   - `index_given_feat_level`: Maps each level for each column in a subset of the categorical features of X into an integer.
@@ -27,18 +28,19 @@ function ordinal_encoder_fit(
     function feature_mapper(col, name)
         feat_levels = levels(col)
         index_given_feat_val =
-            Dict{eltype(feat_levels), output_type}(value => index for (index, value) in enumerate(feat_levels))
+            Dict{eltype(feat_levels), output_type}(
+                value => index for (index, value) in enumerate(feat_levels)
+            )
         return index_given_feat_val
     end
 
     # 2. Pass it to generic_fit
     index_given_feat_level, encoded_features = generic_fit(
         X, features; ignore = ignore, ordered_factor = ordered_factor,
-        feature_mapper = feature_mapper,
-    )
-    cache = Dict(
-        :index_given_feat_level => index_given_feat_level,
-        :encoded_features => encoded_features,
+        feature_mapper = feature_mapper,)
+    cache = (
+      index_given_feat_level = index_given_feat_level,
+      encoded_features = encoded_features,
     )
     return cache
 end
@@ -58,7 +60,7 @@ Encode the levels of a categorical variable in a given table as integers.
 
   - `X_tr`: The table with selected features after the selected features are encoded by ordinal encoding.
 """
-function ordinal_encoder_transform(X, cache::Dict)
-    index_given_feat_level = cache[:index_given_feat_level]
+function ordinal_encoder_transform(X, cache::NamedTuple)
+    index_given_feat_level = cache.index_given_feat_level
     return generic_transform(X, index_given_feat_level)
 end

src/encoders/target_encoding/interface_mlj.jl

@@ -52,7 +52,7 @@ struct TargetEncoderResult{
     task::S            # "Regression", "Classification" 
     num_classes::I     # num_classes in case of classification
     y_classes::A      # y_classes in case of classification
-    
+
 end
 
 
@@ -75,25 +75,24 @@ function MMI.fit(transformer::TargetEncoder, verbosity::Int, X, y)
         m = transformer.m,
     )
     fitresult = TargetEncoderResult(
-        generic_cache[:y_stat_given_feat_level],
-        generic_cache[:task],
-        generic_cache[:num_classes],
-        generic_cache[:y_classes],
+        generic_cache.y_stat_given_feat_level,
+        generic_cache.task,
+        generic_cache.num_classes,
+        generic_cache.y_classes,
     )
-    report = (encoded_features = generic_cache[:encoded_features],)        # report only has list of encoded features
+    report = (encoded_features = generic_cache.encoded_features,)        # report only has list of encoded features
     cache = nothing
     return fitresult, cache, report
 end;
 
 
 # 7. Transform method
 function MMI.transform(transformer::TargetEncoder, fitresult, Xnew)
-    generic_cache = Dict(
-        :y_stat_given_feat_level =>
-            fitresult.y_stat_given_feat_level,
-        :num_classes => fitresult.num_classes,
-        :task => fitresult.task,
-        :y_classes => fitresult.y_classes,
+    generic_cache = (
+        y_stat_given_feat_level = fitresult.y_stat_given_feat_level,
+        num_classes = fitresult.num_classes,
+        task = fitresult.task,
+        y_classes = fitresult.y_classes,
     )
     Xnew_transf = target_encoder_transform(Xnew, generic_cache)
     return Xnew_transf

src/encoders/target_encoding/target_encoding.jl

@@ -211,12 +211,12 @@ function target_encoder_fit(
         feature_mapper = feature_mapper,
     )
 
-    cache = Dict(
-        :task => task,
-        :num_classes => (task == "Regression") ? -1 : length(y_classes),
-        :y_stat_given_feat_level => y_stat_given_feat_level,
-        :encoded_features => encoded_features,
-        :y_classes => (task == "Regression") ? nothing : y_classes,
+    cache = (
+        task = task,
+        num_classes = (task == "Regression") ? -1 : length(y_classes),
+        y_stat_given_feat_level = y_stat_given_feat_level,
+        encoded_features = encoded_features,
+        y_classes = (task == "Regression") ? nothing : y_classes,
     )
     return cache
 end
@@ -242,16 +242,16 @@ every categorical feature as well as other metadata needed for transform
 """
 
 function target_encoder_transform(X, cache)
-    task = cache[:task]
-    y_stat_given_feat_level = cache[:y_stat_given_feat_level]
-    num_classes = cache[:num_classes]
-    y_classes = cache[:y_classes]
+    task = cache.task
+    y_stat_given_feat_level = cache.y_stat_given_feat_level
+    num_classes = cache.num_classes
+    y_classes = cache.y_classes
 
     return generic_transform(
         X,
         y_stat_given_feat_level;
         single_feat = task == "Regression" || (task == "Classification" && num_classes < 3),
         use_levelnames = true,
-        custom_levels = y_classes,)
+        custom_levels = y_classes)
 end
 

src/transformers/cardinality_reducer/cardinality_reducer.jl

@@ -105,9 +105,9 @@ function cardinality_reducer_fit(
         X, features; ignore = ignore, ordered_factor = ordered_factor,
         feature_mapper = feature_mapper,
     )
-    cache = Dict(
-        :new_cat_given_col_val => new_cat_given_col_val,
-        :encoded_features => encoded_features,
+    cache = (
+        new_cat_given_col_val = new_cat_given_col_val,
+        encoded_features = encoded_features,
     )
     return cache
 end
@@ -128,7 +128,12 @@ Apply a fitted cardinality reducer to a table given the output of `cardinality_r
 
   - `X_tr`: The table with selected features after the selected features are transformed by cardinality reducer
 """
-function cardinality_reducer_transform(X, cache::Dict)
-    new_cat_given_col_val = cache[:new_cat_given_col_val]
-    return generic_transform(X, new_cat_given_col_val; ignore_unknown = true, ensure_categorical = true)
+function cardinality_reducer_transform(X, cache::NamedTuple)
+    new_cat_given_col_val = cache.new_cat_given_col_val
+    return generic_transform(
+        X,
+        new_cat_given_col_val;
+        ignore_unknown = true,
+        ensure_categorical = true,
+    )
 end

src/transformers/cardinality_reducer/interface_mlj.jl

@@ -24,7 +24,13 @@ function CardinalityReducer(;
         Char => 'O',
     ),
 )
-    return CardinalityReducer(features, ignore, ordered_factor, min_frequency, label_for_infrequent)
+    return CardinalityReducer(
+        features,
+        ignore,
+        ordered_factor,
+        min_frequency,
+        label_for_infrequent,
+    )
 end;
 
 
@@ -43,19 +49,18 @@ function MMI.fit(transformer::CardinalityReducer, verbosity::Int, X)
         min_frequency = transformer.min_frequency,
         label_for_infrequent = transformer.label_for_infrequent,
     )
-    fitresult = generic_cache[:new_cat_given_col_val]
+    fitresult = generic_cache.new_cat_given_col_val
 
-    report = (encoded_features = generic_cache[:encoded_features],)        # report only has list of encoded features
+    report = (encoded_features = generic_cache.encoded_features,)        # report only has list of encoded features
     cache = nothing
     return fitresult, cache, report
 end;
 
 
 # 6. Transform method
 function MMI.transform(transformer::CardinalityReducer, fitresult, Xnew)
-    generic_cache = Dict(
-        :new_cat_given_col_val =>
-            fitresult,
+    generic_cache = (
+        new_cat_given_col_val = fitresult,
     )
     Xnew_transf = cardinality_reducer_transform(Xnew, generic_cache)
     return Xnew_transf