Data Preprocessing

[jessmed749]

Step 3: Normalized and reshaped the data. Study the architecture of CNNs. Learn about convolutional layers, pooling layers, and fully connected layers.

Details:

• Normalized pixel values from 0-255 to 0-1 by dividing by 255. Reshaped images for CNN input to (28, 28, 1).

• Decide how many layers and what types of layers to use in the model. Activation functions and output layer considerations for classification.

• Outcome: Data preprocessing pipeline created and tested with the initial training and test images.

[jessmed749]

A CNN (Convolutional Neural Network) is a type of deep learning neural network specifically designed for processing structured grid data, such as images.

CNN Architecture Components:

1. Convolutional Layer:

   • Applies a set of learnable filters to the input image.
   • Outputs feature maps that highlight specific patterns or features in the image.

2. Activation Function (usually ReLU):

   • Introduces non-linearity to the model, allowing it to learn complex relationships.

3. Pooling Layer (typically Max Pooling):

   • Reduces the spatial dimensions of the feature maps (down-sampling).
   • Helps in reducing computational load and extracting dominant features.

4. Fully Connected (Dense) Layer:

   • Connects all neurons from the previous layer and is used at the end of the network to output the final classification results.

5. Output Layer:

   • Uses an activation function like Softmax for multi-class classification to give the probability distribution for each class.