google-meridian 1.1.4__py3-none-any.whl → 1.1.6__py3-none-any.whl

{google_meridian-1.1.4.dist-info → google_meridian-1.1.6.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: google-meridian
-Version: 1.1.4
+Version: 1.1.6
 Summary: Google's open source mixed marketing model library, helps you understand your return on investment and direct your ad spend with confidence. 
 Author-email: The Meridian Authors <no-reply@google.com>
 License: 
@@ -397,7 +397,7 @@ To cite this repository:
   author = {Google Meridian Marketing Mix Modeling Team},
   title = {Meridian: Marketing Mix Modeling},
   url = {https://github.com/google/meridian},
-  version = {1.1.4},
+  version = {1.1.6},
   year = {2025},
 }
 ```

{google_meridian-1.1.4.dist-info → google_meridian-1.1.6.dist-info}/RECORD

@@ -1,7 +1,7 @@
-google_meridian-1.1.4.dist-info/licenses/LICENSE,sha256=WNHhf_5RCaeuKWyq_K39vmp9F28LxKsB4SpomwSZ2L0,11357
+google_meridian-1.1.6.dist-info/licenses/LICENSE,sha256=WNHhf_5RCaeuKWyq_K39vmp9F28LxKsB4SpomwSZ2L0,11357
 meridian/__init__.py,sha256=XROKwHNVQvEa371QCXAHik5wN_YKObOdJQX9bJ2c4M4,832
-meridian/constants.py,sha256=U-XNPHsEGBOySsJI8EYnSp7J7bQ_uEIrilFwXO1StZs,17223
-meridian/version.py,sha256=o6Skp0JcQqMgRgK6jJoh12OYsXy0ncCvRZr1jECnVbE,644
+meridian/constants.py,sha256=YE5h3qKH8e2lI3d9vWxc5TsSHUm5bHcz1Lq-2LurJnw,17204
+meridian/version.py,sha256=Po1LKdWVufT1_bpzSPvWjGDDYioK0kiMOjW4fb3VUyM,644
 meridian/analysis/__init__.py,sha256=nGBYz7k9FVdadO_WVGMKJcfq7Yy_TuuP8zgee4i9pSA,836
 meridian/analysis/analyzer.py,sha256=L7XyCTd4e_Bqfi8a0bW1WaXjH2ZvSVTPs0VP12a209c,206559
 meridian/analysis/formatter.py,sha256=ENIdR1CRiaVqIGEXx1HcnsA4ewgDD_nhsYCweJAThaw,7270
@@ -21,7 +21,7 @@ meridian/analysis/templates/summary.html.jinja,sha256=LuENVDHYIpNo4pzloYaCR2K9XN
 meridian/analysis/templates/table.html.jinja,sha256=mvLMZx92RcD2JAS2w2eZtfYG-6WdfwYVo7pM8TbHp4g,1176
 meridian/data/__init__.py,sha256=StIe-wfYnnbfUbKtZHwnAQcRQUS8XCZk_PCaEzw90Ww,929
 meridian/data/arg_builder.py,sha256=Kqlt88bOqFj6D3xNwvWo4MBwNwcDFHzd-wMfEOmLoPU,3741
-meridian/data/data_frame_input_data_builder.py,sha256=1upb0gfEmU-E8GX2C60NABRNEE8_iIPHARwF4OPnbEk,23195
+meridian/data/data_frame_input_data_builder.py,sha256=_hexZMFAuAowgo6FaOGElHSFHqhGnHQwEEBcwnT3zUE,27295
 meridian/data/input_data.py,sha256=teJPKTBfW-AzBWgf_fEO_S_Z1J_veqQkCvctINaid6I,39749
 meridian/data/input_data_builder.py,sha256=tbZjVXPDfmtndVyJA0fmzGzZwZb0RCEjXOTXb-ga8Nc,25648
 meridian/data/load.py,sha256=X2nmYCC-7A0RUgmdolTqCt0TD3NEZabQ5oGv-TugE00,40129
@@ -34,14 +34,14 @@ meridian/model/__init__.py,sha256=9NFfqUE5WgFc-9lQMkbfkwwV-bQIz0tsQ_3Jyq0A4SU,98
 meridian/model/adstock_hill.py,sha256=20A_6rbDUAADEkkHspB7JpCm5tYfYS1FQ6hJMLu21Pk,9283
 meridian/model/knots.py,sha256=KPEgnb-UdQQ4QBugOYEke-zBgEghgTmeCMoeiJ30meY,8054
 meridian/model/media.py,sha256=3BaPX8xYAFMEvf0mz3mBSCIDWViIs7M218nrCklc6Fk,14099
-meridian/model/model.py,sha256=BlLPyskHrEx5D71mUZFbNxS2VjkQgaiaE6hLKvQ5D3A,61489
+meridian/model/model.py,sha256=XxVJaJtfUnCWI6gM7hWC6yC64yXECi91r1LHP2B23SQ,61216
 meridian/model/model_test_data.py,sha256=hDDTEzm72LknW9c5E_dNsy4Mm4Tfs6AirhGf_QxykFs,15552
-meridian/model/posterior_sampler.py,sha256=K49zWTTelME2rL1JLeFAdMPzL0OwrBvyAXA3oR-kgSI,27801
-meridian/model/prior_distribution.py,sha256=IEDU1rabcmKNY8lxwbbO4OUAlMHPIMa7flM_zsu3DLM,42417
-meridian/model/prior_sampler.py,sha256=cmu6jG-bSEkYDkjVUxl3iSxrL7r-LN7a77cb2Vc0LoA,23218
+meridian/model/posterior_sampler.py,sha256=aOYMu4R1ltak3VC0scjrAPig5ExSjkpagk4pjmxKeh4,27884
+meridian/model/prior_distribution.py,sha256=1Qh7jQ2py7tdhLPDyeQzZ0doU6NhQRVaA0lGZNnOVZA,42554
+meridian/model/prior_sampler.py,sha256=by41y2g56jEeJ1cxJi_s45uaUBySgf7wtL5u7-GpVE8,23325
 meridian/model/spec.py,sha256=0HNiMQUWQpYvWYOZr1_fj2ah8tH-bEyfEjoqgBZ9Lc0,18049
 meridian/model/transformers.py,sha256=nRjzq1fQG0ypldxboM7Gqok6WSAXAS1witRXoAzeH9Q,7763
-google_meridian-1.1.4.dist-info/METADATA,sha256=_VAPyn1fgR57O8dJ8nVznDy1EQXbe5FVLdAbibd1GWU,22201
-google_meridian-1.1.4.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-google_meridian-1.1.4.dist-info/top_level.txt,sha256=nwaCebZvvU34EopTKZsjK0OMTFjVnkf4FfnBN_TAc0g,9
-google_meridian-1.1.4.dist-info/RECORD,,
+google_meridian-1.1.6.dist-info/METADATA,sha256=HV2L4mWfmMtz4hTWOEsVJYbTQ-aGa_0DeIr45ScZQEw,22201
+google_meridian-1.1.6.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+google_meridian-1.1.6.dist-info/top_level.txt,sha256=nwaCebZvvU34EopTKZsjK0OMTFjVnkf4FfnBN_TAc0g,9
+google_meridian-1.1.6.dist-info/RECORD,,

meridian/constants.py

@@ -218,9 +218,10 @@ PAID_MEDIA_ROI_PRIOR_TYPES = frozenset(
 # Represents a 1% increase in spend.
 MROI_FACTOR = 1.01
 
-NATIONAL_MODEL_SPEC_ARGS = immutabledict.immutabledict(
-    {MEDIA_EFFECTS_DIST: MEDIA_EFFECTS_NORMAL, UNIQUE_SIGMA_FOR_EACH_GEO: False}
-)
+NATIONAL_MODEL_SPEC_ARGS = immutabledict.immutabledict({
+    MEDIA_EFFECTS_DIST: MEDIA_EFFECTS_NORMAL,
+    UNIQUE_SIGMA_FOR_EACH_GEO: False,
+})
 
 NATIONAL_ANALYZER_PARAMETERS_DEFAULTS = immutabledict.immutabledict(
     {'aggregate_geos': True, 'geos_to_include': None}
@@ -231,7 +232,6 @@ NATIONAL_ANALYZER_PARAMETERS_DEFAULTS = immutabledict.immutabledict(
 CHAIN = 'chain'
 DRAW = 'draw'
 KNOTS = 'knots'
-SIGMA_DIM = 'sigma_dim'
 
 
 # Model parameters.

meridian/data/data_frame_input_data_builder.py

@@ -30,25 +30,112 @@ __all__ = [
 class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
   """Builds `InputData` from DataFrames."""
 
+  def __init__(
+      self,
+      kpi_type: str,
+      default_geo_column: str = constants.GEO,
+      default_time_column: str = constants.TIME,
+      default_media_time_column: str = constants.TIME,
+      default_population_column: str = constants.POPULATION,
+      default_kpi_column: str = constants.KPI,
+      default_revenue_per_kpi_column: str = constants.REVENUE_PER_KPI,
+  ):
+    super().__init__(kpi_type)
+
+    self._default_geo_column = default_geo_column
+    self._default_time_column = default_time_column
+    self._default_media_time_column = default_media_time_column
+    self._default_population_column = default_population_column
+    self._default_kpi_column = default_kpi_column
+    self._default_revenue_per_kpi_column = default_revenue_per_kpi_column
+
+  @property
+  def default_geo_column(self) -> str:
+    """The default geo column name for this builder to use.
+
+    This column name is used when `geo_col` is not explicitly provided to a data
+    setter method.
+
+    By default, this is `"geo"`.
+    """
+    return self._default_geo_column
+
+  @property
+  def default_time_column(self) -> str:
+    """The default time column name for this builder to use.
+
+    This column name is used when `time_col` is not explicitly provided to a
+    data setter method.
+
+    By default, this is `"time"`.
+    """
+    return self._default_time_column
+
+  @property
+  def default_media_time_column(self) -> str:
+    """The default *media* time column name for this builder to use.
+
+    This column name is used when `media_time_col` is not explicitly provided to
+    a data setter method.
+
+    By default, this is also `"time"`, since most input dataframes are likely
+    to use the same time column for both their media execution and media spend
+    data.
+    """
+    return self._default_media_time_column
+
+  @property
+  def default_population_column(self) -> str:
+    """The default population column name for this builder to use.
+
+    This column name is used when `population_col` is not explicitly provided to
+    a data setter method.
+
+    By default, this is `"population"`.
+    """
+    return self._default_population_column
+
+  @property
+  def default_kpi_column(self) -> str:
+    """The default kpi column name for this builder to use.
+
+    This column name is used when `kpi_col` is not explicitly provided to a data
+    setter method.
+
+    By default, this is `"kpi"`.
+    """
+    return self._default_kpi_column
+
+  @property
+  def default_revenue_per_kpi_column(self) -> str:
+    """The default revenue per kpi column name for this builder to use.
+
+    This column name is used when `revenue_per_kpi_col` is not explicitly
+    provided to a data setter method.
+
+    By default, this is `"revenue_per_kpi"`.
+    """
+    return self._default_revenue_per_kpi_column
+
   def with_kpi(
       self,
       df: pd.DataFrame,
-      kpi_col: str = constants.KPI,
-      time_col: str = constants.TIME,
-      geo_col: str = constants.GEO,
+      kpi_col: str | None = None,
+      time_col: str | None = None,
+      geo_col: str | None = None,
   ) -> 'DataFrameInputDataBuilder':
     """Reads KPI data from a DataFrame.
 
     Args:
       df: The DataFrame to read the KPI data from.
       kpi_col: The name of the column containing the KPI values. If not
-        provided, the default name is `kpi`.
+        provided, `self.default_kpi_column` is used.
       time_col: The name of the column containing the time coordinates. If not
-        provided, the default name is `time`.
+        provided, `self.default_time_column` is used.
       geo_col: (Optional) The name of the column containing the geo coordinates.
-        If not provided, the default name is `geo`. If the DataFrame provided
-        has no geo column, a national model data is assumed and a geo dimension
-        will be created internally with a single coordinate value
+        If not provided, `self.default_geo_column` is used. If the DataFrame
+        provided has no geo column, a national model data is assumed and a geo
+        dimension will be created internally with a single coordinate value
         `national_geo`.
 
     Returns:
@@ -56,6 +143,10 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
     """
     kpi_df = df.copy()
 
+    kpi_col = kpi_col or self.default_kpi_column
+    time_col = time_col or self.default_time_column
+    geo_col = geo_col or self.default_geo_column
+
     ### Validate ###
     self._validate_cols(kpi_df, [kpi_col, time_col], [geo_col])
     self._validate_coords(kpi_df, geo_col, time_col)
@@ -73,8 +164,8 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
       self,
       df: pd.DataFrame,
       control_cols: list[str],
-      time_col: str = constants.TIME,
-      geo_col: str = constants.GEO,
+      time_col: str | None = None,
+      geo_col: str | None = None,
   ) -> 'DataFrameInputDataBuilder':
     """Reads controls data from a DataFrame.
 
@@ -82,11 +173,11 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
       df: The DataFrame to read the controls data from.
       control_cols: The names of the columns containing the controls values.
       time_col: The name of the column containing the time coordinates. If not
-        provided, the default name is `time`.
+        provided, `self.default_time_column` is used.
       geo_col: (Optional) The name of the column containing the geo coordinates.
-        If not provided, the default name is `geo`. If the DataFrame provided
-        has no geo column, a national model data is assumed and a geo dimension
-        will be created internally with a single coordinate value
+        If not provided, `self.default_geo_column` is used. If the DataFrame
+        provided has no geo column, a national model data is assumed and a geo
+        dimension will be created internally with a single coordinate value
         `national_geo`.
 
     Returns:
@@ -98,6 +189,9 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
 
     controls_df = df.copy()
 
+    time_col = time_col or self.default_time_column
+    geo_col = geo_col or self.default_geo_column
+
     ### Validate ###
     self._validate_cols(
         controls_df,
@@ -120,19 +214,19 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
   def with_population(
       self,
       df: pd.DataFrame,
-      population_col: str = constants.POPULATION,
-      geo_col: str = constants.GEO,
+      population_col: str | None = None,
+      geo_col: str | None = None,
   ) -> 'DataFrameInputDataBuilder':
     """Reads population data from a DataFrame.
 
     Args:
       df: The DataFrame to read the population data from.
       population_col: The name of the column containing the population values.
-        If not provided, the default name is `population`.
+        If not provided, `self.default_population_column` is used.
       geo_col: (Optional) The name of the column containing the geo coordinates.
-        If not provided, the default name is `geo`. If the DataFrame provided
-        has no geo column, a national model data is assumed and a geo dimension
-        will be created internally with a single coordinate value
+        If not provided, `self.default_geo_column` is used. If the DataFrame
+        provided has no geo column, a national model data is assumed and a geo
+        dimension will be created internally with a single coordinate value
         `national_geo`.
 
     Returns:
@@ -140,6 +234,9 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
     """
     population_df = df.copy()
 
+    population_col = population_col or self.default_population_column
+    geo_col = geo_col or self.default_geo_column
+
     ### Validate ###
     self._validate_cols(population_df, [population_col], [geo_col])
     self._validate_coords(population_df, geo_col)
@@ -161,22 +258,22 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
   def with_revenue_per_kpi(
       self,
       df: pd.DataFrame,
-      revenue_per_kpi_col: str = constants.REVENUE_PER_KPI,
-      time_col: str = constants.TIME,
-      geo_col: str = constants.GEO,
+      revenue_per_kpi_col: str | None = None,
+      time_col: str | None = None,
+      geo_col: str | None = None,
   ) -> 'DataFrameInputDataBuilder':
     """Reads revenue per KPI data from a DataFrame.
 
     Args:
       df: The DataFrame to read the revenue per KPI data from.
       revenue_per_kpi_col: The name of the column containing the revenue per KPI
-        values. If not provided, the default name is `revenue_per_kpi`.
+        values. If not provided, `self.default_revenue_per_kpi_column` is used.
       time_col: The name of the column containing the time coordinates. If not
-        provided, the default name is `time`.
+        provided, `self.default_time_column` is used.
       geo_col: (Optional) The name of the column containing the geo coordinates.
-        If not provided, the default name is `geo`. If the DataFrame provided
-        has no geo column, a national model data is assumed and a geo dimension
-        will be created internally with a single coordinate value
+        If not provided, `self.default_geo_column` is used. If the DataFrame
+        provided has no geo column, a national model data is assumed and a geo
+        dimension will be created internally with a single coordinate value
         `national_geo`.
 
     Returns:
@@ -184,6 +281,12 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
     """
     revenue_per_kpi_df = df.copy()
 
+    revenue_per_kpi_col = (
+        revenue_per_kpi_col or self.default_revenue_per_kpi_column
+    )
+    time_col = time_col or self.default_time_column
+    geo_col = geo_col or self.default_geo_column
+
     ### Validate ###
     self._validate_cols(
         revenue_per_kpi_df,
@@ -213,8 +316,8 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
       media_cols: list[str],
       media_spend_cols: list[str],
       media_channels: list[str],
-      time_col: str = constants.TIME,
-      geo_col: str = constants.GEO,
+      time_col: str | None = None,
+      geo_col: str | None = None,
   ) -> 'DataFrameInputDataBuilder':
     """Reads media and media spend data from a DataFrame.
 
@@ -227,14 +330,15 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
         `media_cols` and `media_spend_cols` in length. These are also index
         mapped.
       time_col: The name of the column containing the time coordinates for media
-        spend and media time coordinates for media. If not provided, the default
-        name is `time`. Media time coordinates will be shorter than time
+        spend and media time coordinates for media. If not provided,
+        `self.default_time_column` is used. Media time coordinates are inferred
+        from the same `time_col` and are potentially shorter than time
         coordinates if media spend values are missing (NaN) for some t in
         `time`. Media time must be equal or a subset of time.
       geo_col: (Optional) The name of the column containing the geo coordinates.
-        If not provided, the default name is `geo`. If the DataFrame provided
-        has no geo column, a national model data is assumed and a geo dimension
-        will be created internally with a single coordinate value
+        If not provided, `self.default_geo_column` is used. If the DataFrame
+        provided has no geo column, a national model data is assumed and a geo
+        dimension will be created internally with a single coordinate value
         `national_geo`.
 
     Returns:
@@ -248,6 +352,9 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
 
     media_df = df.copy()
 
+    time_col = time_col or self.default_time_column
+    geo_col = geo_col or self.default_geo_column
+
     ### Validate ###
     # For a media dataframe, media and media_spend columns may be the same
     # (e.g. if using media spend as media execution value), so here we validate
@@ -290,8 +397,8 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
       frequency_cols: list[str],
       rf_spend_cols: list[str],
       rf_channels: list[str],
-      time_col: str = constants.TIME,
-      geo_col: str = constants.GEO,
+      time_col: str | None = None,
+      geo_col: str | None = None,
   ) -> 'DataFrameInputDataBuilder':
     """Reads reach, frequency, and rf spend data from a DataFrame.
 
@@ -305,13 +412,14 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
         also index mapped.
       time_col: The name of the column containing the time coordinates for rf
         spend and media time coordinates for reach and frequency. If not
-        provided, the default name is `time`. Media time coordinates will be
-        shorter than time coordinates if media spend values are missing (NaN)
-        for some t in `time`. Media time must be equal or a subset of time.
+        provided, `self.default_time_column` is used. Media time coordinates are
+        inferred from the same `time_col` and are potentially shorter than time
+        coordinates if media spend values are missing (NaN) for some t in
+        `time`. Media time must be equal or a subset of time.
       geo_col: (Optional) The name of the column containing the geo coordinates.
-        If not provided, the default name is `geo`. If the DataFrame provided
-        has no geo column, a national model data is assumed and a geo dimension
-        will be created internally with a single coordinate value
+        If not provided, `self.default_geo_column` is used. If the DataFrame
+        provided has no geo column, a national model data is assumed and a geo
+        dimension will be created internally with a single coordinate value
         `national_geo`.
 
     Returns:
@@ -331,6 +439,9 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
 
     reach_df = df.copy()
 
+    time_col = time_col or self.default_time_column
+    geo_col = geo_col or self.default_geo_column
+
     ### Validate ###
     self._validate_cols(
         reach_df,
@@ -389,8 +500,8 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
       df: pd.DataFrame,
       organic_media_cols: list[str],
       organic_media_channels: list[str] | None = None,
-      media_time_col: str = constants.MEDIA_TIME,
-      geo_col: str = constants.GEO,
+      media_time_col: str | None = None,
+      geo_col: str | None = None,
   ) -> 'DataFrameInputDataBuilder':
     """Reads organic media data from a DataFrame.
 
@@ -403,11 +514,11 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
         provided, must match `organic_media_cols` in length. This is index
         mapped.
       media_time_col: The name of the column containing the media time
-        coordinates. If not provided, the default name is `media_time`.
+        coordinates. If not provided, `self.default_media_time_column` is used.
       geo_col: (Optional) The name of the column containing the geo coordinates.
-        If not provided, the default name is `geo`. If the DataFrame provided
-        has no geo column, a national model data is assumed and a geo dimension
-        will be created internally with a single coordinate value
+        If not provided, `self.default_geo_column` is used. If the DataFrame
+        provided has no geo column, a national model data is assumed and a geo
+        dimension will be created internally with a single coordinate value
         `national_geo`.
 
     Returns:
@@ -418,6 +529,9 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
 
     organic_media_df = df.copy()
 
+    media_time_col = media_time_col or self.default_media_time_column
+    geo_col = geo_col or self.default_geo_column
+
     ### Validate ###
     if not organic_media_channels:
       organic_media_channels = organic_media_cols
@@ -456,8 +570,8 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
       organic_reach_cols: list[str],
       organic_frequency_cols: list[str],
       organic_rf_channels: list[str],
-      media_time_col: str = constants.MEDIA_TIME,
-      geo_col: str = constants.GEO,
+      media_time_col: str | None = None,
+      geo_col: str | None = None,
   ) -> 'DataFrameInputDataBuilder':
     """Reads organic reach and organic frequency data from a DataFrame.
 
@@ -471,11 +585,11 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
         match `organic_reach_cols` and `organic_frequency_cols` in length. These
         are also index mapped.
       media_time_col: The name of the column containing the media time
-        coordinates. If not provided, the default name is `media_time`.
+        coordinates. If not provided, `self.default_media_time_column` is used.
       geo_col: (Optional) The name of the column containing the geo coordinates.
-        If not provided, the default name is `geo`. If the DataFrame provided
-        has no geo column, a national model data is assumed and a geo dimension
-        will be created internally with a single coordinate value
+        If not provided, `self.default_geo_column` is used. If the DataFrame
+        provided has no geo column, a national model data is assumed and a geo
+        dimension will be created internally with a single coordinate value
         `national_geo`.
 
     Returns:
@@ -494,6 +608,9 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
 
     organic_reach_frequency_df = df.copy()
 
+    media_time_col = media_time_col or self.default_media_time_column
+    geo_col = geo_col or self.default_geo_column
+
     ### Validate ###
     self._validate_cols(
         organic_reach_frequency_df,
@@ -540,8 +657,8 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
       self,
       df: pd.DataFrame,
       non_media_treatment_cols: list[str],
-      time_col: str = constants.TIME,
-      geo_col: str = constants.GEO,
+      time_col: str | None = None,
+      geo_col: str | None = None,
   ) -> 'DataFrameInputDataBuilder':
     """Reads non-media treatments data from a DataFrame.
 
@@ -550,11 +667,11 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
       non_media_treatment_cols: The names of the columns containing the
         non-media treatments values.
       time_col: The name of the column containing the time coordinates. If not
-        provided, the default name is `time`.
+        provided, `self.default_time_column` is used.
       geo_col: (Optional) The name of the column containing the geo coordinates.
-        If not provided, the default name is `geo`. If the DataFrame provided
-        has no geo column, a national model data is assumed and a geo dimension
-        will be created internally with a single coordinate value
+        If not provided, `self.default_geo_column` is used. If the DataFrame
+        provided has no geo column, a national model data is assumed and a geo
+        dimension will be created internally with a single coordinate value
         `national_geo`.
 
     Returns:
@@ -569,6 +686,9 @@ class DataFrameInputDataBuilder(input_data_builder.InputDataBuilder):
 
     non_media_treatments_df = df.copy()
 
+    time_col = time_col or self.default_time_column
+    geo_col = geo_col or self.default_geo_column
+
     ### Validate ###
     self._validate_cols(
         non_media_treatments_df,

meridian/model/model.py

@@ -295,10 +295,6 @@ class Meridian:
   def is_national(self) -> bool:
     return self.n_geos == 1
 
-  @property
-  def _sigma_shape(self) -> int:
-    return len(self.input_data.geo) if self.unique_sigma_for_each_geo else 1
-
   @functools.cached_property
   def knot_info(self) -> knots.KnotInfo:
     return knots.get_knot_info(
@@ -389,6 +385,7 @@ class Meridian:
   @functools.cached_property
   def unique_sigma_for_each_geo(self) -> bool:
     if self.is_national:
+      # Should evaluate to False.
       return constants.NATIONAL_MODEL_SPEC_ARGS[
           constants.UNIQUE_SIGMA_FOR_EACH_GEO
       ]
@@ -449,7 +446,7 @@ class Meridian:
         n_organic_rf_channels=self.n_organic_rf_channels,
         n_controls=self.n_controls,
         n_non_media_channels=self.n_non_media_channels,
-        sigma_shape=self._sigma_shape,
+        unique_sigma_for_each_geo=self.unique_sigma_for_each_geo,
         n_knots=self.knot_info.n_knots,
         is_national=self.is_national,
         set_total_media_contribution_prior=self._set_total_media_contribution_prior,
@@ -663,10 +660,6 @@ class Meridian:
     self._validate_injected_inference_data_group_coord(
         inference_data, group, constants.TIME, self.n_times
     )
-    if not self.model_spec.unique_sigma_for_each_geo:
-      self._validate_injected_inference_data_group_coord(
-          inference_data, group, constants.SIGMA_DIM, self._sigma_shape
-      )
     self._validate_injected_inference_data_group_coord(
         inference_data,
         group,
@@ -1429,7 +1422,7 @@ class Meridian:
     if self.unique_sigma_for_each_geo:
       inference_dims[constants.SIGMA] = [constants.GEO]
     else:
-      inference_dims[constants.SIGMA] = [constants.SIGMA_DIM]
+      inference_dims[constants.SIGMA] = []
 
     return {
         param: [constants.CHAIN, constants.DRAW] + list(dims)

meridian/model/posterior_sampler.py

@@ -528,7 +528,7 @@ class PosteriorMCMCSampler:
         be a positive integer. For more information, see `tf.while_loop`.
       seed: An `int32[2]` Tensor or a Python list or tuple of 2 `int`s, which
         will be treated as stateless seeds; or a Python `int` or `None`, which
-        will be treated as stateful seeds. See [tfp.random.sanitize_seed]
+        will be converted into a stateless seed. See [tfp.random.sanitize_seed]
         (https://www.tensorflow.org/probability/api_docs/python/tfp/random/sanitize_seed).
       **pins: These are used to condition the provided joint distribution, and
         are passed directly to `joint_dist.experimental_pin(**pins)`.
@@ -547,6 +547,8 @@ class PosteriorMCMCSampler:
           " [tfp.random.sanitize_seed](https://www.tensorflow.org/probability/api_docs/python/tfp/random/sanitize_seed)"
           " for details."
       )
+    if seed is not None and isinstance(seed, int):
+      seed = (seed, seed)
     seed = tfp.random.sanitize_seed(seed) if seed is not None else None
     n_chains_list = [n_chains] if isinstance(n_chains, int) else n_chains
     total_chains = np.sum(n_chains_list)

meridian/model/prior_distribution.py

@@ -509,7 +509,7 @@ class PriorDistribution:
       n_organic_rf_channels: int,
       n_controls: int,
       n_non_media_channels: int,
-      sigma_shape: int,
+      unique_sigma_for_each_geo: bool,
       n_knots: int,
       is_national: bool,
       set_total_media_contribution_prior: bool,
@@ -527,9 +527,9 @@ class PriorDistribution:
         used.
       n_controls: Number of controls used.
       n_non_media_channels: Number of non-media channels used.
-      sigma_shape: A number describing the shape of the sigma parameter. It's
-        either `1` (if `sigma_for_each_geo=False`) or `n_geos` (if
-        `sigma_for_each_geo=True`). For more information, see `ModelSpec`.
+      unique_sigma_for_each_geo: A boolean indicator whether to use the same
+        sigma parameter for all geos. Only used if `n_geos > 1`. For more
+        information, see `ModelSpec`.
       n_knots: Number of knots used.
       is_national: A boolean indicator whether the prior distribution will be
         adapted for a national model.
@@ -801,6 +801,9 @@ class PriorDistribution:
     slope_orf = tfp.distributions.BatchBroadcast(
         self.slope_orf, n_organic_rf_channels, name=constants.SLOPE_ORF
     )
+
+    # If `unique_sigma_for_each_geo == False`, then make a scalar batch.
+    sigma_shape = n_geos if (n_geos > 1 and unique_sigma_for_each_geo) else []
     sigma = tfp.distributions.BatchBroadcast(
         self.sigma, sigma_shape, name=constants.SIGMA
     )

meridian/model/prior_sampler.py

@@ -510,6 +510,8 @@ class PriorDistributionSampler:
       tf.keras.utils.set_random_seed(1)
 
     prior = mmm.prior_broadcast
+    # `sample_shape` is prepended to the shape of each BatchBroadcast in `prior`
+    # when it is sampled.
     sample_shape = [1, n_draws]
     sample_kwargs = {constants.SAMPLE_SHAPE: sample_shape, constants.SEED: seed}
 

meridian/version.py

@@ -14,4 +14,4 @@
 
 """Module for Meridian version."""
 
-__version__ = "1.1.4"
+__version__ = "1.1.6"