ref: unite initialization code of background comptuers in base class

CHANGELOG

@@ -7,6 +7,7 @@
  - fix: BackgroundSparseMed did not work for datasets of length < 100
  - fix: bad f-string in BackgroundSparseMed
  - enh: create a default basin-based output file for background computation
+ - ref: unite initialization code of background comptuers in base class
  - ref: remove functools.cache decorator from HDF5Data
  - tests: add tests for BackgroundSparseMed
 0.12.3

dcnum/feat/feat_background/base.py

@@ -2,15 +2,19 @@
 import inspect
 import multiprocessing as mp
 import pathlib
+import uuid
 
 import h5py
+import hdf5plugin
+import numpy as np
 
 from ...meta import ppid
 from ...write import create_with_basins
 
 
 class Background(abc.ABC):
-    def __init__(self, input_data, output_path, num_cpus=None, **kwargs):
+    def __init__(self, input_data, output_path, compress=True, num_cpus=None,
+                 **kwargs):
         """
 
         Parameters
@@ -24,6 +28,9 @@ def __init__(self, input_data, output_path, num_cpus=None, **kwargs):
             set `output_path` to the same path to write directly to the
             input file. The data are written in the "events/image_bg"
             dataset in the output file.
+        compress: bool
+            Whether to compress background data. Set this to False
+            for faster processing.
         num_cpus: int
             Number of CPUs to use for median computation. Defaults to
             `multiprocessing.cpu_count()`.
@@ -70,6 +77,13 @@ def __init__(self, input_data, output_path, num_cpus=None, **kwargs):
         else:
             self.input_data = input_data
 
+        #: unique identifier
+        self.name = str(uuid.uuid4())
+        #: shape of event images
+        self.image_shape = self.input_data[0].shape
+        #: total number of events
+        self.event_count = len(self.input_data)
+
         if self.h5out is None:
             if not output_path.exists():
                 # If the output path does not exist, then we create
@@ -78,10 +92,28 @@ def __init__(self, input_data, output_path, num_cpus=None, **kwargs):
                                    basin_paths=self.paths_ref)
             # TODO:
             #  - properly setup HDF5 caching
-            #  - create image_bg here instead of in subclasses
-            # "a", because output file is already an .rtdc file
+            # "a", because output file already exists
             self.h5out = h5py.File(output_path, "a", libver="latest")
 
+            # Initialize background data
+            if compress:
+                compression_kwargs = hdf5plugin.Zstd(clevel=5)
+            else:
+                compression_kwargs = {}
+            h5bg = self.h5out.require_dataset(
+                "events/image_bg",
+                shape=self.input_data.shape,
+                dtype=np.uint8,
+                chunks=(min(100, self.event_count),
+                        self.image_shape[0],
+                        self.image_shape[1]),
+                fletcher32=True,
+                **compression_kwargs,
+            )
+            h5bg.attrs.create('CLASS', np.string_('IMAGE'))
+            h5bg.attrs.create('IMAGE_VERSION', np.string_('1.2'))
+            h5bg.attrs.create('IMAGE_SUBCLASS', np.string_('IMAGE_GRAYSCALE'))
+
     @staticmethod
     def get_kwargs_from_ppid(bg_ppid):
         """Return keyword arguments for any subclass from a PPID string"""

dcnum/feat/feat_background/bg_roll_median.py

@@ -1,9 +1,7 @@
 import multiprocessing as mp
 import queue
 import time
-import uuid
 
-import hdf5plugin
 import numpy as np
 from scipy import ndimage
 
@@ -60,6 +58,7 @@ def __init__(self, input_data, output_path, kernel_size=100,
         super(BackgroundRollMed, self).__init__(
             input_data=input_data,
             output_path=output_path,
+            compress=compress,
             num_cpus=num_cpus,
             kernel_size=kernel_size,
             batch_size=batch_size)
@@ -69,12 +68,6 @@ def __init__(self, input_data, output_path, kernel_size=100,
         #: number of events processed at once
         self.batch_size = batch_size
 
-        #: unique identifier
-        self.name = str(uuid.uuid4())
-        #: shape of event images
-        self.image_shape = self.input_data[0].shape
-        #: total number of events
-        self.event_count = len(self.input_data)
         #: mp.RawArray for temporary batch input data
         self.shared_input_raw = mp_spawn.RawArray(
             np.ctypeslib.ctypes.c_uint8,
@@ -112,23 +105,6 @@ def __init__(self, input_data, output_path, kernel_size=100,
                         for _ in range(self.num_cpus)]
         [w.start() for w in self.workers]
 
-        # Initialize background data
-        if compress:
-            compression_kwargs = hdf5plugin.Zstd(clevel=5)
-        else:
-            compression_kwargs = {}
-        h5bg = self.h5out.require_dataset(
-            "events/image_bg",
-            shape=self.input_data.shape,
-            dtype=np.uint8,
-            chunks=(100, self.image_shape[0], self.image_shape[1]),
-            fletcher32=True,
-            **compression_kwargs,
-            )
-        h5bg.attrs.create('CLASS', np.string_('IMAGE'))
-        h5bg.attrs.create('IMAGE_VERSION', np.string_('1.2'))
-        h5bg.attrs.create('IMAGE_SUBCLASS', np.string_('IMAGE_GRAYSCALE'))
-
     def __enter__(self):
         return self
 

dcnum/feat/feat_background/bg_sparse_median.py

@@ -2,9 +2,7 @@
 import multiprocessing as mp
 import queue
 import time
-import uuid
 
-import hdf5plugin
 import numpy as np
 from scipy import ndimage
 
@@ -73,6 +71,7 @@ def __init__(self, input_data, output_path, kernel_size=200,
         super(BackgroundSparseMed, self).__init__(
             input_data=input_data,
             output_path=output_path,
+            compress=compress,
             num_cpus=num_cpus,
             kernel_size=kernel_size,
             split_time=split_time,
@@ -88,14 +87,6 @@ def __init__(self, input_data, output_path, kernel_size=200,
         #: keep at least this many background images from the series
         self.frac_cleansing = frac_cleansing
 
-        #: unique identifier
-        self.name = str(uuid.uuid4())
-        #: shape of event images
-        self.image_shape = self.input_data[0].shape
-
-        #: total number of events
-        self.event_count = len(self.input_data)
-
         # time axis
         self.time = None
         if self.h5in is not None:
@@ -166,25 +157,6 @@ def __init__(self, input_data, output_path, kernel_size=200,
                         for _ in range(self.num_cpus)]
         [w.start() for w in self.workers]
 
-        # Initialize background data
-        if compress:
-            compression_kwargs = hdf5plugin.Zstd(clevel=5)
-        else:
-            compression_kwargs = {}
-        h5bg = self.h5out.require_dataset(
-            "events/image_bg",
-            shape=self.input_data.shape,
-            dtype=np.uint8,
-            chunks=(min(100, self.event_count),
-                    self.image_shape[0],
-                    self.image_shape[1]),
-            fletcher32=True,
-            **compression_kwargs,
-        )
-        h5bg.attrs.create('CLASS', np.string_('IMAGE'))
-        h5bg.attrs.create('IMAGE_VERSION', np.string_('1.2'))
-        h5bg.attrs.create('IMAGE_SUBCLASS', np.string_('IMAGE_GRAYSCALE'))
-
     def __enter__(self):
         return self