Merge branch 'master' into back_to_fixed_window

straxen/analyses/posrec_comparison.py

@@ -13,7 +13,7 @@ def load_corrected_positions(
     alt_s1=False,
     alt_s2=False,
     cmt_version=None,
-    posrec_algos=("mlp"),  # TODO: https://github.com/XENONnT/straxen/issues/1454
+    posrec_algos=("mlp", "cnf"),  # TODO: https://github.com/XENONnT/straxen/issues/1454
 ):
     """Returns the corrected position for each position algorithm available, without the need to
     reprocess event_basics, as the needed information is already stored in event_basics.

straxen/corrections_services.py

@@ -15,11 +15,13 @@
 export, __all__ = strax.exporter()
 
 corrections_w_file = [
-    "mlp_model",
+    "mlp_model",  # there is no cnf_model because CNF is using another jax protocol
     "s2_xy_map_mlp",
-    "s2_xy_map",
+    "s2_xy_map_cnf",
     "s1_xyz_map_mlp",
+    "s1_xyz_map_cnf",
     "fdc_map_mlp",
+    "fdc_map_cnf",
     "s1_aft_xyz_map",
 ]
 
@@ -168,7 +170,15 @@ def _get_correction(self, run_id, correction, version):
                             df = self.interface.interpolate(df, when)
                         values.append(df.loc[df.index == when, version].values[0])
             else:
-                df = self.interface.read_at(correction, when)
+                # TODO: remove this hack when fdc_map_cnf is available
+                if correction == "fdc_map_cnf":
+                    df = self.interface.read_at("fdc_map_mlp", when)
+                elif correction == "s1_xyz_map_cnf":
+                    df = self.interface.read_at("s1_xyz_map_mlp", when)
+                elif correction == "s2_xy_map_cnf":
+                    df = self.interface.read_at("s2_xy_map_mlp", when)
+                else:
+                    df = self.interface.read_at(correction, when)
                 if df[version].isnull().values.any():
                     raise CMTnanValueError(
                         f"For {correction} there are NaN values, this means no correction available"

straxen/plugins/defaults.py

@@ -1,6 +1,6 @@
 """Some shared defaults."""
 
-DEFAULT_POSREC_ALGO = "mlp"
+DEFAULT_POSREC_ALGO = "cnf"
 
 HE_PREAMBLE = """High energy channels: attenuated signals of the top PMT-array\n"""
 

straxen/plugins/events/event_basics_vanilla.py

@@ -207,19 +207,9 @@ def _get_posrec_dtypes(self):
 
         return posrec_dtpye
 
-    @staticmethod
-    def set_nan_defaults(buffer):
-        """When constructing the dtype, take extra care to set values to np.Nan / -1 (for ints) as 0
-        might have a meaning."""
-        for field in buffer.dtype.names:
-            if np.issubdtype(buffer.dtype[field], np.integer):
-                buffer[field][:] = -1
-            else:
-                buffer[field][:] = np.nan
-
     def compute(self, events, peaks):
         result = np.zeros(len(events), dtype=self.dtype)
-        self.set_nan_defaults(result)
+        strax.set_nan_defaults(result)
 
         split_peaks = strax.split_by_containment(peaks, events)
 

straxen/plugins/events/event_nearest_triggering.py

@@ -1,6 +1,5 @@
 import numpy as np
 import strax
-import straxen
 
 export, __all__ = strax.exporter()
 
@@ -77,7 +76,7 @@ def compute(self, events, peaks):
         split_peaks = strax.split_by_containment(peaks, events)
         result = np.zeros(len(events), self.dtype)
 
-        straxen.EventBasicsVanilla.set_nan_defaults(result)
+        strax.set_nan_defaults(result)
 
         # 1. Assign peaks features to main S1 and main S2 in the event
         for event_i, (event, sp) in enumerate(zip(events, split_peaks)):

straxen/plugins/events/event_shadow.py

@@ -1,8 +1,6 @@
 import numpy as np
 import strax
 
-import straxen
-
 export, __all__ = strax.exporter()
 
 
@@ -108,7 +106,7 @@ def compute(self, events, peaks):
         split_peaks = strax.split_by_containment(peaks, events)
         result = np.zeros(len(events), self.dtype)
 
-        straxen.EventBasicsVanilla.set_nan_defaults(result)
+        strax.set_nan_defaults(result)
 
         # 1. Assign peaks features to main S1 and main S2 in the event
         for event_i, (event, sp) in enumerate(zip(events, split_peaks)):

straxen/plugins/peaks/peak_nearest_triggering.py

@@ -79,7 +79,7 @@ def compute_triggering(self, peaks, current_peak):
         _peaks = _peaks[_is_triggering]
         # init result
         result = np.zeros(len(current_peak), self.dtype)
-        straxen.EventBasicsVanilla.set_nan_defaults(result)
+        strax.set_nan_defaults(result)
 
         # use center_time as the anchor of things
         things = np.zeros(len(_peaks), dtype=strax.time_fields)

straxen/plugins/peaks/peak_per_event.py

@@ -1,6 +1,5 @@
 import numpy as np
 import strax
-import straxen
 
 export, __all__ = strax.exporter()
 
@@ -30,7 +29,7 @@ def compute(self, events, peaks):
         split_peaks = strax.split_by_containment(peaks, events)
         split_peaks_ind = strax.fully_contained_in(peaks, events)
         result = np.zeros(len(peaks), self.dtype)
-        straxen.EventBasicsVanilla.set_nan_defaults(result)
+        strax.set_nan_defaults(result)
 
         # Assign peaks features to main S1 and main S2 in the event
         for event_i, (event, sp) in enumerate(zip(events, split_peaks)):

tests/test_contexts.py

@@ -79,5 +79,5 @@ def test_cmt_versions():
     )
 
     test = unittest.TestCase()
-    # We should always work for one offline and the online version
-    test.assertTrue(len(success_for) >= 2)
+    # We should always work for the online version
+    test.assertTrue(len(success_for) >= 1)