WIP csv+docs + make_label + lint

Author: C-Achard

docs/res/code/utils.rst

@@ -13,6 +13,10 @@ get_time
 **************************************
 .. autofunction:: napari_cellseg3d.utils::get_time
 
+time_difference
+**************************************
+.. autofunction:: napari_cellseg3d.utils::time_difference
+
 get_time_filepath
 **************************************
 .. autofunction:: napari_cellseg3d.utils::get_time_filepath

docs/res/guides/inference_module_guide.rst

@@ -31,38 +31,44 @@ Interface and functionalities
 
 * **Loading data** :
 
-  | When launching the module, you will be asked to provide an image folder containing all the volumes you'd like to be labeled.
-  | All images with the chosen (**.tif** or **.tiff** currently supported) extension in this folder will be labeled.
-  | You can then choose an output folder, where all the results will be saved.
+  | When launching the module, you will be asked to provide an **image folder** containing all the volumes you'd like to be labeled.
+  | All images with the chosen extension (**.tif** or **.tiff** currently supported) in this folder will be labeled.
+  | You can then choose an **output folder**, where all the results will be saved.
 
 
 * **Model choice** :
 
-  | You can then choose one of the provided models above, which will be used for inference.
-  | You may also choose to load custom weights rather than the pre-trained ones, simply ensure they are compatible (e.g. produced from the training module for the same model)
+  | You can then choose one of the provided **models** above, which will be used for inference.
+  | You may also choose to **load custom weights** rather than the pre-trained ones, simply ensure they are **compatible** (e.g. produced from the training module for the same model)
 
 
 * **Anisotropy** :
 
-  | If you want to see your results without anisotropy when you have anisotropic images, you can specify that you have anisotropic data
-  | and set the resolution of your image in micron, this wil save and show the results without anisotropy.
+  | If you want to see your results without **anisotropy** when you have anisotropic images, you can specify that you have anisotropic data
+  | and set the **resolution of your imaging method in micron**, this wil save and show the results without anisotropy.
 
 
 * **Thresholding** :
 
-  | You can perform thresholding to binarize your labels, all values beneath the confidence threshold will be set to 0 using this.
-  | If you wish to use instance segmentation it is recommended to use thresholding.
+  | You can perform thresholding to **binarize your labels**,
+  | all values beneath the **confidence threshold** will be set to 0 using this.
 
 * **Instance segmentation** :
 
-  | You can convert the semantic segmentation into instance labels by using either the watershed or connected components method.
+  | You can convert the semantic segmentation into instance labels by using either the `watershed`_ or `connected components`_ method.
   | You can set the probability threshold from which a pixel is considered as a valid instance, as well as the minimum size in pixels for objects. All smaller objects will be removed.
   | Instance labels will be saved (and shown if applicable) separately from other results.
 
+
+.. _watershed: https://scikit-image.org/docs/dev/auto_examples/segmentation/plot_watershed.html
+.. _connected components: https://scikit-image.org/docs/dev/api/skimage.measure.html#skimage.measure.label
+
+
 * **Viewing results** :
 
-  | You can also select whether you'd like to see the results in napari afterwards.
-  | By default the first image processed will be displayed, but you can choose to display up to ten at once. You can also request to see the originals.
+  | You can also select whether you'd like to **see the results** in napari afterwards.
+  | By default the first image processed will be displayed, but you can choose to display up to **ten at once**.
+  | You can also request to see the originals.
 
 
 When you are done choosing your parameters, you can press the **Start** button to begin the inference process.

docs/res/guides/training_module_guide.rst

@@ -39,6 +39,11 @@ The training module is comprised of several tabs.
 * The path to the labels folder
 * The path to the results folder
 
+* Whether to copy results to a zip file (for easier transferability)
+
+* Whether to use pre-trained weights that are provided; if you choose to do so, the model will be initialized with the specified weights, possibly improving performance (transfer learning).
+  You can also load custom weights; simply ensure they are compatible with the model.
+
 2) The second tab, **Augmentation**, lets you define dataset and augmentation parameters such as :
 
 * Whether to use images "as is" (**requires all images to be of the same size and cubic**) or extract patches.
@@ -77,6 +82,8 @@ Generalized Dice loss     `Generalized dice Loss from MONAI`_ with ``sigmoid=tru
 Dice-CE loss              `Dice-CE Loss from MONAI`_ with ``sigmoid=true``
 Tversky loss              `Tversky Loss from MONAI`_ with ``sigmoid=true``
 ========================  ================================================================================================
+
+
 .. _Dice Loss from MONAI: https://docs.monai.io/en/stable/losses.html#diceloss
 .. _Focal Loss from MONAI: https://docs.monai.io/en/stable/losses.html#focalloss
 .. _Dice-focal Loss from MONAI: https://docs.monai.io/en/stable/losses.html#dicefocalloss

napari_cellseg3d/_tests/test_utils.py

@@ -8,6 +8,33 @@
 from napari_cellseg3d import utils
 
 
+def test_fill_list_in_between():
+
+    list = [1, 2, 3, 4, 5, 6]
+    res = [
+        1,
+        "",
+        "",
+        2,
+        "",
+        "",
+        3,
+        "",
+        "",
+        4,
+        "",
+        "",
+        5,
+        "",
+        "",
+        6,
+        "",
+        "",
+    ]
+
+    assert utils.fill_list_in_between(list, 2, "") == res
+
+
 def test_align_array_sizes():
 
     im = np.zeros((128, 512, 256))

napari_cellseg3d/interface.py

@@ -91,7 +91,7 @@ def open_file_dialog(
         return filenames
 
 
-def make_label(name, parent):
+def make_label(name, parent=None):
     """Creates a QLabel
 
     Args:
@@ -101,7 +101,10 @@ def make_label(name, parent):
     Returns: created label
 
     """
-    return QLabel(name, parent)
+    if parent is not None:
+        return QLabel(name, parent)
+    else:
+        return QLabel(name)
 
 
 def make_scrollable(
@@ -302,7 +305,7 @@ def make_combobox(
     Args:
         entries array(str): Entries to add to the dropdown menu. Defaults to None, no entries if None
         parent (QWidget): parent QWidget to add dropdown menu to. Defaults to None, no parent is set if None
-        label (str) : if not None, creates a Qlabel with the contents of 'label', and returns the label as well
+        label (str) : if not None, creates a QLabel with the contents of 'label', and returns the label as well
         fixed (bool): if True, will set the size policy of the dropdown menu to Fixed in h and w. Defaults to True.
 
     Returns:

napari_cellseg3d/launch_review.py

@@ -1,4 +1,3 @@
-import os
 from pathlib import Path
 
 import matplotlib.pyplot as plt

napari_cellseg3d/model_workers.py

@@ -201,7 +201,7 @@ def inference(self):
         pad = utils.get_padding_dim(check)
         # print(pad)
         dims = 128
-        dims = 64
+        # dims = 64 # TODO
 
         model = self.model_dict["class"].get_net()
         if self.model_dict["name"] == "SegResNet":

napari_cellseg3d/plugin_crop.py

@@ -8,7 +8,6 @@
 from magicgui.widgets import Slider
 
 # Qt
-from qtpy.QtWidgets import QLabel
 from qtpy.QtWidgets import QSizePolicy
 from tifffile import imwrite
 
@@ -51,7 +50,7 @@ def __init__(self, viewer: "napari.viewer.Viewer", parent):
 
         self.box_widgets = ui.make_n_spinboxes(3, 1, 1000, DEFAULT_CROP_SIZE)
         self.box_lbl = [
-            QLabel("Size in " + axis + " of cropped volume :")
+            ui.make_label("Size in " + axis + " of cropped volume :", self)
             for axis in "xyz"
         ]
 

napari_cellseg3d/plugin_model_inference.py

@@ -5,9 +5,7 @@
 import numpy as np
 
 # Qt
-from qtpy.QtWidgets import QLabel
 from qtpy.QtWidgets import QSizePolicy
-from qtpy.QtWidgets import QWidget
 
 # local
 from napari_cellseg3d import interface as ui
@@ -28,30 +26,32 @@ def __init__(self, viewer: "napari.viewer.Viewer"):
         * Data :
             * A file extension choice for the images to load from selected folders
 
-            * Two buttons to choose the images folder to run segmentation and save results in, respectively
+            * Two fields to choose the images folder to run segmentation and save results in, respectively
+
+        * Inference options :
+            * A dropdown menu to select which model should be used for inference
+
+            * An option to load custom weights for the selected model (e.g. from training module)
+
 
         * Post-processing :
             * A box to select if data is anisotropic, if checked, asks for resolution in micron for each axis
 
             * A box to choose whether to threshold, if checked asks for a threshold between 0 and 1
 
-        * Display options :
-            * A dropdown menu to select which model should be used for inference
+            * A box to enable instance segmentation. If enabled, displays :
+                * The choice of method to use for instance segmentation
+                
+                * The probability threshold below which to remove objects
+
+                * The size in pixels of small objects to remove
 
-            * A checkbox to choose whether to display results in napari afterwards. Will ask for how many results to display, capped at 10
+        * A checkbox to choose whether to display results in napari afterwards. Will ask for how many results to display, capped at 10
 
         * A button to launch the inference process
 
         * A button to close the widget
 
-        TODO:
-
-        * Verify if way of loading model is  OK
-
-        * Padding OK ?
-
-        * Save toggle ?
-
         Args:
             viewer (napari.viewer.Viewer): napari viewer to display the widget in
         """
@@ -100,7 +100,7 @@ def __init__(self, viewer: "napari.viewer.Viewer"):
         )
 
         self.display_number_choice = ui.make_n_spinboxes(1, 1, 10, 1)
-        self.lbl_display_number = QLabel("How many ? (max. 10)", self)
+        self.lbl_display_number = ui.make_label("How many ? (max. 10)", self)
 
         self.aniso_checkbox = ui.make_checkbox(
             "Anisotropic data", self.toggle_display_aniso
@@ -110,7 +110,8 @@ def __init__(self, viewer: "napari.viewer.Viewer"):
             n=3, min=1.0, max=1000, default=1.5, step=0.5, double=True
         )
         self.aniso_box_lbl = [
-            QLabel("Resolution in " + axis + " (microns) :") for axis in "xyz"
+            ui.make_label("Resolution in " + axis + " (microns) :", self)
+            for axis in "xyz"
         ]
 
         self.aniso_box_widgets[-1].setValue(5.0)  # TODO change default
@@ -147,7 +148,9 @@ def __init__(self, viewer: "napari.viewer.Viewer"):
         self.instance_prob_thresh = ui.make_n_spinboxes(
             n=1, max=0.99, default=0.7, step=0.05, double=True
         )
-        self.instance_prob_thresh_lbl = QLabel("Probability threshold :")
+        self.instance_prob_thresh_lbl = ui.make_label(
+            "Probability threshold :", self
+        )
         self.instance_prob_t_container = ui.combine_blocks(
             second=self.instance_prob_thresh,
             first=self.instance_prob_thresh_lbl,
@@ -157,8 +160,8 @@ def __init__(self, viewer: "napari.viewer.Viewer"):
         self.instance_small_object_thresh = ui.make_n_spinboxes(
             n=1, max=100, default=10, step=5
         )
-        self.instance_small_object_thresh_lbl = QLabel(
-            "Small object removal threshold :"
+        self.instance_small_object_thresh_lbl = ui.make_label(
+            "Small object removal threshold :", self
         )
         self.instance_small_object_t_container = ui.combine_blocks(
             second=self.instance_small_object_thresh,