Inference store objects
Objects and models for inference mapping, inference tracking, and performance metrics. See supported Inference store APIs.
vianops_client.models.inferencestore.inference_mapping
- pydantic model V1InferenceMappingColumnSchema
Bases:
BaseModelShow JSON schema
{ "title": "V1InferenceMappingColumnSchema", "type": "object", "properties": { "feature_id": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Id of the feature, generated by the platform for optimization.", "title": "Feature Id" }, "name": { "description": "Name of column.", "title": "Name", "type": "string" }, "dtype": { "description": "Data type of column. For example, `float64`.", "title": "Dtype", "type": "string" }, "sql_type": { "description": "SQL data type of column. For example, `Float32`.", "title": "Sql Type", "type": "string" }, "feature_type": { "description": "Feature type of column. Acceptable values include `categorical` and `continuous`.", "title": "Feature Type", "type": "string" }, "segmentation": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": true, "description": "If true (the default), the column is enabled for segmentation calculations.", "title": "Segmentation" }, "drift": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": true, "description": "If true (the default), the column is enabled for drift calculations.", "title": "Drift" }, "hotspot": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": false, "description": "If true (the default), the column is enabled for hotspot analysis.", "title": "Hotspot" }, "rca": { "anyOf": [ { "items": { "type": "string" }, "type": "array" }, { "type": "null" } ], "default": [], "description": "List of column names included for root cause analysis (rca). Set by the platform based on feature configuration to identify features available for hotspot analysis.", "title": "Rca" }, "round": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "The number of decimals to round to on a continuous float column.", "title": "Round" } }, "required": [ "name", "dtype", "sql_type", "feature_type" ] }
- Fields:
- field drift: bool | None = True
If true (the default), the column is enabled for drift calculations.
- field dtype: str [Required]
Data type of column. For example, float64.
- field feature_id: int | None = None
Id of the feature, generated by the platform for optimization.
- field feature_type: str [Required]
Feature type of column. Acceptable values include categorical and continuous.
- field hotspot: bool | None = False
If true (the default), the column is enabled for hotspot analysis.
- field name: str [Required]
Name of column.
- field rca: List[str] | None = []
List of column names included for root cause analysis (rca). Set by the platform based on feature configuration to identify features available for hotspot analysis.
- field round: int | None = None
The number of decimals to round to on a continuous float column.
- field segmentation: bool | None = True
If true (the default), the column is enabled for segmentation calculations.
- field sql_type: str [Required]
SQL data type of column. For example, Float32.
- pydantic model V1InferenceMappingJob
Bases:
BaseModelShow JSON schema
{ "title": "V1InferenceMappingJob", "type": "object", "properties": { "deployment": { "description": "Deployment name.", "title": "Deployment", "type": "string" }, "model_name": { "description": "Model name. Default value is deployment name.", "title": "Model Name", "type": "string" }, "model_version": { "default": "1", "description": "Model version. Default value is `1`.", "title": "Model Version", "type": "string" }, "model_stage": { "default": "primary", "description": "Model stage. Default value is `primary`.", "title": "Model Stage", "type": "string" }, "connection": { "description": "Database connection.", "title": "Connection", "type": "string" }, "srcurl": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Source url corresponding to location of data.", "title": "Srcurl" }, "df_schema": { "$ref": "#/$defs/V1InferenceMappingSchema", "description": "Dataframe schema." }, "training_table": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Training table name that needs to be specified for training data preprocessing.", "title": "Training Table" } }, "$defs": { "V1InferenceMappingColumnSchema": { "properties": { "feature_id": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Id of the feature, generated by the platform for optimization.", "title": "Feature Id" }, "name": { "description": "Name of column.", "title": "Name", "type": "string" }, "dtype": { "description": "Data type of column. For example, `float64`.", "title": "Dtype", "type": "string" }, "sql_type": { "description": "SQL data type of column. For example, `Float32`.", "title": "Sql Type", "type": "string" }, "feature_type": { "description": "Feature type of column. Acceptable values include `categorical` and `continuous`.", "title": "Feature Type", "type": "string" }, "segmentation": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": true, "description": "If true (the default), the column is enabled for segmentation calculations.", "title": "Segmentation" }, "drift": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": true, "description": "If true (the default), the column is enabled for drift calculations.", "title": "Drift" }, "hotspot": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": false, "description": "If true (the default), the column is enabled for hotspot analysis.", "title": "Hotspot" }, "rca": { "anyOf": [ { "items": { "type": "string" }, "type": "array" }, { "type": "null" } ], "default": [], "description": "List of column names included for root cause analysis (rca). Set by the platform based on feature configuration to identify features available for hotspot analysis.", "title": "Rca" }, "round": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "The number of decimals to round to on a continuous float column.", "title": "Round" } }, "required": [ "name", "dtype", "sql_type", "feature_type" ], "title": "V1InferenceMappingColumnSchema", "type": "object" }, "V1InferenceMappingSchema": { "properties": { "target_col": { "description": "Name of target column.", "title": "Target Col", "type": "string" }, "identifier_cols": { "default": [], "description": "List of identifier columns.", "items": { "type": "string" }, "title": "Identifier Cols", "type": "array" }, "columns": { "description": "List of all columns.", "items": { "$ref": "#/$defs/V1InferenceMappingColumnSchema" }, "title": "Columns", "type": "array" }, "datetime_col": { "default": "", "description": "Name of datetime column", "title": "Datetime Col", "type": "string" }, "predict_proba_col": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "(Classification models only) Name of column that provides the predict probability.", "title": "Predict Proba Col" } }, "required": [ "target_col", "columns" ], "title": "V1InferenceMappingSchema", "type": "object" } }, "required": [ "deployment", "model_name", "connection", "df_schema" ] }
- Config:
protected_namespaces: tuple = ()
- Fields:
- Validators:
- field connection: str [Required]
Database connection.
- field deployment: str [Required]
Deployment name.
- field df_schema: V1InferenceMappingSchema [Required]
Dataframe schema.
- field model_name: str [Required]
Model name. Default value is deployment name.
- Validated by:
- field model_stage: str = 'primary'
Model stage. Default value is primary.
- field model_version: str = '1'
Model version. Default value is 1.
- field srcurl: str | None = None
Source url corresponding to location of data.
- field training_table: str | None = None
Training table name that needs to be specified for training data preprocessing.
- validator grouped_filters_validator » model_name
- pydantic model V1InferenceMappingModel
Bases:
BaseModelShow JSON schema
{ "title": "V1InferenceMappingModel", "type": "object", "properties": { "index": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Index identifier.", "title": "Index" }, "deployment": { "description": "Deployment name for the related model.", "title": "Deployment", "type": "string" }, "model_name": { "description": "Model name for the related model.", "title": "Model Name", "type": "string" }, "model_version": { "description": "Model version for the related model, e.g., 1.", "title": "Model Version", "type": "string" }, "model_stage": { "description": "Model stage for the related model, e.g., primary.", "title": "Model Stage", "type": "string" }, "connection": { "description": "Database connection for the inference mapping.", "title": "Connection", "type": "string" }, "df_schema": { "$ref": "#/$defs/V1InferenceMappingSchema", "description": "Dictionary containing dataframe schema for model, i.e., target column, identifier column(s), all columns, datetime column, and predict probability column (classification models only)." }, "identifier_cols_schema": { "anyOf": [ {}, { "type": "null" } ], "default": null, "description": "List containing identifier columns schema for model.", "title": "Identifier Cols Schema" }, "inference_table": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Name of inference table for mapping.", "title": "Inference Table" }, "ground_truth_table": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Name of ground truth table for mapping.", "title": "Ground Truth Table" }, "joined_table": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Name of joined table for mapping.", "title": "Joined Table" }, "key_table": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Name of key table for mapping.", "title": "Key Table" }, "create_ddl": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "The query to create the ddl tables.", "title": "Create Ddl" }, "postprocessors": { "anyOf": [ { "items": { "type": "object" }, "type": "array" }, { "type": "null" } ], "default": [], "description": "One or more postprocessor (and related params) to include in inference mapping, to transform output data before sending to backend database. The available postprocessors include `PickHighestProbability`, `PickInferenceMapper`, and `PickProbablityThreshold`. For each postprocessor, specify the classname, method, and initparams.", "title": "Postprocessors" }, "training_set_id": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Training set id that needs to be specified for training data preprocessing.", "title": "Training Set Id" }, "training_table": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Training table name that needs to be specified for training data preprocessing.", "title": "Training Table" } }, "$defs": { "V1InferenceMappingColumnSchema": { "properties": { "feature_id": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Id of the feature, generated by the platform for optimization.", "title": "Feature Id" }, "name": { "description": "Name of column.", "title": "Name", "type": "string" }, "dtype": { "description": "Data type of column. For example, `float64`.", "title": "Dtype", "type": "string" }, "sql_type": { "description": "SQL data type of column. For example, `Float32`.", "title": "Sql Type", "type": "string" }, "feature_type": { "description": "Feature type of column. Acceptable values include `categorical` and `continuous`.", "title": "Feature Type", "type": "string" }, "segmentation": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": true, "description": "If true (the default), the column is enabled for segmentation calculations.", "title": "Segmentation" }, "drift": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": true, "description": "If true (the default), the column is enabled for drift calculations.", "title": "Drift" }, "hotspot": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": false, "description": "If true (the default), the column is enabled for hotspot analysis.", "title": "Hotspot" }, "rca": { "anyOf": [ { "items": { "type": "string" }, "type": "array" }, { "type": "null" } ], "default": [], "description": "List of column names included for root cause analysis (rca). Set by the platform based on feature configuration to identify features available for hotspot analysis.", "title": "Rca" }, "round": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "The number of decimals to round to on a continuous float column.", "title": "Round" } }, "required": [ "name", "dtype", "sql_type", "feature_type" ], "title": "V1InferenceMappingColumnSchema", "type": "object" }, "V1InferenceMappingSchema": { "properties": { "target_col": { "description": "Name of target column.", "title": "Target Col", "type": "string" }, "identifier_cols": { "default": [], "description": "List of identifier columns.", "items": { "type": "string" }, "title": "Identifier Cols", "type": "array" }, "columns": { "description": "List of all columns.", "items": { "$ref": "#/$defs/V1InferenceMappingColumnSchema" }, "title": "Columns", "type": "array" }, "datetime_col": { "default": "", "description": "Name of datetime column", "title": "Datetime Col", "type": "string" }, "predict_proba_col": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "(Classification models only) Name of column that provides the predict probability.", "title": "Predict Proba Col" } }, "required": [ "target_col", "columns" ], "title": "V1InferenceMappingSchema", "type": "object" } }, "example": { "connection": "Internal clickhouse DB", "create_ddl": null, "deployment": "vianai.jt_placeholder_deployment", "df_schema": { "columns": [ { "drift": true, "dtype": "float64", "feature_id": 1, "feature_type": "categorical", "name": "PULocationID", "segmentation": true, "sql_type": "Float32" }, { "drift": true, "dtype": "float64", "feature_id": 2, "feature_type": "categorical", "name": "DOLocationID", "segmentation": true, "sql_type": "Float32" }, { "drift": true, "dtype": "float64", "feature_id": 3, "feature_type": "continuous", "name": "trip_distance", "segmentation": true, "sql_type": "Float32" }, { "drift": true, "dtype": "float64", "feature_id": 4, "feature_type": "continuous", "name": "travel_time", "segmentation": true, "sql_type": "Float32" }, { "drift": false, "dtype": "datetime64[ns]", "feature_id": 5, "feature_type": "unknown", "name": "PUdatetime", "sql_type": "" } ], "datetime_col": "PUdatetime", "identifier_cols": [ "pk" ], "target_col": "total_amount" }, "ground_truth_table": "16794179557100_gt", "identifier_cols_schema": null, "index": 7, "inference_table": "16794179557100_dit", "joined_table": "16794179557100_joined", "key_table": "16794179557100_key", "model_name": "vianai.jt_placeholder_deployment", "model_stage": "primary", "model_version": "1", "postprocessors": [ { "classname": "vianai.postprocessing.classification.PickInferenceMapper", "initparams": { "mapper": "{1:'a',2:'b',3:'c'}" }, "method": "fit_transform" } ] }, "required": [ "deployment", "model_name", "model_version", "model_stage", "connection", "df_schema" ] }
- Config:
protected_namespaces: tuple = ()
from_attributes: bool = True
json_schema_extra: dict = {‘example’: {‘index’: 7, ‘deployment’: ‘vianai.jt_placeholder_deployment’, ‘model_name’: ‘vianai.jt_placeholder_deployment’, ‘model_version’: ‘1’, ‘model_stage’: ‘primary’, ‘connection’: ‘Internal clickhouse DB’, ‘df_schema’: {‘target_col’: ‘total_amount’, ‘identifier_cols’: [‘pk’], ‘columns’: [{‘name’: ‘PULocationID’, ‘dtype’: ‘float64’, ‘sql_type’: ‘Float32’, ‘feature_type’: ‘categorical’, ‘feature_id’: 1, ‘drift’: True, ‘segmentation’: True}, {‘name’: ‘DOLocationID’, ‘dtype’: ‘float64’, ‘sql_type’: ‘Float32’, ‘feature_type’: ‘categorical’, ‘feature_id’: 2, ‘drift’: True, ‘segmentation’: True}, {‘name’: ‘trip_distance’, ‘dtype’: ‘float64’, ‘sql_type’: ‘Float32’, ‘feature_type’: ‘continuous’, ‘feature_id’: 3, ‘drift’: True, ‘segmentation’: True}, {‘name’: ‘travel_time’, ‘dtype’: ‘float64’, ‘sql_type’: ‘Float32’, ‘feature_type’: ‘continuous’, ‘feature_id’: 4, ‘drift’: True, ‘segmentation’: True}, {‘name’: ‘PUdatetime’, ‘dtype’: ‘datetime64[ns]’, ‘sql_type’: ‘’, ‘feature_type’: ‘unknown’, ‘feature_id’: 5, ‘drift’: False}], ‘datetime_col’: ‘PUdatetime’}, ‘identifier_cols_schema’: None, ‘inference_table’: ‘16794179557100_dit’, ‘ground_truth_table’: ‘16794179557100_gt’, ‘joined_table’: ‘16794179557100_joined’, ‘key_table’: ‘16794179557100_key’, ‘create_ddl’: None, ‘postprocessors’: [{‘classname’: ‘vianai.postprocessing.classification.PickInferenceMapper’, ‘method’: ‘fit_transform’, ‘initparams’: {‘mapper’: “{1:’a’,2:’b’,3:’c’}”}}]}}
- Fields:
- field connection: str [Required]
Database connection for the inference mapping.
- field create_ddl: str | None = None
The query to create the ddl tables.
- field deployment: str [Required]
Deployment name for the related model.
- field df_schema: V1InferenceMappingSchema [Required]
Dictionary containing dataframe schema for model, i.e., target column, identifier column(s), all columns, datetime column, and predict probability column (classification models only).
- field ground_truth_table: str | None = None
Name of ground truth table for mapping.
- field identifier_cols_schema: Any | None = None
List containing identifier columns schema for model.
- field index: int | None = None
Index identifier.
- field inference_table: str | None = None
Name of inference table for mapping.
- field joined_table: str | None = None
Name of joined table for mapping.
- field key_table: str | None = None
Name of key table for mapping.
- field model_name: str [Required]
Model name for the related model.
- field model_stage: str [Required]
Model stage for the related model, e.g., primary.
- field model_version: str [Required]
Model version for the related model, e.g., 1.
- field postprocessors: List[dict] | None = []
One or more postprocessor (and related params) to include in inference mapping, to transform output data before sending to backend database. The available postprocessors include PickHighestProbability, PickInferenceMapper, and PickProbablityThreshold. For each postprocessor, specify the classname, method, and initparams.
- field training_set_id: int | None = None
Training set id that needs to be specified for training data preprocessing.
- field training_table: str | None = None
Training table name that needs to be specified for training data preprocessing.
- pydantic model V1InferenceMappingModelList
Bases:
RootModelShow JSON schema
{ "title": "V1InferenceMappingModelList", "type": "array", "$defs": { "V1InferenceMappingColumnSchema": { "properties": { "feature_id": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Id of the feature, generated by the platform for optimization.", "title": "Feature Id" }, "name": { "description": "Name of column.", "title": "Name", "type": "string" }, "dtype": { "description": "Data type of column. For example, `float64`.", "title": "Dtype", "type": "string" }, "sql_type": { "description": "SQL data type of column. For example, `Float32`.", "title": "Sql Type", "type": "string" }, "feature_type": { "description": "Feature type of column. Acceptable values include `categorical` and `continuous`.", "title": "Feature Type", "type": "string" }, "segmentation": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": true, "description": "If true (the default), the column is enabled for segmentation calculations.", "title": "Segmentation" }, "drift": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": true, "description": "If true (the default), the column is enabled for drift calculations.", "title": "Drift" }, "hotspot": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": false, "description": "If true (the default), the column is enabled for hotspot analysis.", "title": "Hotspot" }, "rca": { "anyOf": [ { "items": { "type": "string" }, "type": "array" }, { "type": "null" } ], "default": [], "description": "List of column names included for root cause analysis (rca). Set by the platform based on feature configuration to identify features available for hotspot analysis.", "title": "Rca" }, "round": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "The number of decimals to round to on a continuous float column.", "title": "Round" } }, "required": [ "name", "dtype", "sql_type", "feature_type" ], "title": "V1InferenceMappingColumnSchema", "type": "object" }, "V1InferenceMappingModel": { "example": { "connection": "Internal clickhouse DB", "create_ddl": null, "deployment": "vianai.jt_placeholder_deployment", "df_schema": { "columns": [ { "drift": true, "dtype": "float64", "feature_id": 1, "feature_type": "categorical", "name": "PULocationID", "segmentation": true, "sql_type": "Float32" }, { "drift": true, "dtype": "float64", "feature_id": 2, "feature_type": "categorical", "name": "DOLocationID", "segmentation": true, "sql_type": "Float32" }, { "drift": true, "dtype": "float64", "feature_id": 3, "feature_type": "continuous", "name": "trip_distance", "segmentation": true, "sql_type": "Float32" }, { "drift": true, "dtype": "float64", "feature_id": 4, "feature_type": "continuous", "name": "travel_time", "segmentation": true, "sql_type": "Float32" }, { "drift": false, "dtype": "datetime64[ns]", "feature_id": 5, "feature_type": "unknown", "name": "PUdatetime", "sql_type": "" } ], "datetime_col": "PUdatetime", "identifier_cols": [ "pk" ], "target_col": "total_amount" }, "ground_truth_table": "16794179557100_gt", "identifier_cols_schema": null, "index": 7, "inference_table": "16794179557100_dit", "joined_table": "16794179557100_joined", "key_table": "16794179557100_key", "model_name": "vianai.jt_placeholder_deployment", "model_stage": "primary", "model_version": "1", "postprocessors": [ { "classname": "vianai.postprocessing.classification.PickInferenceMapper", "initparams": { "mapper": "{1:'a',2:'b',3:'c'}" }, "method": "fit_transform" } ] }, "properties": { "index": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Index identifier.", "title": "Index" }, "deployment": { "description": "Deployment name for the related model.", "title": "Deployment", "type": "string" }, "model_name": { "description": "Model name for the related model.", "title": "Model Name", "type": "string" }, "model_version": { "description": "Model version for the related model, e.g., 1.", "title": "Model Version", "type": "string" }, "model_stage": { "description": "Model stage for the related model, e.g., primary.", "title": "Model Stage", "type": "string" }, "connection": { "description": "Database connection for the inference mapping.", "title": "Connection", "type": "string" }, "df_schema": { "$ref": "#/$defs/V1InferenceMappingSchema", "description": "Dictionary containing dataframe schema for model, i.e., target column, identifier column(s), all columns, datetime column, and predict probability column (classification models only)." }, "identifier_cols_schema": { "anyOf": [ {}, { "type": "null" } ], "default": null, "description": "List containing identifier columns schema for model.", "title": "Identifier Cols Schema" }, "inference_table": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Name of inference table for mapping.", "title": "Inference Table" }, "ground_truth_table": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Name of ground truth table for mapping.", "title": "Ground Truth Table" }, "joined_table": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Name of joined table for mapping.", "title": "Joined Table" }, "key_table": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Name of key table for mapping.", "title": "Key Table" }, "create_ddl": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "The query to create the ddl tables.", "title": "Create Ddl" }, "postprocessors": { "anyOf": [ { "items": { "type": "object" }, "type": "array" }, { "type": "null" } ], "default": [], "description": "One or more postprocessor (and related params) to include in inference mapping, to transform output data before sending to backend database. The available postprocessors include `PickHighestProbability`, `PickInferenceMapper`, and `PickProbablityThreshold`. For each postprocessor, specify the classname, method, and initparams.", "title": "Postprocessors" }, "training_set_id": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Training set id that needs to be specified for training data preprocessing.", "title": "Training Set Id" }, "training_table": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Training table name that needs to be specified for training data preprocessing.", "title": "Training Table" } }, "required": [ "deployment", "model_name", "model_version", "model_stage", "connection", "df_schema" ], "title": "V1InferenceMappingModel", "type": "object" }, "V1InferenceMappingSchema": { "properties": { "target_col": { "description": "Name of target column.", "title": "Target Col", "type": "string" }, "identifier_cols": { "default": [], "description": "List of identifier columns.", "items": { "type": "string" }, "title": "Identifier Cols", "type": "array" }, "columns": { "description": "List of all columns.", "items": { "$ref": "#/$defs/V1InferenceMappingColumnSchema" }, "title": "Columns", "type": "array" }, "datetime_col": { "default": "", "description": "Name of datetime column", "title": "Datetime Col", "type": "string" }, "predict_proba_col": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "(Classification models only) Name of column that provides the predict probability.", "title": "Predict Proba Col" } }, "required": [ "target_col", "columns" ], "title": "V1InferenceMappingSchema", "type": "object" } }, "items": { "$ref": "#/$defs/V1InferenceMappingModel" } }
- field root: List[V1InferenceMappingModel] [Required]
- pydantic model V1InferenceMappingModelsPage
Bases:
V1PageModelShow JSON schema
{ "title": "V1InferenceMappingModelsPage", "type": "object", "properties": { "items": { "default": [], "description": "Inference Mappings returned by search in page.", "items": { "$ref": "#/$defs/V1InferenceMappingModel" }, "title": "Items", "type": "array" }, "current_page": { "default": 0, "description": "Current page of items.", "title": "Current Page", "type": "integer" }, "page_size": { "default": 0, "description": "Number of items on a page.", "title": "Page Size", "type": "integer" }, "previous_page": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Number of the previous page.", "title": "Previous Page" }, "next_page": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Number of the next page.", "title": "Next Page" }, "has_previous": { "default": false, "description": "Whether or not there is a previous page.", "title": "Has Previous", "type": "boolean" }, "previous_items": { "default": 0, "title": "Previous Items", "type": "integer" }, "has_next": { "default": false, "description": "Whether or not there is a next page.", "title": "Has Next", "type": "boolean" }, "total": { "default": 0, "description": "Total number of items returned by search over all pages.", "title": "Total", "type": "integer" }, "pages": { "default": 0, "description": "Total number of pages returned by search.", "title": "Pages", "type": "integer" } }, "$defs": { "V1InferenceMappingColumnSchema": { "properties": { "feature_id": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Id of the feature, generated by the platform for optimization.", "title": "Feature Id" }, "name": { "description": "Name of column.", "title": "Name", "type": "string" }, "dtype": { "description": "Data type of column. For example, `float64`.", "title": "Dtype", "type": "string" }, "sql_type": { "description": "SQL data type of column. For example, `Float32`.", "title": "Sql Type", "type": "string" }, "feature_type": { "description": "Feature type of column. Acceptable values include `categorical` and `continuous`.", "title": "Feature Type", "type": "string" }, "segmentation": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": true, "description": "If true (the default), the column is enabled for segmentation calculations.", "title": "Segmentation" }, "drift": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": true, "description": "If true (the default), the column is enabled for drift calculations.", "title": "Drift" }, "hotspot": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": false, "description": "If true (the default), the column is enabled for hotspot analysis.", "title": "Hotspot" }, "rca": { "anyOf": [ { "items": { "type": "string" }, "type": "array" }, { "type": "null" } ], "default": [], "description": "List of column names included for root cause analysis (rca). Set by the platform based on feature configuration to identify features available for hotspot analysis.", "title": "Rca" }, "round": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "The number of decimals to round to on a continuous float column.", "title": "Round" } }, "required": [ "name", "dtype", "sql_type", "feature_type" ], "title": "V1InferenceMappingColumnSchema", "type": "object" }, "V1InferenceMappingModel": { "example": { "connection": "Internal clickhouse DB", "create_ddl": null, "deployment": "vianai.jt_placeholder_deployment", "df_schema": { "columns": [ { "drift": true, "dtype": "float64", "feature_id": 1, "feature_type": "categorical", "name": "PULocationID", "segmentation": true, "sql_type": "Float32" }, { "drift": true, "dtype": "float64", "feature_id": 2, "feature_type": "categorical", "name": "DOLocationID", "segmentation": true, "sql_type": "Float32" }, { "drift": true, "dtype": "float64", "feature_id": 3, "feature_type": "continuous", "name": "trip_distance", "segmentation": true, "sql_type": "Float32" }, { "drift": true, "dtype": "float64", "feature_id": 4, "feature_type": "continuous", "name": "travel_time", "segmentation": true, "sql_type": "Float32" }, { "drift": false, "dtype": "datetime64[ns]", "feature_id": 5, "feature_type": "unknown", "name": "PUdatetime", "sql_type": "" } ], "datetime_col": "PUdatetime", "identifier_cols": [ "pk" ], "target_col": "total_amount" }, "ground_truth_table": "16794179557100_gt", "identifier_cols_schema": null, "index": 7, "inference_table": "16794179557100_dit", "joined_table": "16794179557100_joined", "key_table": "16794179557100_key", "model_name": "vianai.jt_placeholder_deployment", "model_stage": "primary", "model_version": "1", "postprocessors": [ { "classname": "vianai.postprocessing.classification.PickInferenceMapper", "initparams": { "mapper": "{1:'a',2:'b',3:'c'}" }, "method": "fit_transform" } ] }, "properties": { "index": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Index identifier.", "title": "Index" }, "deployment": { "description": "Deployment name for the related model.", "title": "Deployment", "type": "string" }, "model_name": { "description": "Model name for the related model.", "title": "Model Name", "type": "string" }, "model_version": { "description": "Model version for the related model, e.g., 1.", "title": "Model Version", "type": "string" }, "model_stage": { "description": "Model stage for the related model, e.g., primary.", "title": "Model Stage", "type": "string" }, "connection": { "description": "Database connection for the inference mapping.", "title": "Connection", "type": "string" }, "df_schema": { "$ref": "#/$defs/V1InferenceMappingSchema", "description": "Dictionary containing dataframe schema for model, i.e., target column, identifier column(s), all columns, datetime column, and predict probability column (classification models only)." }, "identifier_cols_schema": { "anyOf": [ {}, { "type": "null" } ], "default": null, "description": "List containing identifier columns schema for model.", "title": "Identifier Cols Schema" }, "inference_table": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Name of inference table for mapping.", "title": "Inference Table" }, "ground_truth_table": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Name of ground truth table for mapping.", "title": "Ground Truth Table" }, "joined_table": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Name of joined table for mapping.", "title": "Joined Table" }, "key_table": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Name of key table for mapping.", "title": "Key Table" }, "create_ddl": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "The query to create the ddl tables.", "title": "Create Ddl" }, "postprocessors": { "anyOf": [ { "items": { "type": "object" }, "type": "array" }, { "type": "null" } ], "default": [], "description": "One or more postprocessor (and related params) to include in inference mapping, to transform output data before sending to backend database. The available postprocessors include `PickHighestProbability`, `PickInferenceMapper`, and `PickProbablityThreshold`. For each postprocessor, specify the classname, method, and initparams.", "title": "Postprocessors" }, "training_set_id": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Training set id that needs to be specified for training data preprocessing.", "title": "Training Set Id" }, "training_table": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Training table name that needs to be specified for training data preprocessing.", "title": "Training Table" } }, "required": [ "deployment", "model_name", "model_version", "model_stage", "connection", "df_schema" ], "title": "V1InferenceMappingModel", "type": "object" }, "V1InferenceMappingSchema": { "properties": { "target_col": { "description": "Name of target column.", "title": "Target Col", "type": "string" }, "identifier_cols": { "default": [], "description": "List of identifier columns.", "items": { "type": "string" }, "title": "Identifier Cols", "type": "array" }, "columns": { "description": "List of all columns.", "items": { "$ref": "#/$defs/V1InferenceMappingColumnSchema" }, "title": "Columns", "type": "array" }, "datetime_col": { "default": "", "description": "Name of datetime column", "title": "Datetime Col", "type": "string" }, "predict_proba_col": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "(Classification models only) Name of column that provides the predict probability.", "title": "Predict Proba Col" } }, "required": [ "target_col", "columns" ], "title": "V1InferenceMappingSchema", "type": "object" } }, "example": "{\"items\": [{\"index\": 1, \"deployment\": \"Credit Card Fraud vianai S\", \"model_name\": \"Credit Card Fraud vianai S\", \"model_version\": \"1\", \"model_stage\": \"primary\", \"connection\": \"clickhouse://default:@clickhouse/default\", \"df_schema\": {\"target_col\": \"tx_fraud\", \"identifier_cols\": [\"tx_id\"], \"columns\": [{\"feature_id\": 1, \"name\": \"tx_datetime\", \"dtype\": \"datetime\", \"sql_type\": \"\", \"feature_type\": \"unknown\", \"segmentation\": false, \"drift\": false, \"hotspot\": false, \"rca\": [], \"round\": null}, {\"feature_id\": 2, \"name\": \"predict_proba\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 3, \"name\": \"customer_id\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 4, \"name\": \"terminal_id\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 5, \"name\": \"tx_amount\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 6, \"name\": \"tx_fraud_scenario\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 7, \"name\": \"tx_during_weekend\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 8, \"name\": \"tx_during_night\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 9, \"name\": \"customer_id_nb_tx_1day_window\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 10, \"name\": \"customer_id_avg_amount_1day_window\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 11, \"name\": \"customer_id_nb_tx_7day_window\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 12, \"name\": \"customer_id_avg_amount_7day_window\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 13, \"name\": \"customer_id_nb_tx_30day_window\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 14, \"name\": \"customer_id_avg_amount_30day_window\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 15, \"name\": \"terminal_id_nb_tx_1day_window\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 16, \"name\": \"terminal_id_risk_1day_window\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 17, \"name\": \"terminal_id_nb_tx_7day_window\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 18, \"name\": \"terminal_id_risk_7day_window\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 19, \"name\": \"terminal_id_nb_tx_30day_window\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 20, \"name\": \"terminal_id_risk_30day_window\", \"dtype\": \"float64\", \"sql_type\": \"Float32\", \"feature_type\": \"continuous\", \"segmentation\": true, \"drift\": true, \"hotspot\": false, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 21, \"name\": \"tx_id\", \"dtype\": \"object\", \"sql_type\": \"String\", \"feature_type\": \"categorical\", \"segmentation\": true, \"drift\": true, \"hotspot\": true, \"rca\": [\"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 22, \"name\": \"transaction_cat\", \"dtype\": \"object\", \"sql_type\": \"String\", \"feature_type\": \"categorical\", \"segmentation\": true, \"drift\": true, \"hotspot\": true, \"rca\": [\"tx_id\", \"transaction_subcat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 23, \"name\": \"transaction_subcat\", \"dtype\": \"object\", \"sql_type\": \"String\", \"feature_type\": \"categorical\", \"segmentation\": true, \"drift\": true, \"hotspot\": true, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subsubcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 24, \"name\": \"transaction_subsubcat\", \"dtype\": \"object\", \"sql_type\": \"String\", \"feature_type\": \"categorical\", \"segmentation\": true, \"drift\": true, \"hotspot\": true, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"tx_fraud\"], \"round\": null}, {\"feature_id\": 25, \"name\": \"tx_fraud\", \"dtype\": \"bool\", \"sql_type\": \"Boolean\", \"feature_type\": \"categorical\", \"segmentation\": true, \"drift\": true, \"hotspot\": true, \"rca\": [\"tx_id\", \"transaction_cat\", \"transaction_subcat\", \"transaction_subsubcat\"], \"round\": null}], \"datetime_col\": \"tx_datetime\", \"predict_proba_col\": \"predict_proba\"}, \"identifier_cols_schema\": null, \"inference_table\": null, \"ground_truth_table\": null, \"joined_table\": null, \"key_table\": null, \"create_ddl\": null, \"postprocessors\": []}], \"current_page\": 1, \"page_size\": 100, \"previous_page\": null, \"next_page\": null, \"has_previous\": false, \"previous_items\": 0, \"has_next\": false, \"total\": 1, \"pages\": 1}" }
- Config:
json_schema_extra: dict = {‘example’: ‘{“items”: [{“index”: 1, “deployment”: “Credit Card Fraud vianai S”, “model_name”: “Credit Card Fraud vianai S”, “model_version”: “1”, “model_stage”: “primary”, “connection”: “clickhouse://default:@clickhouse/default”, “df_schema”: {“target_col”: “tx_fraud”, “identifier_cols”: [“tx_id”], “columns”: [{“feature_id”: 1, “name”: “tx_datetime”, “dtype”: “datetime”, “sql_type”: “”, “feature_type”: “unknown”, “segmentation”: false, “drift”: false, “hotspot”: false, “rca”: [], “round”: null}, {“feature_id”: 2, “name”: “predict_proba”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 3, “name”: “customer_id”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 4, “name”: “terminal_id”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 5, “name”: “tx_amount”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 6, “name”: “tx_fraud_scenario”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 7, “name”: “tx_during_weekend”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 8, “name”: “tx_during_night”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 9, “name”: “customer_id_nb_tx_1day_window”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 10, “name”: “customer_id_avg_amount_1day_window”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 11, “name”: “customer_id_nb_tx_7day_window”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 12, “name”: “customer_id_avg_amount_7day_window”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 13, “name”: “customer_id_nb_tx_30day_window”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 14, “name”: “customer_id_avg_amount_30day_window”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 15, “name”: “terminal_id_nb_tx_1day_window”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 16, “name”: “terminal_id_risk_1day_window”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 17, “name”: “terminal_id_nb_tx_7day_window”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 18, “name”: “terminal_id_risk_7day_window”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 19, “name”: “terminal_id_nb_tx_30day_window”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 20, “name”: “terminal_id_risk_30day_window”, “dtype”: “float64”, “sql_type”: “Float32”, “feature_type”: “continuous”, “segmentation”: true, “drift”: true, “hotspot”: false, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 21, “name”: “tx_id”, “dtype”: “object”, “sql_type”: “String”, “feature_type”: “categorical”, “segmentation”: true, “drift”: true, “hotspot”: true, “rca”: [“transaction_cat”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 22, “name”: “transaction_cat”, “dtype”: “object”, “sql_type”: “String”, “feature_type”: “categorical”, “segmentation”: true, “drift”: true, “hotspot”: true, “rca”: [“tx_id”, “transaction_subcat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 23, “name”: “transaction_subcat”, “dtype”: “object”, “sql_type”: “String”, “feature_type”: “categorical”, “segmentation”: true, “drift”: true, “hotspot”: true, “rca”: [“tx_id”, “transaction_cat”, “transaction_subsubcat”, “tx_fraud”], “round”: null}, {“feature_id”: 24, “name”: “transaction_subsubcat”, “dtype”: “object”, “sql_type”: “String”, “feature_type”: “categorical”, “segmentation”: true, “drift”: true, “hotspot”: true, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “tx_fraud”], “round”: null}, {“feature_id”: 25, “name”: “tx_fraud”, “dtype”: “bool”, “sql_type”: “Boolean”, “feature_type”: “categorical”, “segmentation”: true, “drift”: true, “hotspot”: true, “rca”: [“tx_id”, “transaction_cat”, “transaction_subcat”, “transaction_subsubcat”], “round”: null}], “datetime_col”: “tx_datetime”, “predict_proba_col”: “predict_proba”}, “identifier_cols_schema”: null, “inference_table”: null, “ground_truth_table”: null, “joined_table”: null, “key_table”: null, “create_ddl”: null, “postprocessors”: []}], “current_page”: 1, “page_size”: 100, “previous_page”: null, “next_page”: null, “has_previous”: false, “previous_items”: 0, “has_next”: false, “total”: 1, “pages”: 1}’}
- Fields:
- field items: List[V1InferenceMappingModel] = []
Inference Mappings returned by search in page.
- pydantic model V1InferenceMappingSchema
Bases:
BaseModelShow JSON schema
{ "title": "V1InferenceMappingSchema", "type": "object", "properties": { "target_col": { "description": "Name of target column.", "title": "Target Col", "type": "string" }, "identifier_cols": { "default": [], "description": "List of identifier columns.", "items": { "type": "string" }, "title": "Identifier Cols", "type": "array" }, "columns": { "description": "List of all columns.", "items": { "$ref": "#/$defs/V1InferenceMappingColumnSchema" }, "title": "Columns", "type": "array" }, "datetime_col": { "default": "", "description": "Name of datetime column", "title": "Datetime Col", "type": "string" }, "predict_proba_col": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "(Classification models only) Name of column that provides the predict probability.", "title": "Predict Proba Col" } }, "$defs": { "V1InferenceMappingColumnSchema": { "properties": { "feature_id": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Id of the feature, generated by the platform for optimization.", "title": "Feature Id" }, "name": { "description": "Name of column.", "title": "Name", "type": "string" }, "dtype": { "description": "Data type of column. For example, `float64`.", "title": "Dtype", "type": "string" }, "sql_type": { "description": "SQL data type of column. For example, `Float32`.", "title": "Sql Type", "type": "string" }, "feature_type": { "description": "Feature type of column. Acceptable values include `categorical` and `continuous`.", "title": "Feature Type", "type": "string" }, "segmentation": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": true, "description": "If true (the default), the column is enabled for segmentation calculations.", "title": "Segmentation" }, "drift": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": true, "description": "If true (the default), the column is enabled for drift calculations.", "title": "Drift" }, "hotspot": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": false, "description": "If true (the default), the column is enabled for hotspot analysis.", "title": "Hotspot" }, "rca": { "anyOf": [ { "items": { "type": "string" }, "type": "array" }, { "type": "null" } ], "default": [], "description": "List of column names included for root cause analysis (rca). Set by the platform based on feature configuration to identify features available for hotspot analysis.", "title": "Rca" }, "round": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "The number of decimals to round to on a continuous float column.", "title": "Round" } }, "required": [ "name", "dtype", "sql_type", "feature_type" ], "title": "V1InferenceMappingColumnSchema", "type": "object" } }, "required": [ "target_col", "columns" ] }
- Fields:
- Validators:
- field columns: List[V1InferenceMappingColumnSchema] [Required]
List of all columns.
- Validated by:
- field datetime_col: str = ''
Name of datetime column
- field identifier_cols: List[str] = []
List of identifier columns.
- field predict_proba_col: str | None = None
(Classification models only) Name of column that provides the predict probability.
- field target_col: str [Required]
Name of target column.
- pydantic model V1InferenceMappingSearch
Bases:
V1PageSearchShow JSON schema
{ "title": "V1InferenceMappingSearch", "type": "object", "properties": { "page": { "default": 1, "description": "Page number to return. Zero-index based. Cannot be less than zero or greater than number of pages. To make sure you don't exceed the number of pages, set `page` to 1 and run this search operation to return the total number items (`total`). Then run it again to return the exact page (`page`) for the specified number of items per page (`page-size`).", "title": "Page", "type": "integer" }, "page_size": { "default": 25, "description": "Number of items to return on a page. Default is 25 items per page.", "title": "Page Size", "type": "integer" }, "order": { "anyOf": [ { "items": { "$ref": "#/$defs/V1OrderBy" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "Fields to use for ordering returned pages and the sort direction.", "title": "Order" }, "search": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "String to search for (case-insensitive) across all fields in previously created items.", "title": "Search" }, "filters": { "anyOf": [ { "$ref": "#/$defs/V1InferenceMappingFilters" }, { "type": "null" } ], "description": "Specifies different condition(s) for filtering the results. If not provided, all relevant results are retrieved." } }, "$defs": { "V1InferenceMappingFilters": { "example": { "index": [ "1", "2" ], "model_name": [ "Example model 1" ] }, "properties": { "created_by": { "anyOf": [ { "items": { "type": "string" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "List of 'created by' users to search for.", "title": "Created By" }, "modified_by": { "anyOf": [ { "items": { "type": "string" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "List of 'modified by' users to search for.", "title": "Modified By" }, "created_ts_start": { "anyOf": [ { "type": "number" }, { "type": "null" } ], "default": null, "description": "Unix time in milliseconds for searching for all objects created during a specified time period.", "title": "Created Ts Start" }, "created_ts_end": { "anyOf": [ { "type": "number" }, { "type": "null" } ], "default": null, "description": "Unix time in milliseconds for searching for all objects created during a specified time period.", "title": "Created Ts End" }, "modified_ts_start": { "anyOf": [ { "type": "number" }, { "type": "null" } ], "default": null, "description": "Unix time in milliseconds for searching for all objects modified during a specified time period.", "title": "Modified Ts Start" }, "modified_ts_end": { "anyOf": [ { "type": "number" }, { "type": "null" } ], "default": null, "description": "Unix time in milliseconds for searching for all objects modified during a specified time period.", "title": "Modified Ts End" }, "uuid": { "anyOf": [ { "items": { "type": "string" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "List of uuids for this operation.", "title": "Uuid" }, "name": { "anyOf": [ { "items": { "type": "string" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "List of names for this operation.", "title": "Name" }, "status": { "anyOf": [ { "items": { "type": "string" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "List of statuses for this operation.", "title": "Status" }, "index": { "anyOf": [ { "items": { "type": "integer" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "Filter in addition to filters listed in V1Filters for this operation.", "title": "Index" }, "df_schema": { "anyOf": [ { "items": { "type": "object" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "One or more df_schema as a list of dict", "title": "Df Schema" } }, "title": "V1InferenceMappingFilters", "type": "object" }, "V1OrderBy": { "properties": { "field": { "description": "Field to order by.", "title": "Field", "type": "string" }, "direction": { "default": "ASC", "description": "Direction to order the field by. Supported values: `ASC` (ascending) and `DESC` (descending) (case-insensitive).", "enum": [ "ASC", "DESC" ], "title": "Direction", "type": "string" } }, "required": [ "field" ], "title": "V1OrderBy", "type": "object" } }, "example": { "filters": { "index": [ 1, 2 ], "model_name": [ "Example model 1" ] }, "order": [ { "direction": "DESC", "field": "name" } ], "page": 1, "page_size": 100, "search": "vianai" } }
- Config:
json_schema_extra: dict = {‘example’: {‘page’: 1, ‘page_size’: 100, ‘order’: [{‘field’: ‘name’, ‘direction’: ‘DESC’}], ‘search’: ‘vianai’, ‘filters’: {‘index’: [1, 2], ‘model_name’: [‘Example model 1’]}}}
- Fields:
- field filters: V1InferenceMappingFilters | None = <class 'vianops_client.models.inferencestore.inference_mapping.V1InferenceMappingFilters'>
Specifies different condition(s) for filtering the results. If not provided, all relevant results are retrieved.
vianops_client.models.inferencestore.inference_tracking
- pydantic model V1AddGroundTruthModel
Bases:
BaseModelShow JSON schema
{ "title": "V1AddGroundTruthModel", "type": "object", "properties": { "deployment": { "description": "Name of the deployment.", "title": "Deployment", "type": "string" }, "model_name": { "description": "Name of the model. Default value is deployment name.", "title": "Model Name", "type": "string" }, "model_version": { "default": "1", "description": "Model version. Default value is `1`.", "title": "Model Version", "type": "string" }, "model_stage": { "default": "primary", "description": "Model stage. Default value is `primary`.", "title": "Model Stage", "type": "string" }, "ground_truths": { "anyOf": [ { "items": { "type": "object" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "List of all ground-truth features and values.", "title": "Ground Truths" }, "format": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": "raw", "description": "The format of the data sent in. Acceptable values include `file` and `raw`. Defaults to `raw`.", "title": "Format" }, "file_type": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": "csv", "description": "The file type of the data format. Acceptable values include: `csv`, `json`, or `parquet`. Defaults to `csv`.", "title": "File Type" }, "cache_key": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "The cache key, if getting ground truth values from cache.", "title": "Cache Key" } }, "example": { "cache_key": "temp.123456", "deployment": "deployment_name", "file_type": "csv", "format": "file", "ground_truths": [ { "ground_truth": "13.8", "inf_uuid": "65_209_2022_03_01 00:00:00_0" } ], "model_name": "model_name", "model_stage": "primary", "model_version": "1" }, "required": [ "deployment", "model_name" ] }
- Config:
protected_namespaces: tuple = ()
json_schema_extra: dict = {‘example’: {‘deployment’: ‘deployment_name’, ‘model_name’: ‘model_name’, ‘model_stage’: ‘primary’, ‘model_version’: ‘1’, ‘ground_truths’: [{‘ground_truth’: ‘13.8’, ‘inf_uuid’: ‘65_209_2022_03_01 00:00:00_0’}], ‘format’: ‘file’, ‘file_type’: ‘csv’, ‘cache_key’: ‘temp.123456’}}
- Fields:
- Validators:
- field cache_key: str | None = None
The cache key, if getting ground truth values from cache.
- field deployment: str [Required]
Name of the deployment.
- field file_type: str | None = 'csv'
The file type of the data format. Acceptable values include: csv, json, or parquet. Defaults to csv.
- field format: str | None = 'raw'
The format of the data sent in. Acceptable values include file and raw. Defaults to raw.
- field ground_truths: List[dict] | None = None
List of all ground-truth features and values.
- field model_name: str [Required]
Name of the model. Default value is deployment name.
- Validated by:
- field model_stage: str = 'primary'
Model stage. Default value is primary.
- field model_version: str = '1'
Model version. Default value is 1.
- validator grouped_filters_validator » model_name
- pydantic model V1DeleteTracking
Bases:
BaseModelShow JSON schema
{ "title": "V1DeleteTracking", "type": "object", "properties": { "deployment": { "description": "Deployment name.", "title": "Deployment", "type": "string" }, "model_name": { "description": "Model name.", "title": "Model Name", "type": "string" }, "model_version": { "description": "Model version.", "title": "Model Version", "type": "string" }, "delete_tracking": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "default": false, "description": "If `true`, inference trackings are deleted from the inference tracking table. Defaults to `false`.", "title": "Delete Tracking" }, "starttime": { "anyOf": [ { "type": "number" }, { "type": "null" } ], "default": null, "description": "Inference trackings created after this time will be deleted (Unix time in milliseconds).", "title": "Starttime" }, "endtime": { "anyOf": [ { "type": "number" }, { "type": "null" } ], "default": null, "description": "Inference trackings created before this time will be deleted (Unix time in milliseconds).", "title": "Endtime" } }, "example": { "group_list": [ "MLEngineer", "SoftwareEngineer" ] }, "required": [ "deployment", "model_name", "model_version" ] }
- Config:
protected_namespaces: tuple = ()
from_attributes: bool = True
json_schema_extra: dict = {‘example’: {‘group_list’: [‘MLEngineer’, ‘SoftwareEngineer’]}}
- Fields:
- field delete_tracking: bool | None = False
If true, inference trackings are deleted from the inference tracking table. Defaults to false.
- field deployment: str [Required]
Deployment name.
- field endtime: float | None = None
Inference trackings created before this time will be deleted (Unix time in milliseconds).
- field model_name: str [Required]
Model name.
- field model_version: str [Required]
Model version.
- field starttime: float | None = None
Inference trackings created after this time will be deleted (Unix time in milliseconds).
- pydantic model V1InferenceTrackingModel
Bases:
BaseModelShow JSON schema
{ "title": "V1InferenceTrackingModel", "type": "object", "properties": { "uuid": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Inference tracking unique identifier.", "title": "Uuid" }, "deployment": { "description": "Name of the deployment.", "title": "Deployment", "type": "string" }, "model_name": { "description": "Name of the model. Default value is deployment name.", "title": "Model Name", "type": "string" }, "project_name": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Name of the Project name.", "title": "Project Name" }, "model_version": { "default": "1", "description": "Model version. Default value is `1`.", "title": "Model Version", "type": "string" }, "model_stage": { "default": "primary", "description": "Model stage. Default value is `primary`.", "title": "Model Stage", "type": "string" }, "hostname": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Hostname for the model.", "title": "Hostname" }, "starttime": { "anyOf": [ { "type": "number" }, { "format": "date-time", "type": "string" } ], "description": "Timestamp of the first inferences tracked in Unix time in milliseconds", "title": "Starttime" }, "endtime": { "anyOf": [ { "type": "number" }, { "format": "date-time", "type": "string" } ], "description": "Timestamp of the last inferences tracked in Unix time in milliseconds", "title": "Endtime" }, "num_inferences": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Total number of inferences included in this inference tracking.", "title": "Num Inferences" }, "inputs": { "anyOf": [ { "items": { "type": "object" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "List of all input features and values.", "title": "Inputs" }, "outputs": { "anyOf": [ { "items": { "anyOf": [ { "type": "string" }, { "type": "integer" }, { "type": "boolean" }, { "type": "number" } ] }, "type": "array" }, { "type": "null" } ], "default": null, "description": "List of all prediction values.", "title": "Outputs" }, "ground_truth": { "anyOf": [ { "items": { "anyOf": [ { "type": "string" }, { "type": "integer" }, { "type": "boolean" }, { "type": "number" } ] }, "type": "array" }, { "type": "null" } ], "default": null, "description": "List of all ground-truth features and values.", "title": "Ground Truth" }, "xformed_inputs": { "anyOf": [ { "items": { "type": "object" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "List of all transformed-input features and values.", "title": "Xformed Inputs" }, "modified_ts": { "anyOf": [ { "type": "number" }, { "format": "date-time", "type": "string" }, { "type": "null" } ], "default": null, "description": "Timestamp for the last time the inferences were modified in Unix time in milliseconds.", "title": "Modified Ts" }, "modified_indexes": { "anyOf": [ { "items": { "type": "integer" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "List of modified indexes.", "title": "Modified Indexes" }, "inference_mapping_index": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Inference mapping index related to the inference tracking.", "title": "Inference Mapping Index" }, "inference_mapping": { "anyOf": [ { "type": "object" }, { "type": "null" } ], "default": null, "description": "Inference mapping values related to the inference tracking.", "title": "Inference Mapping" } }, "example": { "deployment": "0234_placeholder_deployment", "endtime": "1654703247921.791", "ground_truth": [ 1 ], "hostname": "model_host", "inference_mapping": [ {} ], "inference_mapping_index": 1, "inputs": [ { "feature_a": 1.0, "feature_b": 42, "feature_c": 42, "feature_d": 1.0 } ], "model_name": "0234_placeholder_deployment", "model_stage": "primary", "model_version": "1", "modified_indexes": [ 10000, 20000, 30000 ], "modified_ts": "1654703247907.512", "num_inferences": 4, "outputs": [ 1 ], "project_name": "LLM Project", "starttime": "1654703247907.512", "uuid": "000121d1-e790-46f9-b603-c168eba631ac", "xformed_inputs": [ 1 ] }, "required": [ "deployment", "model_name", "starttime", "endtime" ] }
- Config:
protected_namespaces: tuple = ()
from_attributes: bool = True
json_schema_extra: dict = {‘example’: {‘uuid’: ‘000121d1-e790-46f9-b603-c168eba631ac’, ‘deployment’: ‘0234_placeholder_deployment’, ‘model_name’: ‘0234_placeholder_deployment’, ‘project_name’: ‘LLM Project’, ‘model_stage’: ‘primary’, ‘model_version’: ‘1’, ‘hostname’: ‘model_host’, ‘starttime’: ‘1654703247907.512’, ‘endtime’: ‘1654703247921.791’, ‘num_inferences’: 4, ‘inputs’: [{‘feature_a’: 1.0, ‘feature_b’: 42, ‘feature_c’: 42, ‘feature_d’: 1.0}], ‘outputs’: [1], ‘ground_truth’: [1], ‘xformed_inputs’: [1], ‘modified_ts’: ‘1654703247907.512’, ‘modified_indexes’: [10000, 20000, 30000], ‘inference_mapping_index’: 1, ‘inference_mapping’: [{}]}}
- Fields:
- Validators:
- field deployment: str [Required]
Name of the deployment.
- field endtime: float | datetime [Required]
Timestamp of the last inferences tracked in Unix time in milliseconds
- field ground_truth: List[str | int | bool | float] | None = None
List of all ground-truth features and values.
- field hostname: str | None = None
Hostname for the model.
- field inference_mapping: dict | None = None
Inference mapping values related to the inference tracking.
- field inference_mapping_index: int | None = None
Inference mapping index related to the inference tracking.
- field inputs: List[dict] | None = None
List of all input features and values.
- field model_name: str [Required]
Name of the model. Default value is deployment name.
- Validated by:
- field model_stage: str = 'primary'
Model stage. Default value is primary.
- field model_version: str = '1'
Model version. Default value is 1.
- field modified_indexes: List[int] | None = None
List of modified indexes.
- field modified_ts: float | datetime | None = None
Timestamp for the last time the inferences were modified in Unix time in milliseconds.
- field num_inferences: int | None = None
Total number of inferences included in this inference tracking.
- field outputs: List[str | int | bool | float] | None = None
List of all prediction values.
- field project_name: str | None = None
Name of the Project name.
- field starttime: float | datetime [Required]
Timestamp of the first inferences tracked in Unix time in milliseconds
- field uuid: str | None = None
Inference tracking unique identifier.
- field xformed_inputs: List[dict] | None = None
List of all transformed-input features and values.
- validator grouped_filters_validator » model_name
- pydantic model V1InferenceTrackingModelList
Bases:
RootModelShow JSON schema
{ "title": "V1InferenceTrackingModelList", "type": "array", "$defs": { "V1InferenceTrackingModel": { "example": { "deployment": "0234_placeholder_deployment", "endtime": "1654703247921.791", "ground_truth": [ 1 ], "hostname": "model_host", "inference_mapping": [ {} ], "inference_mapping_index": 1, "inputs": [ { "feature_a": 1.0, "feature_b": 42, "feature_c": 42, "feature_d": 1.0 } ], "model_name": "0234_placeholder_deployment", "model_stage": "primary", "model_version": "1", "modified_indexes": [ 10000, 20000, 30000 ], "modified_ts": "1654703247907.512", "num_inferences": 4, "outputs": [ 1 ], "project_name": "LLM Project", "starttime": "1654703247907.512", "uuid": "000121d1-e790-46f9-b603-c168eba631ac", "xformed_inputs": [ 1 ] }, "properties": { "uuid": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Inference tracking unique identifier.", "title": "Uuid" }, "deployment": { "description": "Name of the deployment.", "title": "Deployment", "type": "string" }, "model_name": { "description": "Name of the model. Default value is deployment name.", "title": "Model Name", "type": "string" }, "project_name": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Name of the Project name.", "title": "Project Name" }, "model_version": { "default": "1", "description": "Model version. Default value is `1`.", "title": "Model Version", "type": "string" }, "model_stage": { "default": "primary", "description": "Model stage. Default value is `primary`.", "title": "Model Stage", "type": "string" }, "hostname": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Hostname for the model.", "title": "Hostname" }, "starttime": { "anyOf": [ { "type": "number" }, { "format": "date-time", "type": "string" } ], "description": "Timestamp of the first inferences tracked in Unix time in milliseconds", "title": "Starttime" }, "endtime": { "anyOf": [ { "type": "number" }, { "format": "date-time", "type": "string" } ], "description": "Timestamp of the last inferences tracked in Unix time in milliseconds", "title": "Endtime" }, "num_inferences": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Total number of inferences included in this inference tracking.", "title": "Num Inferences" }, "inputs": { "anyOf": [ { "items": { "type": "object" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "List of all input features and values.", "title": "Inputs" }, "outputs": { "anyOf": [ { "items": { "anyOf": [ { "type": "string" }, { "type": "integer" }, { "type": "boolean" }, { "type": "number" } ] }, "type": "array" }, { "type": "null" } ], "default": null, "description": "List of all prediction values.", "title": "Outputs" }, "ground_truth": { "anyOf": [ { "items": { "anyOf": [ { "type": "string" }, { "type": "integer" }, { "type": "boolean" }, { "type": "number" } ] }, "type": "array" }, { "type": "null" } ], "default": null, "description": "List of all ground-truth features and values.", "title": "Ground Truth" }, "xformed_inputs": { "anyOf": [ { "items": { "type": "object" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "List of all transformed-input features and values.", "title": "Xformed Inputs" }, "modified_ts": { "anyOf": [ { "type": "number" }, { "format": "date-time", "type": "string" }, { "type": "null" } ], "default": null, "description": "Timestamp for the last time the inferences were modified in Unix time in milliseconds.", "title": "Modified Ts" }, "modified_indexes": { "anyOf": [ { "items": { "type": "integer" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "List of modified indexes.", "title": "Modified Indexes" }, "inference_mapping_index": { "anyOf": [ { "type": "integer" }, { "type": "null" } ], "default": null, "description": "Inference mapping index related to the inference tracking.", "title": "Inference Mapping Index" }, "inference_mapping": { "anyOf": [ { "type": "object" }, { "type": "null" } ], "default": null, "description": "Inference mapping values related to the inference tracking.", "title": "Inference Mapping" } }, "required": [ "deployment", "model_name", "starttime", "endtime" ], "title": "V1InferenceTrackingModel", "type": "object" } }, "default": [], "items": { "$ref": "#/$defs/V1InferenceTrackingModel" } }
- Fields:
- field root: List[V1InferenceTrackingModel] = []
- pydantic model V1InferenceTrackingModelStatsListGroupedByDayOrHour
Bases:
RootModelShow JSON schema
{ "title": "V1InferenceTrackingModelStatsListGroupedByDayOrHour", "type": "array", "$defs": { "V1InferenceTrackingStats": { "properties": { "deployment": { "description": "Name of the deployment.", "title": "Deployment", "type": "string" }, "date": { "description": "The inference trackings within this datetime.", "format": "date-time", "title": "Date", "type": "string" }, "hour": { "anyOf": [ { "type": "number" }, { "type": "null" } ], "description": "The inference trackings within this hour.", "title": "Hour" }, "peak_requests": { "description": "The number of inference tracking requests per minute within specified timeframe.", "title": "Peak Requests", "type": "integer" }, "predicts": { "description": "The number of predictions within timeframe.", "title": "Predicts", "type": "integer" }, "requests": { "description": "The number of requests within timeframe.", "title": "Requests", "type": "integer" }, "mean_execution_time": { "description": "The mean execution time in milliseconds (ms) within timeframe.", "title": "Mean Execution Time", "type": "number" } }, "required": [ "deployment", "date", "hour", "peak_requests", "predicts", "requests", "mean_execution_time" ], "title": "V1InferenceTrackingStats", "type": "object" } }, "items": { "$ref": "#/$defs/V1InferenceTrackingStats" } }
- Fields:
- field root: List[V1InferenceTrackingStats] [Required]
- pydantic model V1RanMetricIndexes
Bases:
BaseModelShow JSON schema
{ "title": "V1RanMetricIndexes", "type": "object", "properties": { "indexes": { "items": { "type": "integer" }, "title": "Indexes", "type": "array" } }, "required": [ "indexes" ] }
- Fields:
- field indexes: List[int] [Required]
- pydantic model V1SeldonPayload
Bases:
BaseModelShow JSON schema
{ "title": "V1SeldonPayload", "type": "object", "properties": { "seldon_headers": { "title": "Seldon Headers" }, "seldon_body": { "title": "Seldon Body" } }, "required": [ "seldon_headers", "seldon_body" ] }
- field seldon_body: Any [Required]
- field seldon_headers: Any [Required]