accept the possibility of composing variants ; this will allow the po…

…ssibility of studying the effect of new theories -which might be differently architectured in some ways- in legacy datasets

validphys2/src/validphys/commondataparser.py

@@ -467,8 +467,16 @@ def check(self):
     def apply_variant(self, variant_name):
         """Return a new instance of this class with the variant applied
 
-        This class also defines how the variant is applied to the commondata
+        This class also defines how the variant is applied to the commondata.
+        If more than a variant is being used, this function will be called recursively
+        until all variants are applied.
         """
+        if not isinstance(variant_name, str):
+            observable = self
+            for single_variant in variant_name:
+                observable = observable.apply_variant(single_variant)
+            return observable
+
         try:
             variant = self.variants[variant_name]
         except KeyError as e:
@@ -484,15 +492,20 @@ def apply_variant(self, variant_name):
 
         # This section should only be used for the purposes of reproducibility
         # of legacy data, no new data should use these
-
         if variant.experiment is not None:
             new_nnpdf_metadata = dict(self._parent.nnpdf_metadata.items())
             new_nnpdf_metadata["experiment"] = variant.experiment
             setmetadata_copy = dataclasses.replace(self._parent, nnpdf_metadata=new_nnpdf_metadata)
             variant_replacement["_parent"] = setmetadata_copy
             variant_replacement["plotting"] = dataclasses.replace(self.plotting)
 
-        return dataclasses.replace(self, applied_variant=variant_name, **variant_replacement)
+        # Keep track of applied variants:
+        if self.applied_variant is None:
+            varname = variant_name
+        else:
+            varname = f"{self.applied_variant}_{variant_name}"
+
+        return dataclasses.replace(self, applied_variant=varname, **variant_replacement)
 
     @property
     def is_positivity(self):
@@ -840,7 +853,7 @@ def parse_new_metadata(metadata_file, observable_name, variant=None):
     # Select one observable from the entire metadata
     metadata = set_metadata.select_observable(observable_name)
 
-    # And apply variant if given
+    # And apply variant or variants if given
     if variant is not None:
         metadata = metadata.apply_variant(variant)
 

validphys2/src/validphys/config.py

@@ -460,14 +460,17 @@ def parse_dataset_input(self, dataset: Mapping):
         if variant is None or map_variant == "legacy_dw":
             variant = map_variant
 
+        if sysnum is not None:
+            log.warning("The key 'sys' is deprecated and will soon be removed")
+
         return DataSetInput(
             name=name,
-            sys=sysnum,
             cfac=cfac,
             frac=frac,
             weight=weight,
             custom_group=custom_group,
             variant=variant,
+            sys=sysnum,
         )
 
     def parse_use_fitcommondata(self, do_use: bool):

validphys2/src/validphys/core.py

@@ -368,17 +368,39 @@ def plot_kinlabels(self):
 
 class DataSetInput(TupleComp):
     """Represents whatever the user enters in the YAML to specify a
-    dataset."""
+    dataset.
+
+    name: str
+        name of the dataset_inputs
+    cfac: tuple
+        cfactors to apply to the final predictions (default: ())
+    frac: float
+        fraction of the data to be used during training (default: 1.0)
+    weight: float
+        extra weight to apply to the dataset (default: 1.0)
+    variant: str or tuple[str]
+        variant or variants to apply (default: None)
+    sysnum: int
+        deprecated, systematic file to load for the dataset
+    """
 
-    def __init__(self, *, name, sys, cfac, frac, weight, custom_group, variant):
+    def __init__(self, *, name, cfac, frac, weight, custom_group, variant, sys=None):
         self.name = name
         self.sys = sys
         self.cfac = cfac
         self.frac = frac
         self.weight = weight
         self.custom_group = custom_group
+
+        # Parse the variant if introduced as a string
+        if isinstance(variant, str):
+            variant = (variant,)
+
+        # Make sure that variant is not a list but, in case, a tuple
+        if isinstance(variant, list):
+            variant = tuple(variant)
         self.variant = variant
-        super().__init__(name, sys, cfac, frac, weight, custom_group, variant)
+        super().__init__(name, cfac, frac, weight, custom_group, variant, sys)
 
     def __str__(self):
         return self.name

validphys2/src/validphys/loader.py

@@ -383,7 +383,7 @@ def check_commondata(
         self, setname, sysnum=None, use_fitcommondata=False, fit=None, variant=None
     ):
         """Prepare the commondata files to be loaded.
-        A commondata is defined by its name (``setname``) and the variant (``variant``)
+        A commondata is defined by its name (``setname``) and the variant(s) (``variant``)
 
         At the moment both old-format and new-format commondata can be utilized and loaded
         however old-format commondata are deprecated and will be removed in future relases.
@@ -423,7 +423,12 @@ def check_commondata(
                         )
                         break
                     # try new commondata format
-                    old_path = fit.path / "filter" / legacy_name / f"filtered_uncertainties_{legacy_name}.yaml"
+                    old_path = (
+                        fit.path
+                        / "filter"
+                        / legacy_name
+                        / f"filtered_uncertainties_{legacy_name}.yaml"
+                    )
                     if old_path.exists():
                         data_path = old_path.with_name(f"filtered_data_{legacy_name}.yaml")
                         unc_path = old_path.with_name(f"filtered_uncertainties_{legacy_name}.yaml")
@@ -481,6 +486,7 @@ def get_commondata(self, setname, sysnum):
         """Get a Commondata from the set name and number."""
         # TODO: check where this is used
         # as this might ignore cfactors or variants
+        raise Exception("Not used")
         cd = self.check_commondata(setname, sysnum)
         return cd.load()
 

validphys2/src/validphys/overfit_metric.py

@@ -1,7 +1,7 @@
 """
 overfit_metric.py
 
-This module contains the functions used to calculate the overfit metric and 
+This module contains the functions used to calculate the overfit metric and
 produce the corresponding tables and figures.
 """
 
@@ -59,7 +59,7 @@ def calculate_chi2s_per_replica(
     preds : list[pd.core.frame.DataFrame]
         List of pandas dataframes, each containing the predictions of the pdf
         replicas for a dataset_input
-    dataset_inputs : list[DatasetInput]
+    dataset_inputs : list[DataSetInput]
     groups_covmat_no_table : pdf.core.frame.DataFrame
 
     Returns
@@ -112,10 +112,7 @@ def calculate_chi2s_per_replica(
 
 
 def array_expected_overfitting(
-    calculate_chi2s_per_replica,
-    replica_data,
-    number_of_resamples=1000,
-    resampling_fraction=0.95,
+    calculate_chi2s_per_replica, replica_data, number_of_resamples=1000, resampling_fraction=0.95
 ):
     """Calculates the expected difference in chi2 between:
     1. The chi2 of a PDF replica calculated using the corresponding pseudodata
@@ -181,10 +178,7 @@ def plot_overfitting_histogram(fit, array_expected_overfitting):
         ax.hist(array_expected_overfitting, bins=50, density=True)
         ax.axvline(x=mean, color="black")
         ax.axvline(x=0, color="black", linestyle="--")
-        xrange = [
-            array_expected_overfitting.min(),
-            array_expected_overfitting.max(),
-        ]
+        xrange = [array_expected_overfitting.min(), array_expected_overfitting.max()]
         xgrid = np.linspace(xrange[0], xrange[1], num=100)
         ax.plot(xgrid, stats.norm.pdf(xgrid, mean, std))
         ax.set_xlabel(r"$\mathcal{R}_O$")