easy_ml 0.2.0.pre.rc104 → 0.2.0.pre.rc105

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 617edee53d32c1340b1a996a48e6a8a60d8cccff4345e27b2e5cf2ffc926c4ac
-  data.tar.gz: 225c133b9365d62e579e39e862ef76efaf7759d1c40a59b89968441381c8c5ee
+  metadata.gz: c0f70db7d7b59423ef34b5f3cd15d783ec22112977b44c1a5c7df390cabbc1c9
+  data.tar.gz: c985391e9231db8870c3a0dde14d41c9a0b75f0643b9ebb4ad30af7b57354230
 SHA512:
-  metadata.gz: 8a31abca5a4086eab323b1dfc112c05f89682bbe1fab0211ed111a0978cdc43f45dfd3ece03c6920001cfb4286d22b41e52421d280f02476e2feb63f0214eefd
-  data.tar.gz: 1f109b18e911eee6f81fe797f80320cf37a06e5ae25e4eb76894ffa98dbc931a66b63264462d36ce58e29ab0a7b95e90d522457e53b4fc15a71ece5473d73389
+  metadata.gz: d06176713f32434dc24b7250090a7c005b79a0c515d96e7ef35b67a28e9c5c6feb58ed2bb6a5136f98897c9ded66b8c9c5ee78018cb8056c50ef1f2491b26e0a
+  data.tar.gz: 8a049b2858d28e0f4eb7776a606a1e3375535c7ab0a13d41ed32a5c253b9fc979d9dcf705c0c7e99322340b8c0d3b21c38513885e79947347efd1b9bafb66347

data/app/models/easy_ml/dataset/learner/lazy/datetime.rb

@@ -11,7 +11,7 @@ module EasyML
 
           def unique_count
             Polars.col(column.name)
-                  .cast(column.polars_datatype)
+                  .cast(datatype)
                   .n_unique.alias("#{column.name}__unique_count")
           end
         end

data/app/models/easy_ml/dataset/learner/lazy/numeric.rb

@@ -6,27 +6,27 @@ module EasyML
           def train_query
             super.concat([
               Polars.col(column.name)
-                    .cast(column.polars_datatype)
+                    .cast(datatype)
                     .mean
                     .alias("#{column.name}__mean"),
 
               Polars.col(column.name)
-                      .cast(column.polars_datatype)
+                      .cast(datatype)
                       .median
                       .alias("#{column.name}__median"),
 
               Polars.col(column.name)
-                    .cast(column.polars_datatype)
+                    .cast(datatype)
                     .min
                     .alias("#{column.name}__min"),
 
               Polars.col(column.name)
-                    .cast(column.polars_datatype)
+                    .cast(datatype)
                     .max
                     .alias("#{column.name}__max"),
 
               Polars.col(column.name)
-                    .cast(column.polars_datatype)
+                    .cast(datatype)
                     .std
                     .alias("#{column.name}__std"),
             ])

data/app/models/easy_ml/dataset/learner/lazy/query.rb

@@ -33,6 +33,14 @@ module EasyML
             end
           end
 
+          def datatype
+            case column.polars_datatype.to_s
+            when /Polars::Categorical/ then Polars::String
+            else
+              column.polars_datatype
+            end
+          end
+
           private
 
           def full_dataset_query
@@ -45,21 +53,21 @@ module EasyML
 
           def null_count
             Polars.col(column.name)
-                  .cast(column.polars_datatype)
+                  .cast(datatype)
                   .null_count
                   .alias("#{column.name}__null_count")
           end
 
           def num_rows
             Polars.col(column.name)
-                  .cast(column.polars_datatype)
+                  .cast(datatype)
                   .len
                   .alias("#{column.name}__num_rows")
           end
 
           def most_frequent_value
             Polars.col(column.name)
-                  .cast(column.polars_datatype)
+                  .cast(datatype)
                   .filter(Polars.col(column.name).is_not_null)
                   .mode
                   .first
@@ -70,7 +78,7 @@ module EasyML
             return unless dataset.date_column.present?
 
             Polars.col(column.name)
-                  .cast(column.polars_datatype)
+                  .cast(datatype)
                   .sort_by(dataset.date_column.name, reverse: true, nulls_last: true)
                   .filter(Polars.col(column.name).is_not_null)
                   .first

data/lib/easy_ml/version.rb

@@ -1,7 +1,7 @@
 # frozen_string_literal: true
 
 module EasyML
-  VERSION = "0.2.0-rc104"
+  VERSION = "0.2.0-rc105"
 
   module Version
   end

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification
 name: easy_ml
 version: !ruby/object:Gem::Version
-  version: 0.2.0.pre.rc104
+  version: 0.2.0.pre.rc105
 platform: ruby
 authors:
 - Brett Shollenberger
 bindir: exe
 cert_chain: []
-date: 2025-03-18 00:00:00.000000000 Z
+date: 2025-03-19 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: activerecord