Merge pull request #3 from dev-rajk/main

updated documentation

.gitignore

@@ -161,3 +161,4 @@ cython_debug/
 .vscode
 YOLOv8_Explainer.egg-info/*
 .code-workspace/
+/Items

README.md

@@ -84,28 +84,28 @@ https://arxiv.org/abs/1710.11063 <br>
 Aditya Chattopadhyay, Anirban Sarkar, Prantik Howlader, Vineeth N Balasubramanian`
 
 https://arxiv.org/abs/1910.01279 <br>
-`Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networks
+`Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networks  
 Haofan Wang, Zifan Wang, Mengnan Du, Fan Yang, Zijian Zhang, Sirui Ding, Piotr Mardziel, Xia Hu`
 
 https://ieeexplore.ieee.org/abstract/document/9093360/ <br>
-`Ablation-cam: Visual explanations for deep convolutional network via gradient-free localization.
+`Ablation-cam: Visual explanations for deep convolutional network via gradient-free localization.  
 Saurabh Desai and Harish G Ramaswamy. In WACV, pages 972–980, 2020`
 
 https://arxiv.org/abs/2008.02312 <br>
-`Axiom-based Grad-CAM: Towards Accurate Visualization and Explanation of CNNs
+`Axiom-based Grad-CAM: Towards Accurate Visualization and Explanation of CNNs  
 Ruigang Fu, Qingyong Hu, Xiaohu Dong, Yulan Guo, Yinghui Gao, Biao Li`
 
 https://arxiv.org/abs/2008.00299 <br>
 `Eigen-CAM: Class Activation Map using Principal Components
 Mohammed Bany Muhammad, Mohammed Yeasin`
 
 http://mftp.mmcheng.net/Papers/21TIP_LayerCAM.pdf <br>
-`LayerCAM: Exploring Hierarchical Class Activation Maps for Localization
+`LayerCAM: Exploring Hierarchical Class Activation Maps for Localization  
 Peng-Tao Jiang; Chang-Bin Zhang; Qibin Hou; Ming-Ming Cheng; Yunchao Wei`
 
 https://arxiv.org/abs/1905.00780 <br>
-`Full-Gradient Representation for Neural Network Visualization
-Suraj Srinivas, Francois Fleuret`
+`Full-Gradient Representation for Neural Network Visualization  
+\n Suraj Srinivas, Francois Fleuret`
 
 https://arxiv.org/abs/1806.10206 <br>
 `Deep Feature Factorization For Concept Discovery

docs/functions.md

@@ -1,6 +1,10 @@
 # Functions
 
-this module contains the funtions that can help generate the heatmaps for 
+##Main Functions
+The core functions that can be used to visualise the different Class Activated Mapping(CAM) are given below. 
 ::: YOLOv8_Explainer.core
 
+##Helper Functions
+The functions that can be used to display images and provide various other functionalities can be found here. 
+
 ::: YOLOv8_Explainer.utils

docs/index.md

@@ -1,7 +1,7 @@
 # Welcome to YOLOv8 Explainer
 
 ## Simplify your understanding of YOLOv8 Results
-This is a package with state of the art methods for Explainable AI for computer vision using YOLOv8. This can be used for diagnosing model predictions, either in production or while developing models. The aim is also to serve as a benchmark of algorithms and metrics for research of new explainability methods.
+This is a package with state of the art Class Activated Mapping(CAM) methods for Explainable AI for computer vision using YOLOv8. This can be used for diagnosing model predictions, either in production or while developing models. The aim is also to serve as a benchmark of algorithms and metrics for research of new explainability methods.
 
 ### Install Environment & Dependencies
 
@@ -37,24 +37,30 @@ model = yolov8_heatmap(
         conf_threshold=0.4, 
         device = "cpu", 
         method = "EigenCAM", 
-        layer=[10, 12, 14, 16, 18, -3],
-        backward_type="all",
+        layer=[12, 17, 21],
         ratio=0.02,
         show_box=True,
         renormalize=False,
 )
 
-imagelist = model(
+images = model(
     img_path="/location/image.jpg", 
     )
 
-display_images(imagelist)
+display_images(images)
 
 ```
+- Here the `from YOLOv8_Explainer import yolov8_heatmap, display_images` allows you to import the required functionalities. 
 
-You can choose between the following CAM Models for version 0.0.2:
+- The line `model = yolov8_heatmap( weight="/location/model.pt", conf_threshold=0.4, device = "cpu", method = "EigenCAM", layer=[12, 17, 21], ratio=0.02,    show_box=True, renormalize=False)` allows the user to pass a pertrained YOLO weight which the CAM is expected to evaluate, along with additional parameters like the desired CAM method target layers, and the confidence threshold.
+
+You can choose between the following CAM Models for version 0.0.5:
 
 `GradCAM` , `HiResCAM`, `GradCAMPlusPlus`, `XGradCAM` , `LayerCAM`, `EigenGradCAM` and `EigenCAM`.
 
+- The line `images = model( img_path="/location/image.jpg" )` passes the images the model will process
+
+- The line `display_images(images)` displays the output of the model, along with the CAM model's output. 
+
 You can add a single image or a directory images to be used by the `Module`. The output will be a corresponding list of images (list containing one PIL Image for a single image input and list containing as many PIL images as Images in the input directory).
 

mkdocs.yml

@@ -4,6 +4,8 @@ site_name: YOLOv8 Explainer
 theme:
   name: material
   features:
+    - search.suggest
+    - search.highlight
     - navigation.tabs
     - navigation.sections
     - navigation.top
@@ -12,25 +14,26 @@ theme:
     - navigation.indexes
     - toc.integrate
     - navigation.top
-    - search.suggest
-    - search.highlight
     - content.tabs.link
     - content.code.annotation
     - content.code.copy
+
   language: en
   palette:
-    - scheme: slate
+    - media: '(prefers-color-scheme: light)'
+      scheme: slate
+      primary: teal
+      accent: amber
       toggle:
-        icon: material/toggle-switch-off-outline 
+        icon: material/lightbulb
         name: Switch to light mode
+    - media: '(prefers-color-scheme: dark)'
+      scheme: default
       primary: teal
-      accent: purple 
-    - scheme: default
+      accent: amber
       toggle:
-        icon: material/toggle-switch
-        name: Switch to dark mode    
-      primary: teal
-      accent: lime
+        icon: material/lightbulb-outline
+        name: Switch to dark mode
 
 
 extra:
@@ -64,6 +67,7 @@ copyright: |
   &copy 2024 Proyash Paban Sarma Borah and Devraj Kashyap
 
 plugins: 
+ - search: null 
  - mkdocstrings:
     default_handler: python
     handlers: