Step 1: Load activity datasets |
Step 2: Encode predictive class values with one-hot encoding |
Step 3: Split data into train, validation, and test with (a 60:20:20) ratio using the stratification technique |
Step 4: Create classification model, M |
Step 5: Compile M with value-set for optimization technique, k-fold, and metrics |
Step 6: Fit model M with training data |
Step 7: Improve the model with a grid-search technique |
Step 8: Calculate accuracy and other classification metrics |
Step 9: Select the best learning parameters as computed with the Grid Search technique |
Step 10: Classify input data into respective output classes |