Merge pull request #41 from Leengit/example_notebook

ENH: Example notebook to create tensorflow Dataset via histomics_stream

Author: Leengit

example/TA232-mask.png

@@ -0,0 +1,22 @@
+# Data source
+The file `example/TA232.svs` comes from the zip file available as
+```
+https://stanfordmedicine.box.com/s/ub8e0wlhsdenyhdsuuzp6zhj0i82xrb1
+```
+from the web page
+```
+https://github.com/stanfordmlgroup/DLBCL-Morph
+```
+It is in that zip file as
+```
+DLBCL-Morph/TMA/MYC/TA232.svs
+```
+
+The corresponding mask `example/TA232-mask.png` is randomly generated in Python with
+```python
+import numpy as np
+from PIL import Image
+arr = np.random.randint(0, 2, (number_of_rows_of_mask, number_of_columns_of_mask), dtype=np.int8)
+im = Image.fromarray(arr)
+im.save("TA232-mask.png")
+```

example/TA232-source.md

@@ -1,21 +1,5 @@
 {
  "cells": [
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "25408cb3-b6d6-4244-9346-6f3b3166cdb6",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "# install histomics_stream\n",
-    "!pip install -e /tf/notebooks/histomics_stream\n",
-    "\n",
-    "# add to system path\n",
-    "import sys\n",
-    "\n",
-    "sys.path.append(\"/tf/notebooks/histomics_stream/\")"
-   ]
-  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -38,9 +22,9 @@
     "\n",
     "# Add a slide to the study, including slide-wide information with it.\n",
     "my_slide0 = my_slides[\"Slide_0\"] = {}\n",
-    "my_slide0[\"filename\"] = \"/tf/notebooks/histomics_stream/example/TCGA-BH-A0BZ-01Z-00-DX1.45EB3E93-A871-49C6-9EAE-90D98AE01913.svs\"\n",
-    "my_slide0[\"slide_name\"] = \"TCGA-BH-A0BZ-01Z-00-DX1\"\n",
-    "my_slide0[\"slide_group\"] = \"TCGA-BH-A0BZ\"\n",
+    "my_slide0[\"filename\"] = \"TA232.svs\"\n",
+    "my_slide0[\"slide_name\"] = \"TA232\"\n",
+    "my_slide0[\"slide_group\"] = \"MYC\"\n",
     "my_slide0[\"number_pixel_rows_for_chunk\"] = 2048\n",
     "my_slide0[\"number_pixel_columns_for_chunk\"] = 2048\n",
     "\n",
@@ -88,8 +72,8 @@
     "    my_study_tiles_by_grid_and_mask,\n",
     "    number_pixel_overlap_rows_for_tile=0,\n",
     "    number_pixel_overlap_columns_for_tile=0,\n",
-    "    mask_filename=\"/tf/notebooks/histomics_stream/example/TCGA-BH-A0BZ-01Z-00-DX1.45EB3E93-A871-49C6-9EAE-90D98AE01913-mask.png\",\n",
-    "    randomly_select=100,\n",
+    "    mask_filename=\"TA232-mask.png\",\n",
+    "    randomly_select=50,\n",
     ")\n",
     "# We could apply this to a subset of the slides, but we will apply it to all slides in\n",
     "# this example.\n",
@@ -165,7 +149,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.0"
+   "version": "3.8.10"
   }
  },
  "nbformat": 4,

example/tensorflow_stream.ipynb

@@ -1,6 +1,6 @@
 """Whole-slide image streamer for TensorFlow."""
 
-__version__ = "2.1.0"
+__version__ = "2.1.1"
 
 """
 

histomics_stream/__init__.py

@@ -5,7 +5,7 @@
 
 setuptools.setup(
     name="histomics_stream",
-    version="2.1.0",
+    version="2.1.1",
     author="Lee Newberg",
     author_email="lee.newberg@kitware.com",
     description="A TensorFlow 2 package for reading whole slide images",