fiddler-client 3.0.0.dev14__tar.gz → 3.0.0.dev15__tar.gz

{fiddler-client-3.0.0.dev14/fiddler_client.egg-info → fiddler-client-3.0.0.dev15}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: fiddler-client
-Version: 3.0.0.dev14
+Version: 3.0.0.dev15
 Summary: Python client for Fiddler Platform
 Home-page: https://fiddler.ai
 Author: Fiddler Labs

fiddler-client-3.0.0.dev15/fiddler/VERSION

@@ -0,0 +1 @@
+3.0.0.dev15

{fiddler-client-3.0.0.dev14 → fiddler-client-3.0.0.dev15}/fiddler/entities/events.py

@@ -2,7 +2,7 @@ from __future__ import annotations
 
 import tempfile
 from pathlib import Path
-from typing import Any
+from typing import Any, Callable
 from uuid import UUID
 
 import pandas as pd
@@ -17,6 +17,9 @@ from fiddler.entities.job import Job
 from fiddler.schemas.dataset import EnvType
 from fiddler.schemas.events import EventsSource, FileSource
 from fiddler.schemas.job import JobCompactResp
+from fiddler.utils.logger import get_logger
+
+logger = get_logger(__name__)
 
 
 class EventPublisher(ConnectionMixin):
@@ -51,40 +54,72 @@ class EventPublisher(ConnectionMixin):
 
         :return: list[UUID] for list of dicts source and Job object for file path or dataframe source.
         """
+
+        publish_method = self._get_publish_method(source=source)
+
+        return publish_method(
+            source=source,
+            environment=environment,
+            dataset_name=dataset_name,
+            update=update,
+        )
+
+    def _get_publish_method(
+        self, source: list[dict[str, Any]] | str | Path | pd.DataFrame
+    ) -> Callable:
         if isinstance(source, (str, Path)):
-            return self._publish_file(
-                source=source,
-                environment=environment,
-                dataset_name=dataset_name,
-                update=update,
-            )
+            return self._publish_file
 
         if isinstance(source, pd.DataFrame):
-            with tempfile.NamedTemporaryFile(suffix='.csv') as temp_file:
-                source.to_csv(temp_file.name, index=False)
+            return self._publish_df
+
+        if isinstance(source, list):
+            return self._publish_stream
+
+        raise ValueError(f'Unsupported source - {type(source)}')
+
+    def _publish_df(
+        self,
+        source: pd.DataFrame,
+        environment: EnvType,
+        dataset_name: str | None = None,
+        update: bool = False,
+    ) -> Job:
+        with tempfile.NamedTemporaryFile(suffix='.parquet') as temp_file:
+            try:
+                source.to_parquet(temp_file.name, index=False)
                 return self._publish_file(
                     source=temp_file.name,
                     environment=environment,
                     dataset_name=dataset_name,
                     update=update,
                 )
-
-        elif isinstance(source, list):
-            return self._publish_stream(source=source, update=update)
-
-        raise ValueError(f'Unsupported source - {type(source)}')
+            except Exception:
+                logger.warning(
+                    'Failed to convert input dataframe to parquet format. Retrying as a CSV file.'
+                )
+        with tempfile.NamedTemporaryFile(suffix='.csv') as temp_file:
+            source.to_csv(temp_file.name, index=False)
+            return self._publish_file(
+                source=temp_file.name,
+                environment=environment,
+                dataset_name=dataset_name,
+                update=update,
+            )
 
     def _publish_stream(
         self,
         source: list[dict[str, Any]],
+        environment: EnvType = EnvType.PRODUCTION,
+        dataset_name: str | None = None,
         update: bool = False,
     ) -> list[UUID]:
         event_ids = []
         for i in tqdm(range(0, len(source), self.STREAM_LIMIT)):
             response = self._publish_call(
                 source=EventsSource(events=source[i : i + self.STREAM_LIMIT]),
-                environment=EnvType.PRODUCTION,
-                dataset_name=None,
+                environment=environment,
+                dataset_name=dataset_name,
                 update=update,
             )
             event_ids.extend(response.json()['data']['event_ids'])

{fiddler-client-3.0.0.dev14 → fiddler-client-3.0.0.dev15}/fiddler/schemas/custom_features.py

@@ -1,4 +1,4 @@
-from typing import Any, Dict, List, Optional, Type, TypeVar
+from typing import Any, Dict, List, Optional, Type, TypeVar, Literal
 
 from pydantic import BaseModel, validator
 
@@ -10,18 +10,12 @@ CustomFeatureTypeVar = TypeVar('CustomFeatureTypeVar', bound='CustomFeature')
 
 class CustomFeature(BaseModel):
     name: str
-    type: CustomFeatureType
-    n_clusters: Optional[int] = DEFAULT_NUM_CLUSTERS
-    centroids: Optional[List] = None
-
-    @validator('n_clusters')
-    def validate_n_clusters(cls, value: int) -> int:  # noqa: N805
-        if value < 0:
-            raise ValueError('n_clusters must be greater than 0')
-        return value
+    type: Any
 
     class Config:
         allow_mutation = False
+        use_enum_values = True
+        discriminator = 'type'
 
     @classmethod
     def from_columns(
@@ -53,12 +47,13 @@ class CustomFeature(BaseModel):
         return_dict: Dict[str, Any] = {
             'name': self.name,
             'type': self.type.value,
-            'n_clusters': self.n_clusters,
         }
         if isinstance(self, Multivariate):
             return_dict['columns'] = self.columns
+            return_dict['n_clusters'] = self.n_clusters
         elif isinstance(self, VectorFeature):
             return_dict['column'] = self.column
+            return_dict['n_clusters'] = self.n_clusters
             if isinstance(self, (ImageEmbedding, TextEmbedding)):
                 return_dict['source_column'] = self.source_column
                 if isinstance(self, TextEmbedding):
@@ -74,7 +69,9 @@ class CustomFeature(BaseModel):
 
 
 class Multivariate(CustomFeature):
-    type: CustomFeatureType = CustomFeatureType.FROM_COLUMNS
+    type: Literal['FROM_COLUMNS'] = CustomFeatureType.FROM_COLUMNS.value
+    n_clusters: Optional[int] = DEFAULT_NUM_CLUSTERS
+    centroids: Optional[List] = None
     columns: List[str]
     monitor_components: bool = False
 
@@ -84,24 +81,32 @@ class Multivariate(CustomFeature):
             raise ValueError('Multivariate columns must be greater than 1')
         return value
 
+    @validator('n_clusters')
+    def validate_n_clusters(cls, value: int) -> int:  # noqa: N805
+        if value < 0:
+            raise ValueError('n_clusters must be greater than 0')
+        return value
+
 
 class VectorFeature(CustomFeature):
-    type: CustomFeatureType = CustomFeatureType.FROM_VECTOR
-    source_column: Optional[str] = None
+    type: Literal['FROM_VECTOR'] = CustomFeatureType.FROM_VECTOR.value
+    n_clusters: Optional[int] = DEFAULT_NUM_CLUSTERS
+    centroids: Optional[List] = None
     column: str
 
+    @validator('n_clusters')
+    def validate_n_clusters(cls, value: int) -> int:  # noqa: N805
+        if value < 0:
+            raise ValueError('n_clusters must be greater than 0')
+        return value
+
 
 class TextEmbedding(VectorFeature):
-    type: CustomFeatureType = CustomFeatureType.FROM_TEXT_EMBEDDING
+    type: Literal['FROM_TEXT_EMBEDDING'] = CustomFeatureType.FROM_TEXT_EMBEDDING.value  # type: ignore
+    source_column: str
     n_tags: Optional[int] = DEFAULT_NUM_TAGS
     tf_idf: Optional[Dict[str, List]] = None
 
-    @validator('source_column')
-    def validate_source_column(cls, value: str) -> str:  # noqa: N805
-        if value is None:
-            raise ValueError('source_column must be specified')
-        return value
-
     @validator('n_tags')
     def validate_n_tags(cls, value: int) -> int:  # noqa: N805
         if value < 0:
@@ -110,13 +115,8 @@ class TextEmbedding(VectorFeature):
 
 
 class ImageEmbedding(VectorFeature):
-    type: CustomFeatureType = CustomFeatureType.FROM_IMAGE_EMBEDDING
-
-    @validator('source_column')
-    def validate_source_column(cls, value: str) -> str:  # noqa: N805
-        if value is None:
-            raise ValueError('source_column must be specified')
-        return value
+    type: Literal['FROM_IMAGE_EMBEDDING'] = CustomFeatureType.FROM_IMAGE_EMBEDDING.value  # type: ignore
+    source_column: str
 
 
 class Enrichment(CustomFeature):
@@ -135,7 +135,7 @@ class Enrichment(CustomFeature):
     """
 
     # Setting the feature type to ENRICHMENT
-    type: CustomFeatureType = CustomFeatureType.ENRICHMENT
+    type: Literal['ENRICHMENT'] = CustomFeatureType.ENRICHMENT.value
 
     # List of input column names used to generate the enrichment
     columns: List[str]

fiddler-client-3.0.0.dev15/fiddler/schemas/model_spec.py

@@ -0,0 +1,32 @@
+from typing import List, Union
+
+from pydantic import BaseModel, Field
+
+from fiddler.schemas.custom_features import Multivariate, VectorFeature, TextEmbedding, ImageEmbedding, Enrichment
+
+
+class ModelSpec(BaseModel):
+    """Model spec defines how model columns are used along with model task"""
+
+    schema_version: int = 1
+    """Schema version"""
+
+    inputs: List[str] = Field(default_factory=list)
+    """Feature columns"""
+
+    outputs: List[str] = Field(default_factory=list)
+    """Prediction columns"""
+
+    targets: List[str] = Field(default_factory=list)
+    """Label columns"""
+
+    decisions: List[str] = Field(default_factory=list)
+    """Decisions columns"""
+
+    metadata: List[str] = Field(default_factory=list)
+    """Metadata columns"""
+
+    custom_features: List[Union[Multivariate, VectorFeature, TextEmbedding, ImageEmbedding, Enrichment]] = Field(
+        default_factory=list
+    )
+    """Custom feature definitions"""

{fiddler-client-3.0.0.dev14 → fiddler-client-3.0.0.dev15/fiddler_client.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: fiddler-client
-Version: 3.0.0.dev14
+Version: 3.0.0.dev15
 Summary: Python client for Fiddler Platform
 Home-page: https://fiddler.ai
 Author: Fiddler Labs

fiddler-client-3.0.0.dev14/fiddler/VERSION

@@ -1 +0,0 @@
-3.0.0.dev14

fiddler-client-3.0.0.dev14/fiddler/schemas/model_spec.py

@@ -1,30 +0,0 @@
-from typing import List
-
-from pydantic import BaseModel
-
-from fiddler.schemas.custom_features import CustomFeature
-
-
-class ModelSpec(BaseModel):
-    """Model spec defines how model columns are used along with model task"""
-
-    schema_version: int = 1
-    """Schema version"""
-
-    inputs: List[str] = []
-    """Feature columns"""
-
-    outputs: List[str] = []
-    """Prediction columns"""
-
-    targets: List[str] = []
-    """Label columns"""
-
-    decisions: List[str] = []
-    """Decisions columns"""
-
-    metadata: List[str] = []
-    """Metadata columns"""
-
-    custom_features: List[CustomFeature] = []
-    """Custom feature definitions"""