workbench.py

import curses
from curses import wrapper
from typing import Any
import torch
import torch.nn as nn
import torch.optim as optim
from cgtnnlib.datasets import breast_cancer_dataset, car_evaluation_dataset, student_performance_factors_dataset
from cgtnnlib.AugmentedReLUNetwork import AugmentedReLUNetwork
from cgtnnlib.RegularNetwork import RegularNetwork
from cgtnnlib.training import train_model
from cgtnnlib.common import evaluate_classification_model, evaluate_regression_model

# App's state

models: list[Any] = []
evaluations = []

def main(stdscr):
    curses.curs_set(0)  # Hide the cursor
    stdscr.clear()
    stdscr.refresh()

    current_row = 0
    menu = ['Train Model', 'List Models', 'Run Evaluation', 'Exit']

    while True:
        stdscr.clear()
        h, w = stdscr.getmaxyx()

        for idx, row in enumerate(menu):
            x = w // 2 - len(row) // 2
            y = h // 2 - len(menu) // 2 + idx
            if idx == current_row:
                stdscr.attron(curses.color_pair(1))
                stdscr.addstr(y, x, row)
                stdscr.attroff(curses.color_pair(1))
            else:
                stdscr.addstr(y, x, row)

        stdscr.refresh()

        key = stdscr.getch()

        if key == curses.KEY_UP and current_row > 0:
            current_row -= 1
        elif key == curses.KEY_DOWN and current_row < len(menu) - 1:
            current_row += 1
        elif key == curses.KEY_ENTER or key in [10, 13]:
            if current_row == 0:
                train_model_menu(stdscr)
            elif current_row == 1:
                list_models_menu(stdscr)
            elif current_row == 2:
                run_evaluation_menu(stdscr)
            elif current_row == 3:
                break

def train_model_menu(stdscr):
    stdscr.clear()
    stdscr.addstr(0, 0, "Select Dataset:")
    datasets = ['Breast Cancer', 'Car Evaluation', 'Student Performance Factors']
    for idx, dataset in enumerate(datasets):
        stdscr.addstr(idx + 1, 0, f"{idx + 1}. {dataset}")

    stdscr.refresh()
    key = stdscr.getch()
    selected_dataset = int(chr(key)) - 1

    if selected_dataset == 0:
        X, y = breast_cancer_dataset()
    elif selected_dataset == 1:
        X, y = car_evaluation_dataset()
    elif selected_dataset == 2:
        X, y = student_performance_factors_dataset()

    model = AugmentedReLUNetwork(inputs_count=X.shape[1], outputs_count=1, p=0.5)
    optimizer = optim.Adam(model.parameters(), lr=0.001)
    train_model(model, X, y, optimizer, epochs=10)
    models.append(model)

def list_models_menu(stdscr):
    stdscr.clear()
    stdscr.addstr(0, 0, "List of Models:")
    for idx, model in enumerate(models):
        stdscr.addstr(idx + 1, 0, f"{idx + 1}. {model}")
    stdscr.refresh()
    stdscr.getch()

def run_evaluation_menu(stdscr):
    stdscr.clear()
    stdscr.addstr(0, 0, "Select Model to Evaluate:")
    for idx, model in enumerate(models):
        stdscr.addstr(idx + 1, 0, f"{idx + 1}. {model}")

    stdscr.refresh()
    key = stdscr.getch()
    selected_model = int(chr(key)) - 1

    if selected_model < len(models):
        model = models[selected_model]
        X, y = breast_cancer_dataset()  # Example dataset
        evaluation = evaluate_classification_model(model, X, y)
        evaluations.append(evaluation)
        stdscr.clear()
        stdscr.addstr(0, 0, "Evaluation Results:")
        for idx, result in enumerate(evaluation):
            stdscr.addstr(idx + 1, 0, f"{result}")
        stdscr.refresh()
        stdscr.getch()

if __name__ == "__main__":
    wrapper(main)