Merge branch 'dev'

# Conflicts: # poetry.lock # pyproject.toml
MontpellierRessourcesImagerie · Oct 16, 2024 · 146d6fc · 146d6fc
2 parents edf5a9d + 9680ce3
commit 146d6fc
Show file tree

Hide file tree

Showing 22 changed files with 2,407 additions and 1,063 deletions.
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -10,11 +10,12 @@ repos:
         # pre-commit's default_language_version, see
         # https://pre-commit.com/#top_level-default_language_version
         language_version: python3.11
-
+        files: ^(src|test)/.*\.py$
 
   - repo: https://github.com/pycqa/isort
     rev: 5.13.2
     hooks:
       - id: isort
         args: ["--profile", "black"]
         name: isort (python)
+        files: ^(src|test)/.*\.py$
diff --git a/docs/examples/new_sample.py b/docs/examples/new_sample.py
@@ -3,14 +3,15 @@
 See official documentation at https://scikit-image.org/docs/0.7.0/api/skimage.transform.html#probabilistic-hough
 
 The procedure to follow is, in short:
-- import everything from the samples module
+- import everything from the analyses module
 - import the types that you might be using from the typing module
 - import any necessary libraries that you will need for your analysis
-- Create one or more subclasses of the Analysis abstract class of samples. Within each class:
+- Create one or more subclasses of the Analysis abstract class of analyses. Within each class:
     - define your input requirements
     - define a 'run' method that will implement the logic of your analysis
     - if desired, define a 'plot' method returning a plot showing the results of the analysis
 """
+
 from math import atan2
 
 # import the types that you may be using
@@ -25,7 +26,7 @@
 from skimage.transform import probabilistic_hough_line
 
 # import the sample functionality
-from microscopemetrics.samples import *
+from microscopemetrics.analyses import *
 
 
 class DetectLinesAnalysis(

diff --git a/docs/notebooks/microscope-metrics_tutorial.ipynb b/docs/notebooks/microscope-metrics_tutorial.ipynb
@@ -805,4 +805,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 1
-}
+}
diff --git a/poetry.lock b/poetry.lock
diff --git a/pyproject.toml b/pyproject.toml
@@ -14,7 +14,7 @@ documentation = "https://github.com/MontpellierRessourcesImagerie/microscope-met
 keywords = ["microscopy", "metrology"]
 
 [tool.pytest.ini_options]
-addopts = "--hypothesis-show-statistics --hypothesis-profile=dev --workers=auto"
+addopts = "--hypothesis-show-statistics --hypothesis-profile=dev" # --workers=auto"
 testpaths = [
     "tests"
 ]
@@ -30,13 +30,17 @@ filterwarnings = [
     "ignore:.*urllib3 v2 only supports OpenSSL.*:Warning",
 ]
 
+[tool.isort]
+profile = "black"
+
 [tool.poetry.dependencies]
-python = ">=3.10.12 <3.12"
+python = ">=3.9.18 <3.12"
 numpy = "^1"
 scikit-image = "^0"
 scipy = "^1"
 pandas = "^1"
-microscopemetrics-schema = {git = "https://github.com/juliomateoslangerak/microscopemetrics-schema.git"}
+matplotlib = "^3"
+microscopemetrics-schema = {git = "https://github.com/juliomateoslangerak/microscopemetrics-schema.git", branch = "dev"}
 
 [tool.poetry.group.test.dependencies]
 pytest = "^7"
@@ -48,6 +52,7 @@ pytest-parallel = "^0.1.1"
 [tool.poetry.group.dev.dependencies]
 black = "^23"
 poetry = "^1"
+jupyter = "^1"
 isort = "^5"
 pre-commit = "^3"
 

diff --git a/src/microscopemetrics/samples/__init__.py → src/microscopemetrics/analyses/__init__.py b/src/microscopemetrics/samples/__init__.py → src/microscopemetrics/analyses/__init__.py
@@ -1,4 +1,4 @@
-# Main samples module defining the sample superclass
+# Main analyses module defining the sample superclass
 import logging
 from abc import ABC, abstractmethod
 from datetime import datetime
@@ -8,51 +8,8 @@
 import numpy as np
 import pandas as pd
 
-# We are defining some global dictionaries to register the different analysis types
-IMAGE_ANALYSIS_REGISTRY = {}
-DATASET_ANALYSIS_REGISTRY = {}
-PROGRESSION_ANALYSIS_REGISTRY = {}
-
-
-# Decorators to register exposed analysis functions
-def register_image_analysis(cls):
-    IMAGE_ANALYSIS_REGISTRY[cls.__name__] = cls
-    return cls
-
-
-def register_dataset_analysis(cls):
-    DATASET_ANALYSIS_REGISTRY[cls.__name__] = cls
-    return cls
-
-
-def register_progression_analysis(cls):
-    PROGRESSION_ANALYSIS_REGISTRY[cls.__name__] = cls
-    return cls
-
-
 # Create a logging service
 logger = logging.getLogger(__name__)
-# TODO: work on the loggers
-
-
-# def get_references(
-#     objects: Union[mm_schema.MetricsObject, list[mm_schema.MetricsObject]]
-# ) -> List[mm_schema.DataReference]:
-#     """Get the references of a metrics object or a list of metrics objects"""
-#     if isinstance(objects, mm_schema.MetricsObject):
-#         return mm_schema.DataReference(
-#             data_uri=objects.data_uri,
-#             # HACK: This is a temporary fix to get the first element of the tuple
-#             omero_host=objects.omero_host[0],
-#             omero_port=objects.omero_port[0],
-#             omero_object_type=objects.omero_object_type[0],
-#             omero_object_id=objects.omero_object_id,
-#         )
-#
-#     elif isinstance(objects, list):
-#         return [get_references(obj) for obj in objects]
-#     else:
-#         raise ValueError("Input should be a metrics object or a list of metrics objects")
 
 
 def get_object_id(
@@ -152,72 +109,6 @@ def numpy_to_mm_image(
     )
 
 
-# def numpy_to_mask_inlined(
-#     array: np.ndarray,
-#     name: str = None,
-#     description: str = None,
-#     image_url: str = None,
-#     source_image_url: Union[str, List[str]] = None,
-# ) -> mm_schema.ImageMask:
-#     """Converts a bool numpy array with dimensions order yx to an inlined mask"""
-#     if array.ndim != 2:
-#         raise ValueError("Input array should be 2D")
-#     if array.dtype != bool and array.dtype == np.uint8:
-#         try:
-#             array = array.astype(bool)
-#         except ValueError:
-#             raise ValueError("Input array could not be casted to type bool")
-#     if array.dtype != bool:
-#         raise ValueError("Input array should be of type bool")
-#
-#     return mm_schema.ImageMask(
-#         name=name,
-#         description=description,
-#         image_url=image_url,
-#         source_image_url=source_image_url,
-#         data=array.flatten().tolist(),
-#         shape_y=array.shape[0],
-#         shape_x=array.shape[1],
-#     )
-#
-#
-# def numpy_to_image_inlined(
-#     array: np.ndarray,
-#     name: str = None,
-#     description: str = None,
-#     image_url: str = None,
-#     source_image_url: Union[str, List[str]] = None,
-# ) -> mm_schema.ImageInline:
-#     """Converts a numpy array with dimensions order tzyxc or yx to an inlined image"""
-#     if array.ndim == 5:
-#         return mm_schema.Image5D(
-#             name=name,
-#             description=description,
-#             image_url=image_url,
-#             source_image_url=source_image_url,
-#             data=array.flatten().tolist(),
-#             shape_t=array.shape[0],
-#             shape_z=array.shape[1],
-#             shape_y=array.shape[2],
-#             shape_x=array.shape[3],
-#             shape_c=array.shape[4],
-#         )
-#     elif array.ndim == 2:
-#         return mm_schema.Image2D(
-#             name=name,
-#             description=description,
-#             image_url=image_url,
-#             source_image_url=source_image_url,
-#             data=array.flatten().tolist(),
-#             shape_y=array.shape[0],
-#             shape_x=array.shape[1],
-#         )
-#     else:
-#         raise NotImplementedError(
-#             f"Array of dimension {array.ndim} is not supported by this function"
-#         )
-
-
 def _create_table(
     data: Union[dict[str, list], pd.DataFrame],
     name: str,

diff --git a/src/microscopemetrics/samples/argolight.py → src/microscopemetrics/analyses/argolight.py b/src/microscopemetrics/samples/argolight.py → src/microscopemetrics/analyses/argolight.py
@@ -1,24 +1,25 @@
 from datetime import datetime
 from itertools import product
-from typing import Any, Dict, List, Tuple, Union
+from typing import Dict, List, Tuple
 
 import microscopemetrics_schema.datamodel as mm_schema
 import numpy as np
 import pandas as pd
 from numpy import float64, int64, ndarray
 from pandas import DataFrame
-from scipy.interpolate import griddata
 from scipy.optimize import curve_fit
 from scipy.signal import find_peaks
 from skimage.transform import hough_line  # hough_line_peaks, probabilistic_hough_line
 
-from microscopemetrics.analysis.tools import (
+from microscopemetrics.analyses import logger, numpy_to_mm_image, validate_requirements
+from microscopemetrics.analyses.tools import (
+    airy_fun,
     compute_distances_matrix,
     compute_spots_properties,
+    is_saturated,
+    multi_airy_fun,
     segment_image,
 )
-from microscopemetrics.samples import logger, numpy_to_mm_image, validate_requirements
-from microscopemetrics.utilities.utilities import airy_fun, is_saturated, multi_airy_fun
 
 
 def _profile_to_columns(profile: ndarray, channel: int) -> List[Dict[str, Dict[str, List[float]]]]:
@@ -223,12 +224,12 @@ def _compute_resolution(
     )
 
 
-def analise_argolight_b(dataset: mm_schema.ArgolightBDataset) -> bool:
+def analyse_argolight_b(dataset: mm_schema.ArgolightBDataset) -> bool:
     validate_requirements()
 
     # Check image shape
     logger.info("Checking image shape...")
-    image = dataset.input.argolight_b_image.data
+    image = dataset.input_data.argolight_b_image.data
     if len(image.shape) != 5:
         logger.error("Image must be 5D")
         return False
@@ -239,8 +240,8 @@ def analise_argolight_b(dataset: mm_schema.ArgolightBDataset) -> bool:
     for c in range(image.shape[-1]):
         if is_saturated(
             channel=image[:, :, :, :, c],
-            threshold=dataset.input.saturation_threshold,
-            detector_bit_depth=dataset.input.bit_depth,
+            threshold=dataset.input_parameters.saturation_threshold,
+            detector_bit_depth=dataset.input_parameters.bit_depth,
         ):
             logger.error(f"Channel {c} is saturated")
             saturated_channels.append(c)
@@ -249,32 +250,36 @@ def analise_argolight_b(dataset: mm_schema.ArgolightBDataset) -> bool:
         return False
 
     # Calculating the distance between spots in pixels with a security margin
-    min_distance = round(dataset.input.spots_distance * 0.3)
+    min_distance = round(dataset.input_parameters.spots_distance * 0.3)
 
     # Calculating the maximum tolerated distance in microns for the same spot in a different channels
-    max_distance = dataset.input.spots_distance * 0.4
+    max_distance = dataset.input_parameters.spots_distance * 0.4
 
     labels = segment_image(
         image=image,
         min_distance=min_distance,
-        sigma=(dataset.input.sigma_z, dataset.input.sigma_y, dataset.input.sigma_x),
+        sigma=(
+            dataset.input_parameters.sigma_z,
+            dataset.input_parameters.sigma_y,
+            dataset.input_parameters.sigma_x,
+        ),
         method="local_max",
-        low_corr_factors=dataset.input.lower_threshold_correction_factors,
-        high_corr_factors=dataset.input.upper_threshold_correction_factors,
+        low_corr_factors=dataset.input_parameters.lower_threshold_correction_factors,
+        high_corr_factors=dataset.input_parameters.upper_threshold_correction_factors,
     )
 
     dataset.output.spots_labels_image = numpy_to_mm_image(  # TODO: this should be a mask
         array=labels,
-        name=f"{dataset.input.argolight_b_image.name}_spots_labels",
-        description=f"Spots labels of {dataset.input.argolight_b_image.image_url}",
-        image_url=dataset.input.argolight_b_image.image_url,
-        source_image_url=dataset.input.argolight_b_image.image_url,
+        name=f"{dataset.input_data.argolight_b_image.name}_spots_labels",
+        description=f"Spots labels of {dataset.input_data.argolight_b_image.image_url}",
+        image_url=dataset.input_data.argolight_b_image.image_url,
+        source_image_url=dataset.input_data.argolight_b_image.image_url,
     )
 
     spots_properties, spots_positions = compute_spots_properties(
         image=image,
         labels=labels,
-        remove_center_cross=dataset.input.remove_center_cross,
+        remove_center_cross=dataset.input_parameters.remove_center_cross,
     )
 
     distances_df = compute_distances_matrix(
@@ -338,7 +343,7 @@ def analise_argolight_b(dataset: mm_schema.ArgolightBDataset) -> bool:
         spots_centroids.append(
             mm_schema.Roi(
                 label=f"Centroids_ch{ch:02}",
-                image=dataset.input.argolight_b_image.image_url,
+                image=dataset.input_data.argolight_b_image.image_url,
                 shapes=channel_shapes,
             )
         )
@@ -398,7 +403,7 @@ def analise_argolight_b(dataset: mm_schema.ArgolightBDataset) -> bool:
     return True
 
 
-def analise_argolight_e(dataset: mm_schema.ArgolightEDataset) -> bool:
+def analyse_argolight_e(dataset: mm_schema.ArgolightEDataset) -> bool:
     validate_requirements()
 
     # Check image shape
@@ -410,11 +415,11 @@ def analise_argolight_e(dataset: mm_schema.ArgolightEDataset) -> bool:
     # Check for axis value
     pass  # TODO: implement
 
-    image = dataset.input.argolight_e_image.data
+    image = dataset.input_data.argolight_e_image.data
     image = image[0]  # if there is a time dimension, take the first one
-    axis = dataset.input.orientation_axis
-    pixel_size = dataset.input.argolight_e_image.voxel_size_x_micron
-    measured_band = dataset.input.measured_band
+    axis = dataset.input_parameters.orientation_axis
+    pixel_size = dataset.input_data.argolight_e_image.voxel_size_x_micron
+    measured_band = dataset.input_parameters.measured_band
 
     (
         profiles,
@@ -428,7 +433,7 @@ def analise_argolight_e(dataset: mm_schema.ArgolightEDataset) -> bool:
         image=image,
         axis=axis,
         measured_band=measured_band,
-        prominence=dataset.input.prominence_threshold,
+        prominence=dataset.input_parameters.prominence_threshold,
         do_angle_refinement=False,  # TODO: implement angle refinement
     )
     key_values = {
@@ -498,7 +503,7 @@ def analise_argolight_e(dataset: mm_schema.ArgolightEDataset) -> bool:
         rois.append(
             mm_schema.Roi(
                 label=f"ch{ch:02}_peaks",
-                image=dataset.input.argolight_e_image.image_url,
+                image=dataset.input_data.argolight_e_image.image_url,
                 shapes=shapes,
             )
         )
-Original file line number
+Diff line change
@@ Expand Up / @@ -805,4 +805,4 @@ @@
      },
      "nbformat": 4,
      "nbformat_minor": 1
-    }
+    }