data-designer 0.1.3__py3-none-any.whl → 0.1.5__py3-none-any.whl

data_designer/_version.py

@@ -28,7 +28,7 @@ version_tuple: VERSION_TUPLE
 commit_id: COMMIT_ID
 __commit_id__: COMMIT_ID
 
-__version__ = version = '0.1.3'
-__version_tuple__ = version_tuple = (0, 1, 3)
+__version__ = version = '0.1.5'
+__version_tuple__ = version_tuple = (0, 1, 5)
 
 __commit_id__ = commit_id = None

data_designer/config/analysis/column_profilers.py

@@ -10,16 +10,16 @@ from rich.panel import Panel
 from rich.table import Column, Table
 from typing_extensions import TypeAlias
 
-from ..base import ConfigBase
-from ..utils.visualization import ColorPalette
-from .column_statistics import (
+from data_designer.config.analysis.column_statistics import (
     CategoricalDistribution,
     CategoricalHistogramData,
     ColumnDistributionType,
     MissingValue,
     NumericalDistribution,
 )
-from .utils.reporting import TITLE_STYLE, create_judge_score_summary_table
+from data_designer.config.analysis.utils.reporting import TITLE_STYLE, create_judge_score_summary_table
+from data_designer.config.base import ConfigBase
+from data_designer.config.utils.visualization import ColorPalette
 
 
 class ColumnProfilerType(str, Enum):

data_designer/config/analysis/column_statistics.py

@@ -11,11 +11,11 @@ from pandas import Series
 from pydantic import BaseModel, ConfigDict, create_model, field_validator, model_validator
 from typing_extensions import Self, TypeAlias
 
-from ...plugin_manager import PluginManager
-from ..column_types import DataDesignerColumnType
-from ..sampler_params import SamplerType
-from ..utils.constants import EPSILON
-from ..utils.numerical_helpers import is_float, is_int, prepare_number_for_reporting
+from data_designer.config.column_types import DataDesignerColumnType
+from data_designer.config.sampler_params import SamplerType
+from data_designer.config.utils.constants import EPSILON
+from data_designer.config.utils.numerical_helpers import is_float, is_int, prepare_number_for_reporting
+from data_designer.plugin_manager import PluginManager
 
 
 class MissingValue(str, Enum):

data_designer/config/analysis/dataset_profiler.py

@@ -7,12 +7,12 @@ from typing import Annotated, Optional, Union
 
 from pydantic import BaseModel, Field, field_validator
 
-from ..column_types import DataDesignerColumnType, get_column_display_order
-from ..utils.constants import EPSILON
-from ..utils.numerical_helpers import prepare_number_for_reporting
-from .column_profilers import ColumnProfilerResultsT
-from .column_statistics import ColumnStatisticsT
-from .utils.reporting import ReportSection, generate_analysis_report
+from data_designer.config.analysis.column_profilers import ColumnProfilerResultsT
+from data_designer.config.analysis.column_statistics import ColumnStatisticsT
+from data_designer.config.analysis.utils.reporting import ReportSection, generate_analysis_report
+from data_designer.config.column_types import DataDesignerColumnType, get_column_display_order
+from data_designer.config.utils.constants import EPSILON
+from data_designer.config.utils.numerical_helpers import prepare_number_for_reporting
 
 
 class DatasetProfilerResults(BaseModel):

data_designer/config/analysis/utils/errors.py

@@ -1,7 +1,7 @@
 # SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 
-from ...errors import DataDesignerError
+from data_designer.errors import DataDesignerError
 
 
 class AnalysisReportError(DataDesignerError):

data_designer/config/analysis/utils/reporting.py

@@ -14,18 +14,18 @@ from rich.rule import Rule
 from rich.table import Column, Table
 from rich.text import Text
 
-from ...analysis.column_statistics import CategoricalHistogramData
-from ...column_types import COLUMN_TYPE_EMOJI_MAP, DataDesignerColumnType, get_column_display_order
-from ...utils.visualization import (
+from data_designer.config.analysis.column_statistics import CategoricalHistogramData
+from data_designer.config.analysis.utils.errors import AnalysisReportError
+from data_designer.config.column_types import COLUMN_TYPE_EMOJI_MAP, DataDesignerColumnType, get_column_display_order
+from data_designer.config.utils.visualization import (
     ColorPalette,
     convert_to_row_element,
     create_rich_histogram_table,
     pad_console_element,
 )
-from .errors import AnalysisReportError
 
 if TYPE_CHECKING:
-    from ...analysis.dataset_profiler import DatasetProfilerResults
+    from data_designer.config.analysis.dataset_profiler import DatasetProfilerResults
 
 HEADER_STYLE = "dim"
 RULE_STYLE = f"bold {ColorPalette.NVIDIA_GREEN.value}"

data_designer/config/base.py

@@ -6,10 +6,10 @@ from __future__ import annotations
 from pathlib import Path
 from typing import Any, Optional, Union
 
-from pydantic import BaseModel, ConfigDict
 import yaml
+from pydantic import BaseModel, ConfigDict
 
-from .utils.io_helpers import serialize_data
+from data_designer.config.utils.io_helpers import serialize_data
 
 
 class ConfigBase(BaseModel):

data_designer/config/column_configs.py

@@ -7,14 +7,14 @@ from typing import Annotated, Literal, Optional, Type, Union
 from pydantic import BaseModel, Discriminator, Field, model_validator
 from typing_extensions import Self
 
-from .base import ConfigBase
-from .errors import InvalidConfigError
-from .models import ImageContext
-from .sampler_params import SamplerParamsT, SamplerType
-from .utils.code_lang import CodeLang
-from .utils.constants import REASONING_TRACE_COLUMN_POSTFIX
-from .utils.misc import assert_valid_jinja2_template, get_prompt_template_keywords
-from .validator_params import ValidatorParamsT, ValidatorType
+from data_designer.config.base import ConfigBase
+from data_designer.config.errors import InvalidConfigError
+from data_designer.config.models import ImageContext
+from data_designer.config.sampler_params import SamplerParamsT, SamplerType
+from data_designer.config.utils.code_lang import CodeLang
+from data_designer.config.utils.constants import REASONING_TRACE_COLUMN_POSTFIX
+from data_designer.config.utils.misc import assert_valid_jinja2_template, get_prompt_template_keywords
+from data_designer.config.validator_params import ValidatorParamsT, ValidatorType
 
 
 class SingleColumnConfig(ConfigBase, ABC):

data_designer/config/column_types.py

@@ -5,8 +5,7 @@ from typing import Union
 
 from typing_extensions import TypeAlias
 
-from ..plugin_manager import PluginManager
-from .column_configs import (
+from data_designer.config.column_configs import (
     ExpressionColumnConfig,
     LLMCodeColumnConfig,
     LLMJudgeColumnConfig,
@@ -16,9 +15,14 @@ from .column_configs import (
     SeedDatasetColumnConfig,
     ValidationColumnConfig,
 )
-from .errors import InvalidColumnTypeError, InvalidConfigError
-from .sampler_params import SamplerType
-from .utils.type_helpers import SAMPLER_PARAMS, create_str_enum_from_discriminated_type_union, resolve_string_enum
+from data_designer.config.errors import InvalidColumnTypeError, InvalidConfigError
+from data_designer.config.sampler_params import SamplerType
+from data_designer.config.utils.type_helpers import (
+    SAMPLER_PARAMS,
+    create_str_enum_from_discriminated_type_union,
+    resolve_string_enum,
+)
+from data_designer.plugin_manager import PluginManager
 
 plugin_manager = PluginManager()
 

data_designer/config/config_builder.py

@@ -13,30 +13,30 @@ from pygments.formatters import HtmlFormatter
 from pygments.lexers import PythonLexer
 from typing_extensions import Self
 
-from .analysis.column_profilers import ColumnProfilerConfigT
-from .base import ExportableConfigBase
-from .column_configs import SeedDatasetColumnConfig
-from .column_types import (
+from data_designer.config.analysis.column_profilers import ColumnProfilerConfigT
+from data_designer.config.base import ExportableConfigBase
+from data_designer.config.column_configs import SeedDatasetColumnConfig
+from data_designer.config.column_types import (
     ColumnConfigT,
     DataDesignerColumnType,
     column_type_is_llm_generated,
     get_column_config_from_kwargs,
     get_column_display_order,
 )
-from .data_designer_config import DataDesignerConfig
-from .dataset_builders import BuildStage
-from .datastore import DatastoreSettings, fetch_seed_dataset_column_names
-from .default_model_settings import get_default_model_configs, resolve_seed_default_model_settings
-from .errors import BuilderConfigurationError, InvalidColumnTypeError, InvalidConfigError
-from .models import ModelConfig, load_model_configs
-from .processors import ProcessorConfig, ProcessorType, get_processor_config_from_kwargs
-from .sampler_constraints import (
+from data_designer.config.data_designer_config import DataDesignerConfig
+from data_designer.config.dataset_builders import BuildStage
+from data_designer.config.datastore import DatastoreSettings, fetch_seed_dataset_column_names
+from data_designer.config.default_model_settings import get_default_model_configs
+from data_designer.config.errors import BuilderConfigurationError, InvalidColumnTypeError, InvalidConfigError
+from data_designer.config.models import ModelConfig, load_model_configs
+from data_designer.config.processors import ProcessorConfig, ProcessorType, get_processor_config_from_kwargs
+from data_designer.config.sampler_constraints import (
     ColumnConstraintT,
     ColumnInequalityConstraint,
     ConstraintType,
     ScalarInequalityConstraint,
 )
-from .seed import (
+from data_designer.config.seed import (
     DatastoreSeedDatasetReference,
     IndexRange,
     LocalSeedDatasetReference,
@@ -45,19 +45,15 @@ from .seed import (
     SeedConfig,
     SeedDatasetReference,
 )
-from .utils.constants import DEFAULT_REPR_HTML_STYLE, REPR_HTML_TEMPLATE
-from .utils.info import ConfigBuilderInfo
-from .utils.io_helpers import serialize_data, smart_load_yaml
-from .utils.misc import can_run_data_designer_locally, json_indent_list_of_strings, kebab_to_snake
-from .utils.type_helpers import resolve_string_enum
-from .utils.validation import ViolationLevel, rich_print_violations, validate_data_designer_config
+from data_designer.config.utils.constants import DEFAULT_REPR_HTML_STYLE, REPR_HTML_TEMPLATE
+from data_designer.config.utils.info import ConfigBuilderInfo
+from data_designer.config.utils.io_helpers import serialize_data, smart_load_yaml
+from data_designer.config.utils.misc import can_run_data_designer_locally, json_indent_list_of_strings, kebab_to_snake
+from data_designer.config.utils.type_helpers import resolve_string_enum
+from data_designer.config.utils.validation import ViolationLevel, rich_print_violations, validate_data_designer_config
 
 logger = logging.getLogger(__name__)
 
-# Resolve default model settings on import to ensure they are available when the library is used.
-if can_run_data_designer_locally():
-    resolve_seed_default_model_settings()
-
 
 class BuilderConfig(ExportableConfigBase):
     """Configuration container for Data Designer builder.
@@ -143,11 +139,8 @@ class DataDesignerConfigBuilder:
                 - A list of ModelConfig objects
                 - A string or Path to a model configuration file
         """
-        if not can_run_data_designer_locally() and (model_configs is None or len(model_configs) == 0):
-            raise BuilderConfigurationError("🛑 Model configurations are required!")
-
         self._column_configs = {}
-        self._model_configs = load_model_configs(model_configs or get_default_model_configs())
+        self._model_configs = _load_model_configs(model_configs)
         self._processor_configs: list[ProcessorConfig] = []
         self._seed_config: Optional[SeedConfig] = None
         self._constraints: list[ColumnConstraintT] = []
@@ -658,3 +651,15 @@ class DataDesignerConfigBuilder:
         highlighted_html = highlight(repr_string, PythonLexer(), formatter)
         css = formatter.get_style_defs(".code")
         return REPR_HTML_TEMPLATE.format(css=css, highlighted_html=highlighted_html)
+
+
+def _load_model_configs(model_configs: Optional[Union[list[ModelConfig], str, Path]] = None) -> list[ModelConfig]:
+    """Resolves the provided model_configs, which may be a string or Path to a model configuration file.
+    If None or empty, returns default model configurations if possible, otherwise raises an error.
+    """
+    if model_configs:
+        return load_model_configs(model_configs)
+    elif can_run_data_designer_locally():
+        return get_default_model_configs()
+    else:
+        raise BuilderConfigurationError("🛑 Model configurations are required!")

data_designer/config/data_designer_config.py

@@ -7,13 +7,13 @@ from typing import Annotated, Optional
 
 from pydantic import Field
 
-from .analysis.column_profilers import ColumnProfilerConfigT
-from .base import ExportableConfigBase
-from .column_types import ColumnConfigT
-from .models import ModelConfig
-from .processors import ProcessorConfig
-from .sampler_constraints import ColumnConstraintT
-from .seed import SeedConfig
+from data_designer.config.analysis.column_profilers import ColumnProfilerConfigT
+from data_designer.config.base import ExportableConfigBase
+from data_designer.config.column_types import ColumnConfigT
+from data_designer.config.models import ModelConfig
+from data_designer.config.processors import ProcessorConfig
+from data_designer.config.sampler_constraints import ColumnConstraintT
+from data_designer.config.seed import SeedConfig
 
 
 class DataDesignerConfig(ExportableConfigBase):

data_designer/config/datastore.py

@@ -7,16 +7,16 @@ import logging
 from pathlib import Path
 from typing import TYPE_CHECKING, Optional, Union
 
-from huggingface_hub import HfApi, HfFileSystem
 import pandas as pd
 import pyarrow.parquet as pq
+from huggingface_hub import HfApi, HfFileSystem
 from pydantic import BaseModel, Field
 
-from .errors import InvalidConfigError, InvalidFileFormatError, InvalidFilePathError
-from .utils.io_helpers import VALID_DATASET_FILE_EXTENSIONS, validate_path_contains_files_of_type
+from data_designer.config.errors import InvalidConfigError, InvalidFileFormatError, InvalidFilePathError
+from data_designer.config.utils.io_helpers import VALID_DATASET_FILE_EXTENSIONS, validate_path_contains_files_of_type
 
 if TYPE_CHECKING:
-    from .seed import SeedDatasetReference
+    from data_designer.config.seed import SeedDatasetReference
 
 logger = logging.getLogger(__name__)
 

data_designer/config/default_model_settings.py

@@ -2,21 +2,21 @@
 # SPDX-License-Identifier: Apache-2.0
 
 
-from functools import lru_cache
 import logging
 import os
+from functools import lru_cache
 from pathlib import Path
 from typing import Any, Literal, Optional
 
-from .models import InferenceParameters, ModelConfig, ModelProvider
-from .utils.constants import (
+from data_designer.config.models import InferenceParameters, ModelConfig, ModelProvider
+from data_designer.config.utils.constants import (
     MANAGED_ASSETS_PATH,
     MODEL_CONFIGS_FILE_PATH,
     MODEL_PROVIDERS_FILE_PATH,
     PREDEFINED_PROVIDERS,
     PREDEFINED_PROVIDERS_MODEL_MAP,
 )
-from .utils.io_helpers import load_config_file, save_config_file
+from data_designer.config.utils.io_helpers import load_config_file, save_config_file
 
 logger = logging.getLogger(__name__)
 

data_designer/config/errors.py

@@ -1,7 +1,7 @@
 # SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 
-from ..errors import DataDesignerError
+from data_designer.errors import DataDesignerError
 
 
 class BuilderConfigurationError(DataDesignerError): ...

data_designer/config/exports.py

@@ -0,0 +1,133 @@
+# SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+
+from data_designer.config.analysis.column_profilers import JudgeScoreProfilerConfig
+from data_designer.config.column_configs import (
+    ExpressionColumnConfig,
+    LLMCodeColumnConfig,
+    LLMJudgeColumnConfig,
+    LLMStructuredColumnConfig,
+    LLMTextColumnConfig,
+    SamplerColumnConfig,
+    Score,
+    SeedDatasetColumnConfig,
+    ValidationColumnConfig,
+)
+from data_designer.config.column_types import DataDesignerColumnType
+from data_designer.config.config_builder import DataDesignerConfigBuilder
+from data_designer.config.data_designer_config import DataDesignerConfig
+from data_designer.config.dataset_builders import BuildStage
+from data_designer.config.datastore import DatastoreSettings
+from data_designer.config.models import (
+    ImageContext,
+    ImageFormat,
+    InferenceParameters,
+    ManualDistribution,
+    ManualDistributionParams,
+    Modality,
+    ModalityContext,
+    ModalityDataType,
+    ModelConfig,
+    ModelProvider,
+    UniformDistribution,
+    UniformDistributionParams,
+)
+from data_designer.config.processors import (
+    DropColumnsProcessorConfig,
+    ProcessorType,
+    SchemaTransformProcessorConfig,
+)
+from data_designer.config.sampler_constraints import ColumnInequalityConstraint, ScalarInequalityConstraint
+from data_designer.config.sampler_params import (
+    BernoulliMixtureSamplerParams,
+    BernoulliSamplerParams,
+    BinomialSamplerParams,
+    CategorySamplerParams,
+    DatetimeSamplerParams,
+    GaussianSamplerParams,
+    PersonFromFakerSamplerParams,
+    PersonSamplerParams,
+    PoissonSamplerParams,
+    SamplerType,
+    ScipySamplerParams,
+    SubcategorySamplerParams,
+    TimeDeltaSamplerParams,
+    UniformSamplerParams,
+    UUIDSamplerParams,
+)
+from data_designer.config.seed import (
+    DatastoreSeedDatasetReference,
+    IndexRange,
+    PartitionBlock,
+    SamplingStrategy,
+    SeedConfig,
+)
+from data_designer.config.utils.code_lang import CodeLang
+from data_designer.config.utils.info import InfoType
+from data_designer.config.validator_params import (
+    CodeValidatorParams,
+    RemoteValidatorParams,
+    ValidatorType,
+)
+
+
+def get_config_exports() -> list[str]:
+    return [
+        SchemaTransformProcessorConfig.__name__,
+        BernoulliMixtureSamplerParams.__name__,
+        BernoulliSamplerParams.__name__,
+        BinomialSamplerParams.__name__,
+        CategorySamplerParams.__name__,
+        CodeLang.__name__,
+        CodeValidatorParams.__name__,
+        ColumnInequalityConstraint.__name__,
+        DataDesignerColumnType.__name__,
+        DataDesignerConfig.__name__,
+        DataDesignerConfigBuilder.__name__,
+        BuildStage.__name__,
+        DatastoreSeedDatasetReference.__name__,
+        DatastoreSettings.__name__,
+        DatetimeSamplerParams.__name__,
+        DropColumnsProcessorConfig.__name__,
+        ExpressionColumnConfig.__name__,
+        GaussianSamplerParams.__name__,
+        IndexRange.__name__,
+        InfoType.__name__,
+        ImageContext.__name__,
+        ImageFormat.__name__,
+        InferenceParameters.__name__,
+        JudgeScoreProfilerConfig.__name__,
+        LLMCodeColumnConfig.__name__,
+        LLMJudgeColumnConfig.__name__,
+        LLMStructuredColumnConfig.__name__,
+        LLMTextColumnConfig.__name__,
+        ManualDistribution.__name__,
+        ManualDistributionParams.__name__,
+        Modality.__name__,
+        ModalityContext.__name__,
+        ModalityDataType.__name__,
+        ModelConfig.__name__,
+        ModelProvider.__name__,
+        PartitionBlock.__name__,
+        PersonSamplerParams.__name__,
+        PersonFromFakerSamplerParams.__name__,
+        PoissonSamplerParams.__name__,
+        ProcessorType.__name__,
+        RemoteValidatorParams.__name__,
+        SamplerColumnConfig.__name__,
+        SamplerType.__name__,
+        SamplingStrategy.__name__,
+        ScalarInequalityConstraint.__name__,
+        ScipySamplerParams.__name__,
+        Score.__name__,
+        SeedConfig.__name__,
+        SeedDatasetColumnConfig.__name__,
+        SubcategorySamplerParams.__name__,
+        TimeDeltaSamplerParams.__name__,
+        UniformDistribution.__name__,
+        UniformDistributionParams.__name__,
+        UniformSamplerParams.__name__,
+        UUIDSamplerParams.__name__,
+        ValidationColumnConfig.__name__,
+        ValidatorType.__name__,
+    ]

data_designer/config/interface.py

@@ -8,14 +8,14 @@ from typing import TYPE_CHECKING, Generic, Protocol, TypeVar
 
 import pandas as pd
 
-from .models import ModelConfig, ModelProvider
-from .utils.constants import DEFAULT_NUM_RECORDS
-from .utils.info import InterfaceInfo
+from data_designer.config.models import ModelConfig, ModelProvider
+from data_designer.config.utils.constants import DEFAULT_NUM_RECORDS
+from data_designer.config.utils.info import InterfaceInfo
 
 if TYPE_CHECKING:
-    from .analysis.dataset_profiler import DatasetProfilerResults
-    from .config_builder import DataDesignerConfigBuilder
-    from .preview_results import PreviewResults
+    from data_designer.config.analysis.dataset_profiler import DatasetProfilerResults
+    from data_designer.config.config_builder import DataDesignerConfigBuilder
+    from data_designer.config.preview_results import PreviewResults
 
 
 class ResultsProtocol(Protocol):