added additional comments and features to functions

Author: DerikVo

Created_images/Baseline confusion matrix.png

@@ -0,0 +1,153 @@
+import os
+import cv2
+import numpy as np
+import pandas as pd
+from tensorflow.keras.preprocessing.image import img_to_array, load_img
+from sklearn.metrics import confusion_matrix, precision_score, recall_score, f1_score
+import seaborn as sns
+import matplotlib.pyplot as plt
+
+
+'''
+This portion uses code from a previous project from this [notebook](https://github.com/DerikVo/DSI_project_4_plant_disease/blob/main/notebooks/01_Potato_PlantVillageEDA.ipynb).
+The code was originally developed by chat GPT 4 with the prompt:  "I have an image data set that I want to do EDA on. How can I average out the pixel values of all the images in a class. python keras."
+
+This function takes two arguments the dataset: training or testing, and the sub_folder for the type of tumor e.g. ['glioma', 'meningioma', 'notumor', 'pituitary']
+This function is used to find the average pixel values of each class
+The purpose is to find if there is a difference in each class
+
+'''
+def avg_images(class_name, dataset='Training'):
+    '''
+    This function is used to find the average pixel value of each class
+    
+    Users will need to assign the images to a variable. 
+    For example:
+    meningioma_tumor = eda.avg_images('meningioma')
+    =============================================================================    
+    Keyword arguments:
+    folders -- The sub folder containing the classifcation for tumor type ( 'glioma',
+    'meningioma', 'notumor', 'pituitary' )
+    dataset -- The main folder either Train and Test folder (default = Training)
+    '''
+    path = f'../Images/{dataset}/{class_name}/'
+    image_files = os.listdir(path)
+    num_images = len(image_files)
+    average_image = np.zeros((256, 256, 1), dtype=np.float32)
+
+    for image_file in image_files:
+        image_path = os.path.join(path, image_file)
+        image = load_img(image_path, color_mode='grayscale', target_size=(256, 256))
+        image_array = img_to_array(image)
+        average_image += image_array / num_images
+
+    return average_image
+
+'''
+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.
+'''
+def find_closest_class(mean_pixel_value, class_averages):
+    #initialize the cloest class variable
+    closest_class = None
+    #initialize the cloest class variable
+    closest_distance = float('inf')
+    for class_name, average in class_averages.items():
+        #finds the distance between the mean pixel value and the class average
+        distance = np.linalg.norm(mean_pixel_value - average)
+        # Finds the smaller distance
+        if distance < closest_distance:
+            #update the closest distance
+            closest_distance = distance
+            #updates the closest class
+            closest_class = class_name
+
+    return closest_class
+
+'''
+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.
+'''
+def classify_images(test_folder_path, class_paths):
+    #create a list for the actual images
+    actual_classes = []
+    #create a list for the predictions
+    predicted_classes = []
+    #store the average pixel values for each class
+    class_averages = {}
+    for class_name, class_path in class_paths.items():
+        #calculates the average pixcel value for each class (function should default to the training data set)
+        average_image = avg_images(class_name)
+        #finds the mean pixel value and stores it as a key value pair
+        class_averages[class_name] = np.mean(average_image)
+
+    for class_name, class_path in class_paths.items():
+        #gets the path to the class folders under the testing data set
+        class_folder_path = os.path.join(test_folder_path, class_name)
+        #gets a list of the images within the folder
+        image_files = os.listdir(class_folder_path)
+
+        for image_file in image_files:
+            #gets the path an individual image
+            image_path = os.path.join(class_folder_path, image_file)
+            #reads the image path using Open CV
+            test_image = cv2.imread(image_path)
+            #gets the mean pixcel value of the image
+            mean_pixel_value = np.mean(test_image, axis=(0, 1))
+            #uses the find_cloest_class function to find what class its closest to
+            closest_class = find_closest_class(mean_pixel_value, class_averages)
+            #appends the actual class to the actual classes list
+            actual_classes.append(class_name)
+            #appends the predicted class to the predicted classes list
+            predicted_classes.append(closest_class)
+
+    return actual_classes, predicted_classes
+'''
+This portion reuses code from prior projects. The confusion matrix used the project: https://github.com/DerikVo/DSI_project_4_plant_disease/blob/main/notebooks/02_plant_village_potato_modeling.ipynb
+which prompted ChatGPT 4 to help grab the labels information from the validation dataset and get it into a numpy array, so that we can use that to make a confusion matrix.
+
+The creatation of the data frame was taken from this project: https://github.com/DerikVo/NN_hackathon/blob/main/Code/Training/pre-trained-models.ipynb
+
+portions of the code have been adapted to work with file pathways.
+
+Prompted ChatGPT 3 with incorporating the code as a function that uses the classify_images function to get the confusion matrix and classification metrics. Then prompted and correct some syntax errors. 
+'''
+
+def calculate_metrics(actual_classes, predicted_classes, class_paths):
+    #creates the confusion matrix
+    cm = confusion_matrix(actual_classes, predicted_classes,
+    # gets the label of each class
+    labels=list(class_paths.keys()))
+    #Finds the weighted scores for each metric
+    precision = precision_score(actual_classes, predicted_classes, average='weighted')
+    recall = recall_score(actual_classes, predicted_classes, average='weighted')
+    f1 = f1_score(actual_classes, predicted_classes, average='weighted')
+    #adds the scores into a data frame
+    data = {'Precision': [precision], 'Recall': [recall], 'F1 Score': [f1]}
+    metrics_df = pd.DataFrame(data, index=['baseline'])
+
+    return cm, metrics_df
+
+'''
+This portion reuses code from prior projects. The confusion matrix used the project: https://github.com/DerikVo/DSI_project_4_plant_disease/blob/main/notebooks/02_plant_village_potato_modeling.ipynb
+which prompted ChatGPT 4 to help grab the labels information from the validation dataset and get it into a numpy array, so that we can use that to make a confusion matrix.
+
+There were slight modifications to fit the purposes of this code such as assigning class paths and a title parameter.
+'''
+def plot_confusion_matrix(confusion_matrix, class_paths, title):
+    #sets the figure size
+    plt.figure(figsize=(10,10))
+    #Plots the confusion matrix and assigns the class names on the axis ticks
+    sns.heatmap(confusion_matrix, annot=True, cmap='Blues', fmt='g',
+    xticklabels=list(class_paths.keys()), yticklabels=list(class_paths.keys()))
+    #labels the axis
+    plt.xlabel('Predicted Class')
+    plt.ylabel('True Class')
+    #sets the title
+    plt.title(f'{title} Confusion Matrix')
+    #saves the figure
+    plt.savefig(f'../Created_images/{title} confusion matrix.png')
+    #displays the image
+    plt.show()