Merge branch 'main' into add_more_reports

examples/fedopt_example/client_data.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import json
 from pathlib import Path
 from typing import Dict, List, Optional, Tuple
@@ -20,7 +22,7 @@ def __init__(self, classes: List[str], label_to_class: Dict[int, str], class_to_
         self.class_to_label = class_to_label
 
     @staticmethod
-    def encoder_from_dataframe(df: pd.DataFrame, class_column: str) -> "LabelEncoder":
+    def encoder_from_dataframe(df: pd.DataFrame, class_column: str) -> LabelEncoder:
         categories = df[class_column].astype("category")
         categories_str = [str(category) for category in categories.to_list()]
         label_to_class = dict(set(zip(categories.cat.codes, categories_str)))
@@ -29,7 +31,7 @@ def encoder_from_dataframe(df: pd.DataFrame, class_column: str) -> "LabelEncoder
         return LabelEncoder(classes, label_to_class, class_to_label)
 
     @staticmethod
-    def from_json(json_str: str) -> "LabelEncoder":
+    def from_json(json_str: str) -> LabelEncoder:
         attributes = json.loads(json_str)
         # need to cast string keys to int
         label_to_class = {int(label): category for label, category in json.loads(attributes["label_to_class"]).items()}
@@ -59,7 +61,7 @@ def __init__(self, vocabulary_dict: Optional[Dict[str, int]], train_set: Optiona
         elif train_set is not None:
             self._create_vocabulary(train_set)
         else:
-            raise ValueError("Must provide either precumputed dictionary or training set to create vocabulary")
+            raise ValueError("Must provide either precomputed dictionary or training set to create vocabulary")
         self.vocabulary_size = len(self.word2index)
 
     def _create_vocabulary(self, train_set: List[List[str]]) -> None:
@@ -95,7 +97,7 @@ def to_json(self) -> str:
         return json.dumps(self.word2index)
 
     @staticmethod
-    def from_json(json_str: str) -> "Vocabulary":
+    def from_json(json_str: str) -> Vocabulary:
         return Vocabulary(json.loads(json_str), None)
 
 

examples/fedopt_example/metrics.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import json
 from logging import INFO
 from typing import Dict, List, Optional
@@ -43,7 +45,7 @@ def summarize(self) -> Dict[str, float]:
         }
 
     @staticmethod
-    def from_results_dict(class_name: str, stats_string: str) -> "Outcome":
+    def from_results_dict(class_name: str, stats_string: str) -> Outcome:
         outcome = Outcome(class_name)
         stats = json.loads(stats_string)
         outcome.true_positive = stats[0]
@@ -52,7 +54,7 @@ def from_results_dict(class_name: str, stats_string: str) -> "Outcome":
         return outcome
 
     @staticmethod
-    def merge_outcomes(outcome_1: "Outcome", outcome_2: "Outcome") -> "Outcome":
+    def merge_outcomes(outcome_1: "Outcome", outcome_2: "Outcome") -> Outcome:
         assert outcome_1.class_name == outcome_2.class_name
         outcome_1.true_positive += outcome_2.true_positive
         outcome_1.false_negative += outcome_2.false_negative

fl4health/feature_alignment/string_columns_transformer.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 from typing import Optional
 
 import pandas as pd
@@ -15,7 +17,7 @@ class TextMulticolumnTransformer(BaseEstimator, TransformerMixin):
     def __init__(self, transformer: TextFeatureTransformer):
         self.transformer = transformer
 
-    def fit(self, X: pd.DataFrame, y: Optional[pd.DataFrame] = None) -> "TextMulticolumnTransformer":
+    def fit(self, X: pd.DataFrame, y: Optional[pd.DataFrame] = None) -> TextMulticolumnTransformer:
         joined_X = X.apply(lambda x: " ".join(x), axis=1)
         self.transformer.fit(joined_X)
         return self
@@ -34,7 +36,7 @@ class TextColumnTransformer(BaseEstimator, TransformerMixin):
     def __init__(self, transformer: TextFeatureTransformer):
         self.transformer = transformer
 
-    def fit(self, X: pd.DataFrame, y: Optional[pd.DataFrame] = None) -> "TextColumnTransformer":
+    def fit(self, X: pd.DataFrame, y: Optional[pd.DataFrame] = None) -> TextColumnTransformer:
         assert isinstance(X, pd.DataFrame) and X.shape[1] == 1
         self.transformer.fit(X[X.columns[0]])
         return self

fl4health/feature_alignment/tab_features_info_encoder.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import json
 from typing import Dict, List, Optional, Union
 
@@ -87,7 +89,7 @@ def encoder_from_dataframe(
         id_column: str,
         target_columns: Union[str, List[str]],
         fill_values: Optional[Dict[str, Scalar]] = None,
-    ) -> "TabularFeaturesInfoEncoder":
+    ) -> TabularFeaturesInfoEncoder:
         features_list = sorted(df.columns.values.tolist())
         features_list.remove(id_column)
         # Leverage cyclops to perform type inference
@@ -119,7 +121,7 @@ def to_json(self) -> str:
         )
 
     @staticmethod
-    def from_json(json_str: str) -> "TabularFeaturesInfoEncoder":
+    def from_json(json_str: str) -> TabularFeaturesInfoEncoder:
         attributes = json.loads(json_str)
         return TabularFeaturesInfoEncoder(
             [TabularFeature.from_json(tab_str) for tab_str in json.loads(attributes["tabular_features"])],

fl4health/feature_alignment/tabular_feature.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import json
 from typing import Optional, Union
 
@@ -70,7 +72,7 @@ def to_json(self) -> str:
         )
 
     @staticmethod
-    def from_json(json_str: str) -> "TabularFeature":
+    def from_json(json_str: str) -> TabularFeature:
         attributes = json.loads(json_str)
         return TabularFeature(
             json.loads(attributes["feature_name"]),