add simple example (#5)

* add simple example * add simple example
lapix-ufsc · Aug 4, 2022 · 2e26100 · 2e26100
1 parent 8a2b78b
commit 2e26100
Show file tree

Hide file tree

Showing 8 changed files with 182 additions and 29 deletions.
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -8,4 +8,5 @@ repos:
       args: [--py37-plus]
     - id: nbqa-isort
     - id: nbqa-black
-    - id: nbqa-flake8
+    - id: nbqa-flake8
+      args: ['--config=settings.ini']
diff --git a/README.md b/README.md
@@ -1,34 +1,41 @@
-# dl-tutorials
-This repository contains demonstrations done with deep learning computer vision models.
-
-## Datasets
-
-1. [CCAgT: Images of Cervical Cells with AgNOR Stain Technique](https://doi.org/10.17632/wg4bpm33hj.2) 
-2. [UFSC OCPap: Papanicolaou Stained Oral Cytology Dataset (v4)](doi.org/10.17632/dr7ydy9xbk.1)
-3. 
+Deep Learning tutorials
+================
 
+<!-- WARNING: THIS FILE WAS AUTOGENERATED! DO NOT EDIT! -->
 
-## Semantic Segmentation
+## Content
 
-### Models
+This package/repository is not a library, but a set of tutorials using
+several libraries with models and utilities for computer vision using
+deep learning.
 
-1. 
-2. 
+These tutorials were built in a way that their contents are
+self-contained, and that they can be used as a basis for other
+experiments.
 
-## Object detection
+The tutorials will cover models and tools for semantic segmentation,
+object detection, image classification, tracking, augmentation, model
+evaluation, among other topics.
 
-### Models
-
-1.
-2.
+## Datasets
 
-## Image Classification
+These tutorials are the fruit of different Lapix researchers, who
+throughout their masters or doctoral degrees developed several computer
+vision datasets.
 
-### Models
+Therefore, these tutorials were created from experiments using the
+following datasets:
 
-1.
-2.
+1.  [CCAgT: Images of Cervical Cells with AgNOR Stain
+    Technique](https://doi.org/10.17632/wg4bpm33hj.2)
+2.  [Clouds-1000](https://doi.org/10.17632/4pw8vfsnpx.2)
+3.  [UFSC OCPap: Papanicolaou Stained Oral Cytology Dataset
+    (v4)](https://doi.org/10.17632/dr7ydy9xbk.1)
 
-## Testing models
+## Authors
 
-[LapixDL](https://doi.org/10.5281/zenodo.5963342)
+|             Name             | GitHub                                   | Orcid                                                        |
+|:----------------------------:|------------------------------------------|--------------------------------------------------------------|
+|     Aldo von Wangenheim      | [@awangenh](https://github.com/awangenh) | [0000-0003-4532-1417](https://orcid.org/0000-0003-4532-1417) |
+| João Gustavo Atkinson Amorim | [@johnnv1](http://github.com/johnnv1)    | [0000-0003-3361-6891](https://orcid.org/0000-0003-3361-6891) |
+|                              |                                          |                                                              |
diff --git a/dl-tutorials/Datasets/CCAgT.ipynb b/dl-tutorials/Datasets/CCAgT.ipynb
@@ -0,0 +1,69 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# CCAgT: Images of Cervical Cells with AgNOR Stain Technique"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "This dataset was available at Mendelay data in the link: [https://doi.org/10.17632/wg4bpm33hj.2](https://doi.org/10.17632/wg4bpm33hj.2)\n",
+    "\n",
+    "\n",
+    "Contains 9339 images with resolution of 1600×1200 where each pixel is 0.111µmX0.111µm from 15 different slides stained with AgNOR technique, having at least one label per image. Have more than sixty-three thousand annotations. The images from patients of Gynecology and Colonoscopy Outpatient Clinic of the Polydoro Ernani de São Thiago University Hospital of the Universidade Federal de Santa Catarina (HU-UFSC). This research was approved by the UFSC Research Ethics Committee (CEPSH), protocol number 57423616.3.0000.0121. First, all patients involved were informed about the objectives of the study, and those who agreed to participate signed an informed consent form."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Using Hugging Faces datasets"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Installing HF datasets lib\n",
+    "!pip install datasets"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%nbdev_collapse_output\n",
+    "from datasets import load_dataset\n",
+    "\n",
+    "dataset = load_dataset(\"lapix/CCAgT\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3.8.10 ('venv': venv)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "name": "python",
+   "version": "3.8.10"
+  },
+  "orig_nbformat": 4,
+  "vscode": {
+   "interpreter": {
+    "hash": "14fc90b036d17ab8931f0afc7b06851673501038f7e6077c614cdf5f462d6df3"
+   }
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/dl-tutorials/Semantic segmentation/Fastai/unet.ipynb b/dl-tutorials/Semantic segmentation/Fastai/unet.ipynb
@@ -0,0 +1,30 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Unet"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3.10.6 64-bit",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "name": "python",
+   "version": "3.10.6"
+  },
+  "orig_nbformat": 4,
+  "vscode": {
+   "interpreter": {
+    "hash": "97cc609b13305c559618ec78a438abc56230b9381f827f22d070313b9a1f3777"
+   }
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/dl-tutorials/Semantic segmentation/index.ipynb b/dl-tutorials/Semantic segmentation/index.ipynb
@@ -0,0 +1,37 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# General Explanation"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "It is called “semantic” because it associates an individual pixel directly with its “meaning”: each pixel is classified accordingly to the class of object to which it pertains and not only aggregated into meaningless regions based on syntactic criteria, such as color homogeneity or borders and color variation. Semantic segmentation eliminates the need for additional pattern recognition for the classification of the regions that result from a segmentation."
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3.8.10 ('venv': venv)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "name": "python",
+   "version": "3.8.10"
+  },
+  "orig_nbformat": 4,
+  "vscode": {
+   "interpreter": {
+    "hash": "14fc90b036d17ab8931f0afc7b06851673501038f7e6077c614cdf5f462d6df3"
+   }
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/dl-tutorials/_quarto.yml b/dl-tutorials/_quarto.yml
@@ -9,7 +9,7 @@ project:
 
 format:
   html:
-    theme: cosmo
+    theme: flatly
     css: styles.css
     toc: true
     toc-depth: 4

diff --git a/dl-tutorials/sidebar.yml b/dl-tutorials/sidebar.yml
@@ -1,4 +1,13 @@
 website:
   sidebar:
     contents:
-      - index.ipynb
+      - index.ipynb
+      - section: Datasets
+        contents:
+          - Datasets/CCAgT.ipynb
+      - section: Semantic segmentation
+        contents:
+          - Semantic segmentation/index.ipynb
+          - section: Fastai
+            contents:
+              - Semantic segmentation/Fastai/unet.ipynb
diff --git a/settings.ini b/settings.ini
@@ -30,9 +30,9 @@ title = %(lib_name)s
 ### OPTIONAL ### see https://github.com/fastai/nbdev/blob/master/settings.ini for examples
 
 # requirements = 
-# dev_requirements = 
+dev_requirements = datasets Pillow
 # console_scripts =
 
-
-
+[flake8]
+ignore = E402