Skip to content
Snippets Groups Projects
Select Git revision
  • e7e0aa41bb4b2bfb93105fea0747ae953a23eabc
  • master default protected
2 results

main.py

Blame
  • Code owners
    Assign users and groups as approvers for specific file changes. Learn more.
    ANN_Training.py 9.35 KiB
    # -*- coding: utf-8 -*-
    """
    @author: Laura C. Kühle, Soraya Terrab (sorayaterrab)
    
    Code-Style: E226, W503
    Docstring-Style: D200, D400
    
    TODO: Test new ANN set-up with Soraya
    TODO: Remove object set-up (for more flexibility)
    TODO: Add documentation
    TODO: Improve log output (timer, bit of text) -> Done
    TODO: Throw exception for error due to missing classes
    TODO: Allow multiple approximations in one config
    
    """
    import numpy as np
    import time
    import matplotlib
    from matplotlib import pyplot as plt
    import os
    import torch
    from torch.utils.data import TensorDataset, DataLoader, random_split
    from sklearn.model_selection import KFold
    from sklearn.metrics import accuracy_score, precision_recall_fscore_support, roc_auc_score
    
    import ANN_Model
    from Plotting import plot_classification_accuracy, plot_boxplot
    
    matplotlib.use('Agg')
    
    
    class ModelTrainer(object):
        def __init__(self, config):
            self._reset(config)
    
        def _reset(self, config):
            self._dir = config.pop('dir', 'test_data')
            self._model_name = config.pop('model_name', '0')
            self._training_data = read_training_data(self._dir)
    
            self._batch_size = config.pop('batch_size', min(len(self._training_data)//2, 500))
            self._num_epochs = config.pop('num_epochs', 1000)
            self._threshold = config.pop('threshold', 1e-5)
            self._model = config.pop('model', 'ThreeLayerReLu')
            self._model_config = config.pop('model_config', {})
            self._loss_function = config.pop('loss_function', 'BCELoss')
            self._loss_config = config.pop('loss_config', {})
            self._optimizer = config.pop('optimizer', 'Adam')
            self._optimizer_config = config.pop('optimizer_config', {})
    
            # Set learning rate
            self._learning_rate = config.pop('learning_rate', 1e-2)
            self._optimizer_config['lr'] = self._learning_rate
    
            if not hasattr(ANN_Model, self._model):
                raise ValueError('Invalid model: "%s"' % self._model)
            if not hasattr(torch.nn.modules.loss, self._loss_function):
                raise ValueError('Invalid loss function: "%s"' % self._loss_function)
            if not hasattr(torch.optim, self._optimizer):
                raise ValueError('Invalid optimizer: "%s"' % self._optimizer)
    
            self._model = getattr(ANN_Model, self._model)(self._model_config)
            self._loss_function = getattr(torch.nn.modules.loss, self._loss_function)(
                **self._loss_config)
            self._optimizer = getattr(torch.optim, self._optimizer)(
                self._model.parameters(), **self._optimizer_config)
            self._validation_loss = torch.zeros(self._num_epochs//10)
    
        def epoch_training(self, dataset=None, num_epochs=None, verbose=True):
            tic = time.perf_counter()
            # Split data into training and validation set
            if dataset is None:
                dataset = self._training_data
            if num_epochs is None:
                num_epochs = self._num_epochs
            num_samples = len(dataset)
            if verbose:
                print('Splitting data randomly into training and validation set.')
            train_ds, valid_ds = random_split(dataset, [round(num_samples*0.8), round(num_samples*0.2)])
    
            # Load sets
            train_dl = DataLoader(train_ds, batch_size=self._batch_size, shuffle=True)
            valid_dl = DataLoader(valid_ds, batch_size=self._batch_size * 2)
    
            # Training with Validation
            if verbose:
                print('\nTraining model...')
                print('Number of epochs:', num_epochs)
            tic_train = time.perf_counter()
            for epoch in range(num_epochs):
                self._model.train()
                for x_batch, y_batch in train_dl:
                    pred = self._model(x_batch.float())
                    loss = self._loss_function(pred, y_batch.float()).mean()
    
                    # Run back propagation, update the weights, and zero gradients for next epoch
                    loss.backward()
                    self._optimizer.step()
                    self._optimizer.zero_grad()
    
                self._model.eval()
                with torch.no_grad():
                    valid_loss = sum(
                        self._loss_function(self._model(x_batch_valid.float()), y_batch_valid.float())
                        for x_batch_valid, y_batch_valid in valid_dl)
    
                    if (epoch+1) % 100 == 0:
                        self._validation_loss[int((epoch+1) / 100)-1] = valid_loss / len(valid_dl)
                        if verbose:
                            print(epoch+1, 'epochs completed. Loss:', valid_loss / len(valid_dl))
    
                    if valid_loss / len(valid_dl) < self._threshold:
                        break
            toc_train = time.perf_counter()
            if verbose:
                print('Finished training model!')
                print(f'Training time: {toc_train-tic_train:0.4f}s\n')
            toc = time.perf_counter()
            if verbose:
                print(f'Total runtime: {toc-tic:0.4f}s\n')
    
        def test_model(self, training_set, test_set):
            self.epoch_training(training_set, num_epochs=100, verbose=False)
            self._model.eval()
    
            x_test, y_test = test_set
            model_score = self._model(x_test.float())
            model_output = torch.argmax(model_score, dim=1)
    
            y_true = y_test.detach().numpy()[:, 1]
            y_pred = model_output.detach().numpy()
            accuracy = accuracy_score(y_true, y_pred)
            precision, recall, f_score, support = precision_recall_fscore_support(y_true, y_pred)
            auroc = roc_auc_score(y_true, y_pred)
    
            return {'Precision_Smooth': precision[0], 'Precision_Troubled': precision[1],
                    'Recall_Smooth': recall[0], 'Recall_Troubled': recall[1],
                    'F-Score_Smooth': f_score[0], 'F-Score_Troubled': f_score[1],
                    'Accuracy': accuracy, 'AUROC': auroc}
    
        def save_model(self):
            # Saving Model
            name = self._model_name
    
            # Set paths for files if not existing already
            model_dir = self._dir + '/trained models'
            if not os.path.exists(model_dir):
                os.makedirs(model_dir)
    
            torch.save(self._model.state_dict(), model_dir + '/model__' + name + '.pt')
            torch.save(self._validation_loss, model_dir + '/loss__' + name + '.pt')
    
    
    def read_training_data(directory, normalized=True):
        # Get training dataset from saved file and map to Torch tensor and dataset
        input_file = directory + ('/normalized_input_data.npy' if normalized else '/input_data.npy')
        output_file = directory + '/output_data.npy'
        return TensorDataset(*map(torch.tensor, (np.load(input_file), np.load(output_file))))
    
    
    def evaluate_models(models, directory, num_iterations=100, colors=None,
                        compare_normalization=False):
        tic = time.perf_counter()
        if colors is None:
            colors = {'Accuracy': 'magenta', 'Precision_Smooth': 'red',
                      'Precision_Troubled': '#8B0000', 'Recall_Smooth': 'blue',
                      'Recall_Troubled': '#00008B', 'F-Score_Smooth': 'green',
                      'F-Score_Troubled': '#006400', 'AUROC': 'yellow'}
    
        print('Read normalized training data.')
        datasets = {'normalized': read_training_data(directory)}
        if compare_normalization:
            print('Read raw, non-normalized training data.')
            datasets['raw'] = read_training_data(directory, False)
        classification_stats = {measure: {model + ' (' + dataset + ')': [] for model in models
                                          for dataset in datasets} for measure in colors}
    
        print('\nTraining models with 5-fold cross validation...')
        print('Number of iterations:', num_iterations)
        tic_train = time.perf_counter()
        for iteration in range(num_iterations):
            for train_index, test_index in KFold(
                    n_splits=5, shuffle=True).split(datasets['normalized']):
                for dataset in datasets.keys():
                    training_set = TensorDataset(*datasets[dataset][train_index])
                    test_set = datasets[dataset][test_index]
    
                    for model in models:
                        result = models[model].test_model(training_set, test_set)
                        for measure in colors:
                            classification_stats[measure][model + ' (' + dataset + ')'].append(
                                result[measure])
            if iteration+1%max(10, 10*(num_iterations//100)):
                print(iteration+1, 'iterations completed.')
        toc_train = time.perf_counter()
        print('Finished training models with 5-fold cross validation!')
        print(f'Training time: {toc_train - tic_train:0.4f}s\n')
    
        print('Plotting evaluation of trained models.')
        plot_boxplot(classification_stats, colors)
        classification_stats = {measure: {model + ' (' + dataset + ')': np.array(
            classification_stats[measure][model + ' (' + dataset + ')']).mean() for model in models
                                          for dataset in datasets} for measure in colors}
        plot_classification_accuracy(classification_stats, colors)
    
        # Set paths for plot files if not existing already
        plot_dir = directory + '/model evaluation'
        if not os.path.exists(plot_dir):
            os.makedirs(plot_dir)
    
        # Save plots
        print('Saving plots.')
        for identifier in plt.get_figlabels():
            # Set path for figure directory if not existing already
            if not os.path.exists(plot_dir + '/' + identifier):
                os.makedirs(plot_dir + '/' + identifier)
    
            plt.figure(identifier)
            plt.savefig(plot_dir + '/' + identifier + '/' + '_'.join(models.keys()) + '.pdf')
        toc = time.perf_counter()
        print(f'Total runtime: {toc - tic:0.4f}s')