easy_ml 0.2.0.pre.rc85 → 0.2.0.pre.rc88
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/app/controllers/easy_ml/datasets_controller.rb +18 -2
- data/app/frontend/components/dataset/PreprocessingConfig.tsx +523 -150
- data/app/frontend/pages/DatasetsPage.tsx +0 -1
- data/app/frontend/types/dataset.ts +5 -2
- data/app/models/easy_ml/column/imputers/base.rb +23 -2
- data/app/models/easy_ml/column/imputers/embedding_encoder.rb +18 -0
- data/app/models/easy_ml/column/imputers/imputer.rb +1 -0
- data/app/models/easy_ml/column/imputers/most_frequent.rb +1 -1
- data/app/models/easy_ml/column/imputers/one_hot_encoder.rb +1 -1
- data/app/models/easy_ml/column/imputers/ordinal_encoder.rb +1 -1
- data/app/models/easy_ml/column/imputers.rb +47 -41
- data/app/models/easy_ml/column/selector.rb +2 -2
- data/app/models/easy_ml/column.rb +260 -56
- data/app/models/easy_ml/column_history.rb +6 -0
- data/app/models/easy_ml/column_list.rb +30 -1
- data/app/models/easy_ml/dataset/learner/lazy/embedding.rb +10 -0
- data/app/models/easy_ml/dataset/learner/lazy/query.rb +2 -0
- data/app/models/easy_ml/dataset/learner.rb +11 -0
- data/app/models/easy_ml/dataset.rb +6 -19
- data/app/models/easy_ml/lineage_history.rb +17 -0
- data/app/models/easy_ml/model.rb +11 -1
- data/app/models/easy_ml/models/xgboost.rb +37 -7
- data/app/models/easy_ml/pca_model.rb +21 -0
- data/app/models/easy_ml/prediction.rb +2 -1
- data/app/serializers/easy_ml/column_serializer.rb +13 -1
- data/config/initializers/inflections.rb +1 -0
- data/lib/easy_ml/data/dataset_manager/writer/append_only.rb +6 -8
- data/lib/easy_ml/data/dataset_manager/writer/base.rb +15 -2
- data/lib/easy_ml/data/dataset_manager/writer/partitioned.rb +0 -1
- data/lib/easy_ml/data/dataset_manager/writer.rb +2 -0
- data/lib/easy_ml/data/embeddings/compressor.rb +179 -0
- data/lib/easy_ml/data/embeddings/embedder.rb +226 -0
- data/lib/easy_ml/data/embeddings.rb +61 -0
- data/lib/easy_ml/data/polars_column.rb +3 -0
- data/lib/easy_ml/data/polars_reader.rb +54 -23
- data/lib/easy_ml/data/polars_schema.rb +28 -2
- data/lib/easy_ml/data/splits/file_split.rb +7 -2
- data/lib/easy_ml/data.rb +1 -0
- data/lib/easy_ml/embedding_store.rb +92 -0
- data/lib/easy_ml/engine.rb +4 -2
- data/lib/easy_ml/predict.rb +42 -20
- data/lib/easy_ml/railtie/generators/migration/migration_generator.rb +5 -0
- data/lib/easy_ml/railtie/templates/migration/add_is_primary_key_to_easy_ml_columns.rb.tt +9 -0
- data/lib/easy_ml/railtie/templates/migration/add_metadata_to_easy_ml_predictions.rb.tt +6 -0
- data/lib/easy_ml/railtie/templates/migration/add_pca_model_id_to_easy_ml_columns.rb.tt +9 -0
- data/lib/easy_ml/railtie/templates/migration/add_workflow_status_to_easy_ml_dataset_histories.rb.tt +13 -0
- data/lib/easy_ml/railtie/templates/migration/create_easy_ml_pca_models.rb.tt +14 -0
- data/lib/easy_ml/version.rb +1 -1
- data/lib/easy_ml.rb +1 -0
- data/public/easy_ml/assets/.vite/manifest.json +2 -2
- data/public/easy_ml/assets/assets/Application-DfPoyRr8.css +1 -0
- data/public/easy_ml/assets/assets/entrypoints/Application.tsx-KENNRQpC.js +533 -0
- data/public/easy_ml/assets/assets/entrypoints/Application.tsx-KENNRQpC.js.map +1 -0
- metadata +59 -6
- data/lib/tasks/profile.rake +0 -40
- data/public/easy_ml/assets/assets/Application-nnn_XLuL.css +0 -1
- data/public/easy_ml/assets/assets/entrypoints/Application.tsx-CD8voxfL.js +0 -522
- data/public/easy_ml/assets/assets/entrypoints/Application.tsx-CD8voxfL.js.map +0 -1
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: f9a3eb82ce50657d230c9d00a75f84f7f09ce96b68d58fa8a1d615ad4fd31d77
|
|
4
|
+
data.tar.gz: f0ded08a9c73232bcea3eb23b9e678bc392647873e53c425c6b21f62104c983b
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 76b6f960c24f0032d4959f7435770f64245a5699f97150118127fac618e46965267ae996c3be3b4bb21a0734177c039ec2357b690a69ede3117c834e3ff09083
|
|
7
|
+
data.tar.gz: 241ae30e3e2934eef6a74aae16aa05cfb1e10ed9145562045ccb93403c292b46449003badee5a0798a70987394469d6781b6ef8d14d5a552ee9da954efc75f44
|
|
@@ -73,7 +73,23 @@ module EasyML
|
|
|
73
73
|
|
|
74
74
|
# Iterate over columns to check and update preprocessing_steps
|
|
75
75
|
dataset_params[:columns_attributes]&.each do |_, column_attrs|
|
|
76
|
-
|
|
76
|
+
if column_attrs.dig(:preprocessing_steps, :training, :method) == "none"
|
|
77
|
+
column_attrs[:preprocessing_steps] = nil
|
|
78
|
+
elsif column_attrs.dig(:preprocessing_steps, :training)
|
|
79
|
+
# Ensure encoding is properly set for categorical columns
|
|
80
|
+
training_config = column_attrs.dig(:preprocessing_steps, :training)
|
|
81
|
+
if training_config[:params]
|
|
82
|
+
# Remove old encoding params as they're now part of the encoding field
|
|
83
|
+
training_config[:params].delete(:one_hot)
|
|
84
|
+
training_config[:params].delete(:ordinal_encoding)
|
|
85
|
+
end
|
|
86
|
+
|
|
87
|
+
# Ensure embedding params are present when encoding is embedding
|
|
88
|
+
if training_config[:encoding] == "embedding" && training_config[:params]
|
|
89
|
+
training_config[:params][:llm] ||= "openai"
|
|
90
|
+
training_config[:params][:model] ||= "text-embedding-3-small"
|
|
91
|
+
end
|
|
92
|
+
end
|
|
77
93
|
end
|
|
78
94
|
|
|
79
95
|
# Handle feature ID assignment for existing features
|
|
@@ -165,7 +181,7 @@ module EasyML
|
|
|
165
181
|
private
|
|
166
182
|
|
|
167
183
|
def preprocessing_params
|
|
168
|
-
[:method, { params: [:constant, :categorical_min, :
|
|
184
|
+
[:method, :encoding, { params: [:constant, :categorical_min, :llm, :model, :preset, :dimensions, { clip: %i[min max] }] }]
|
|
169
185
|
end
|
|
170
186
|
|
|
171
187
|
def dataset_params
|