Glass-Classification-DL-ANN

For Glass Classification using a Deep Learning Model (ANN), here’s a breakdown of the key topics:

1- CSV Files & Data Splitting (pandas/sklearn):

• Load the dataset using pandas from a CSV file.
• Split the dataset into training and testing sets using train_test_split from the sklearn library.

2- Data Preprocessing:

• Normalize the features using StandardScaler to standardize the input features for better convergence during training.

3- Artificial Neural Network (ANN):

• A sequential model from Keras will be used to build the ANN with layers defined using Dense layers.
• Layers include input, hidden, and output layers, with activation functions such as ReLU (for hidden layers) and softmax (for multi-class classification).

4- Optimizer (Adam):

• Adam is chosen as the optimizer, which is efficient for deep learning models, combining the advantages of both RMSProp and momentum.

5- Loss Function:

• For multi-class classification, use categorical_crossentropy as the loss function.

6- Metrics:

• Track performance using metrics like accuracy, precision, recall, and F1-score.
• Evaluation is done using the Keras model’s evaluate method.

7- Epochs and Training:

• Train the model over a defined number of epochs, adjusting the batch size as needed.
• Use the training and validation data to observe how the model improves over epochs.

8- Predictions:

• After training, make predictions on the test set using model.predict and evaluate the model's performance on unseen data.

Links:

• Project Notebook
• Dataset