Merge pull request #8 from DerikVo/readme

added more information to the conclusions in the evaluation notebook

Author: DerikVo

Created_images/Neural Network confusion matrix.png

@@ -634,17 +634,11 @@
    "metadata": {},
    "source": [
     "_________________________________________________________________________________________-\n",
-    "# Conclusion"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "id": "d267e36b-9eb2-4301-b186-c09e4ce65af0",
-   "metadata": {},
-   "source": [
+    "# Conclusion\n",
+    "\n",
     "Overall, our classes are balanced between the classes 'glioma', 'meningioma', 'notumor', and 'pituitary' in both our training and testing datasets. Additionally when looking at the average pixel value and the contrast between those average we can see some distinct features within those classes. This indicates that there are features that our model can learn to distinguish images into classes.\n",
     "\n",
-    "We will know move onto building [A baseine model](../Notebooks/02_Baseline Model.ipynb)"
+    "We will know move onto building [A baseline model](../Notebooks/01_EDA.ipynb) to compare our results."
    ]
   }
  ],

Created_images/Neural Network with regularization confusion matrix.png

@@ -159,7 +159,9 @@
     "# Conclusion:\n",
     "\n",
     "\n",
-    "Our baseline model has an accuracy of 46% which is better than simply guessing one class. This model simply compares which class an image belongs to by seeing the avg pixel value of a class and the image and seeing which class the image is closest to. \n",
+    "Our baseline model has an accuracy of 46% which is better than simply guessing one class. This model simply compares which class an image belongs to by seeing the avg pixel value of a class and the image and seeing which class the image is closest to.\n",
+    "\n",
+    "With our baseline our most accurate classification is no tumor and glioma.\n",
     "\n",
     "We will now move into a more complex mode, a [neural network](../Notebooks/03_Neural_network.ipynb), to classify our images."
    ]

Notebooks/01_EDA.ipynb

@@ -1,6 +1,4 @@
 ,Accuracy,Precision,Recall,F1 Score
 baseline,0.4698703279938978,0.4683084749604378,0.4698703279938978,0.4635669509975083
-Neural Network: No regularization,0.8779557589626239,0.875790866125634,0.8779557589626239,0.8729339270233065
-Neural Network: With Regularization,0.8527841342486652,0.849915756851124,0.8527841342486652,0.8447039064139434
-Neural Network,0.8749046529366895,0.8738431092045584,0.8749046529366895,0.8672969669842382
-Neural Network: regularization,0.8421052631578947,0.8403497250779882,0.8421052631578947,0.8293207661173666
+Neural Network: No regularization,0.8749046529366895,0.8738431092045584,0.8749046529366895,0.8672969669842382
+Neural Network: With Regularization,0.8421052631578947,0.8403497250779882,0.8421052631578947,0.8293207661173666

Notebooks/02_Baseline Model.ipynb

@@ -46,7 +46,7 @@ def avg_images(class_name, dataset='Training'):
 '''
 The Code was originally developed by chat GPT 3 with the prompt: "I want to find the average pixel value of each class and then use the mean of an image to find which class it belongs to. The path to the class looks like '../Images/Training/glioma/' the classes are 'glioma', 'meningioma', 'notumor', 'pituitary'"
 
-was later prompted to adjust the code to be able to pass a parameter to the classify_images function. Took a total of 8 prompts and manual adjustments.
+was later prompted to adjust the code to be able to pass a parameter to the classify_images function. Took a total of 8 prompts and  3 manual adjustments.
 '''
 def find_closest_class(mean_pixel_value, class_averages):
     '''
@@ -75,7 +75,7 @@ def find_closest_class(mean_pixel_value, class_averages):
 '''
 The Code was originally developed by chat GPT 3 with the prompt: "How do I dynamically classify images using the folder they are in as a class. Please use the OS module"
 
-was later prompted to adjust the code to be able to pass a parameter to the find_closest_class function. Took a total of 8 prompts and manual adjustments.
+was later prompted to adjust the code to be able to pass a parameter to the find_closest_class function.
 '''
 def classify_images(test_folder_path, class_paths):
     '''
@@ -185,7 +185,16 @@ def plot_confusion_matrix(confusion_matrix, class_paths, title):
     #displays the image
     plt.show()
 
-    
+'''
+
+Creating a the data frame borrowed ideas from a one day hack-a-thon
+https://github.com/DerikVo/NN_hackathon
+
+Converting a dictionary was taken from: https://stackoverflow.com/questions/18837262/convert-python-dict-into-a-dataframe
+
+originally tried using functions that's only found in pandas 2.0
+
+'''
 
 def model_metrics(true_classes, predicted_classes, title):
     '''