Merge branch 'dev/richard/wlw' of github.com:whylabs/whylogs into dev…

…/richard/wlw

python/.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 1.3.28
+current_version = 1.3.29
 tag = False
 parse = (?P<major>\d+)\.(?P<minor>\d+)\.(?P<patch>\d+)(\-(?P<release>[a-z]+)(?P<build>\d+))?
 serialize = 

python/Makefile

@@ -5,7 +5,7 @@ src.proto.dir := ../proto/src
 src.proto := $(shell find $(src.proto.dir) -type f -name "*.proto")
 src.proto.v0.dir := ../proto/v0
 src.proto.v0 := $(shell find $(src.proto.v0.dir) -type f -name "*.proto")
-version := 1.3.28
+version := 1.3.29
 
 dist.dir := dist
 egg.dir := .eggs

python/docs/conf.py

@@ -8,7 +8,7 @@
     print("Pandoc is required to build our documentation.")
     sys.exit(1)
 
-version = "1.3.28"
+version = "1.3.29"
 
 project = "whylogs"
 author = "whylogs developers"

python/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "whylogs"
-version = "1.3.28"
+version = "1.3.29"
 description = "Profile and monitor your ML data pipeline end-to-end"
 authors = ["WhyLabs.ai <support@whylabs.ai>"]
 license = "Apache-2.0"

python/tests/core/test_declarative_schema.py

@@ -208,6 +208,25 @@ def test_declarative_schema_with_additional_resolvers(pandas_dataframe):
     }.issubset(colset)
 
 
+def test_declarative_schema_add_multiple_resolvers(pandas_dataframe):
+    schema = DeclarativeSchema([])
+    schema.add_resolver_spec("animal", None, [MetricSpec(StandardMetric.cardinality.value)])
+    schema.add_resolver_spec({"legs", "weight"}, None, [MetricSpec(StandardMetric.distribution.value)])
+    results = why.log(pandas_dataframe, schema=schema).view()
+    metrics = set(results.get_column("animal").get_metric_names())
+    assert metrics == {"cardinality"}
+    metrics = set(results.get_column("legs").get_metric_names())
+    assert metrics == {"distribution"}
+    metrics = set(results.get_column("weight").get_metric_names())
+    assert metrics == {"distribution"}
+
+
+def test_invalid_column_name():
+    schema = DeclarativeSchema([])
+    with pytest.raises(ValueError):
+        schema.add_resolver_spec(42)
+
+
 def test_additional_metrics_nonexistent(pandas_dataframe):
     count_spec = ResolverSpec(
         column_name="nonexistent_columns",

python/tests/migration/test_uncompound.py

@@ -141,3 +141,10 @@ def test_uncompounded_condition_count() -> None:
             assert metric.null.value == 0
             assert metric.nan.value == 0
             assert metric.inf.value == 0
+            metric = uncompounded._columns[column_name]._metrics["types"]
+            assert metric.integral.value == 2
+            assert metric.fractional.value == 0
+            assert metric.boolean.value == 0
+            assert metric.string.value == 0
+            assert metric.object.value == 0
+            assert metric.tensor.value == 0

python/tests/smoketest.py

@@ -12,7 +12,7 @@
 not a development environment.
 """
 
-current_version = "1.3.28"
+current_version = "1.3.29"
 
 
 def test_package_version() -> None:

python/whylogs/core/schema.py

@@ -1,7 +1,7 @@
 import logging
 from copy import deepcopy
 from dataclasses import dataclass, field
-from typing import Any, Dict, List, Mapping, Optional, Tuple, TypeVar
+from typing import Any, Dict, List, Mapping, Optional, Set, Tuple, TypeVar, Union
 
 import whylogs.core.resolvers as res
 from whylogs.core.datatypes import StandardTypeMapper, TypeMapper
@@ -232,10 +232,19 @@ def add_resolver(self, resolver_spec: ResolverSpec):
 
     def add_resolver_spec(
         self,
-        column_name: Optional[str] = None,
+        column_name: Optional[Union[str, Set[str]]] = None,
         column_type: Optional[Any] = None,
         metrics: Optional[List[MetricSpec]] = None,
     ):
+        if column_name is not None and not isinstance(column_name, (str, set)):
+            raise ValueError("column_name must be a stirng or set of strings")
+
+        if isinstance(column_name, set):
+            for name in column_name:
+                spec = ResolverSpec(column_name=name, column_type=column_type, metrics=metrics or [])
+                self.add_resolver(spec)
+            return
+
         spec = ResolverSpec(column_name=column_name, column_type=column_type, metrics=metrics or [])
         self.add_resolver(spec)
 

python/whylogs/migration/uncompound.py

@@ -4,7 +4,7 @@
 
 from whylogs.core import ColumnProfileView, DatasetProfileView
 from whylogs.core.metrics import Metric
-from whylogs.core.metrics.column_metrics import ColumnCountsMetric
+from whylogs.core.metrics.column_metrics import ColumnCountsMetric, TypeCountersMetric
 from whylogs.core.metrics.compound_metric import CompoundMetric
 from whylogs.core.metrics.condition_count_metric import ConditionCountMetric
 from whylogs.core.metrics.metric_components import IntegralComponent
@@ -118,6 +118,14 @@ def _uncompound_condition_count(
         return dict()
 
     result: Dict[str, ColumnProfileView] = dict()
+    type_metric = TypeCountersMetric(
+        integral=metric.total,  # total condition evaluations
+        fractional=IntegralComponent(0),
+        boolean=IntegralComponent(0),
+        string=IntegralComponent(0),
+        object=IntegralComponent(0),
+        tensor=IntegralComponent(0),
+    )
     for condition_name, count_component in metric.matches.items():
         new_col_name = f"{_condition_count_magic_string()}{col_name}.{condition_name}.total"
         new_metric = ColumnCountsMetric(
@@ -126,23 +134,38 @@ def _uncompound_condition_count(
             nan=IntegralComponent(0),  # unused
             inf=IntegralComponent(0),  # unused
         )
-        result[new_col_name] = ColumnProfileView({ColumnCountsMetric.get_namespace(): new_metric})
+        result[new_col_name] = ColumnProfileView(
+            {
+                ColumnCountsMetric.get_namespace(): new_metric,
+                TypeCountersMetric.get_namespace(): type_metric,
+            }
+        )
         new_col_name = f"{_condition_count_magic_string()}{col_name}.{condition_name}.matches"
         new_metric = ColumnCountsMetric(
             n=count_component,  # count of evaluations that matched condition
             null=IntegralComponent(0),  # unused
             nan=IntegralComponent(0),  # unused
             inf=IntegralComponent(0),  # unused
         )
-        result[new_col_name] = ColumnProfileView({ColumnCountsMetric.get_namespace(): new_metric})
+        result[new_col_name] = ColumnProfileView(
+            {
+                ColumnCountsMetric.get_namespace(): new_metric,
+                TypeCountersMetric.get_namespace(): type_metric,
+            }
+        )
         new_col_name = f"{_condition_count_magic_string()}{col_name}.{condition_name}.non_matches"
         new_metric = ColumnCountsMetric(
             n=IntegralComponent(metric.total.value - count_component.value),  # evaluations that didn't match
             null=IntegralComponent(0),  # unused
             nan=IntegralComponent(0),  # unused
             inf=IntegralComponent(0),  # unused
         )
-        result[new_col_name] = ColumnProfileView({ColumnCountsMetric.get_namespace(): new_metric})
+        result[new_col_name] = ColumnProfileView(
+            {
+                ColumnCountsMetric.get_namespace(): new_metric,
+                TypeCountersMetric.get_namespace(): type_metric,
+            }
+        )
 
     return result