How to use the anna.datasets.supervised_data_loader.SupervisedDataContainer function in anna

To help you get started, we’ve selected a few anna examples, based on popular ways it is used in public projects.

Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.

ifp-uiuc / do-neural-networks-learn-faus-iccvw-2015 / ck_plus / ck_plus_checkpoint_checker.py View on Github

# Load data
    train_folds, val_fold, _ = data_fold_loader.load_fold_assignment(test_fold=test_split)
    X_val, y_val = data_fold_loader.load_folds(data_paths.ck_plus_data_path, [val_fold])
    X_test, y_test = data_fold_loader.load_folds(data_paths.ck_plus_data_path, [test_split])
    print X_val.shape, y_val.shape
    print X_test.shape, y_test.shape

    X_val = numpy.float32(X_val)
    X_val /= 255.0
    X_val *= 2.0

    X_test = numpy.float32(X_test)
    X_test /= 255.0
    X_test *= 2.0

    val_data_container = SupervisedDataContainer(X_val, y_val)
    test_data_container = SupervisedDataContainer(X_test, y_test)

    # Construct evaluator
    preprocessor = [util.Normer3(filter_size=5, num_channels=1)]

    checkpoint_file_list = sorted(
        glob.glob(os.path.join(checkpoint_dir, '*.pkl')))
    val_evaluator = util.Evaluator(model, val_data_container,
                                   checkpoint_file_list[0], preprocessor)
    test_evaluator = util.Evaluator(model, test_data_container,
                                    checkpoint_file_list[0], preprocessor)

    # For each checkpoint, compute the overall val accuracy
    val_accuracies = []
    for checkpoint in checkpoint_file_list:
        print 'Checkpoint: %s' % os.path.split(checkpoint)[1]

ifp-uiuc / do-neural-networks-learn-faus-iccvw-2015 / ck_plus / ck_plus_single_checkpoint_evaluator.py View on Github

# Load dataset
    train_folds, val_fold, _ = data_fold_loader.load_fold_assignment(test_fold=test_split)
    X_val, y_val = data_fold_loader.load_folds(data_paths.ck_plus_data_path, [val_fold])
    X_test, y_test = data_fold_loader.load_folds(data_paths.ck_plus_data_path, [test_split])
    print X_val.shape, y_val.shape
    print X_test.shape, y_test.shape

    X_val = numpy.float32(X_val)
    X_val /= 255.0
    X_val *= 2.0

    X_test = numpy.float32(X_test)
    X_test /= 255.0
    X_test *= 2.0

    val_data_container = SupervisedDataContainer(X_val, y_val)
    test_data_container = SupervisedDataContainer(X_test, y_test)

    # Construct evaluator
    preprocessor = [util.Normer3(filter_size=5, num_channels=1)]

    val_evaluator = util.Evaluator(model, val_data_container,
                                   checkpoint_file, preprocessor)

    test_evaluator = util.Evaluator(model, test_data_container,
                                    checkpoint_file, preprocessor)

    # For the inputted checkpoint, compute the overall val accuracy
    print 'Checkpoint: %s' % os.path.split(checkpoint_file)[1]
    val_evaluator.set_checkpoint(checkpoint_file)
    val_accuracy = val_evaluator.run()
    print 'Val Accuracy: %f\n' % val_accuracy

ifp-uiuc / do-neural-networks-learn-faus-iccvw-2015 / ck_plus / ck_plus_single_checkpoint_evaluator.py View on Github

train_folds, val_fold, _ = data_fold_loader.load_fold_assignment(test_fold=test_split)
    X_val, y_val = data_fold_loader.load_folds(data_paths.ck_plus_data_path, [val_fold])
    X_test, y_test = data_fold_loader.load_folds(data_paths.ck_plus_data_path, [test_split])
    print X_val.shape, y_val.shape
    print X_test.shape, y_test.shape

    X_val = numpy.float32(X_val)
    X_val /= 255.0
    X_val *= 2.0

    X_test = numpy.float32(X_test)
    X_test /= 255.0
    X_test *= 2.0

    val_data_container = SupervisedDataContainer(X_val, y_val)
    test_data_container = SupervisedDataContainer(X_test, y_test)

    # Construct evaluator
    preprocessor = [util.Normer3(filter_size=5, num_channels=1)]

    val_evaluator = util.Evaluator(model, val_data_container,
                                   checkpoint_file, preprocessor)

    test_evaluator = util.Evaluator(model, test_data_container,
                                    checkpoint_file, preprocessor)

    # For the inputted checkpoint, compute the overall val accuracy
    print 'Checkpoint: %s' % os.path.split(checkpoint_file)[1]
    val_evaluator.set_checkpoint(checkpoint_file)
    val_accuracy = val_evaluator.run()
    print 'Val Accuracy: %f\n' % val_accuracy

ifp-uiuc / do-neural-networks-learn-faus-iccvw-2015 / ck_plus_six_class / ck_plus_checkpoint_checker.py View on Github

test_mask = numpy.logical_and(y_test != 0, y_test != 2)
    X_test = X_test[test_mask, :, :, :]
    y_test = y_test[test_mask]
    y_test = reindex_labels(y_test)
    num_test_samples = len(y_test)

    print 'Reduced Val Data: ', X_val.shape, y_val.shape
    print 'Reduced Test Data: ', X_test.shape, y_test.shape

    if test_split == 9:
        X_test, y_test = add_padding(X_test, y_test)
    elif test_split == 8:
        X_val, y_val = add_padding(X_val, y_val)

    val_data_container = SupervisedDataContainer(X_val, y_val)
    test_data_container = SupervisedDataContainer(X_test, y_test)

    # Construct evaluator
    preprocessor = [util.Normer3(filter_size=5, num_channels=1)]

    checkpoint_file_list = sorted(
        glob.glob(os.path.join(checkpoint_dir, '*.pkl')))
    val_evaluator = util.Evaluator(model, val_data_container,
                                   checkpoint_file_list[0], preprocessor)
    test_evaluator = util.Evaluator(model, test_data_container,
                                    checkpoint_file_list[0], preprocessor)

    # For each checkpoint, compute the overall val accuracy
    val_accuracies = []
    for checkpoint in checkpoint_file_list:
        print 'Checkpoint: %s' % os.path.split(checkpoint)[1]

ifp-uiuc / do-neural-networks-learn-faus-iccvw-2015 / ck_plus / ck_plus_checkpoint_checker.py View on Github

train_folds, val_fold, _ = data_fold_loader.load_fold_assignment(test_fold=test_split)
    X_val, y_val = data_fold_loader.load_folds(data_paths.ck_plus_data_path, [val_fold])
    X_test, y_test = data_fold_loader.load_folds(data_paths.ck_plus_data_path, [test_split])
    print X_val.shape, y_val.shape
    print X_test.shape, y_test.shape

    X_val = numpy.float32(X_val)
    X_val /= 255.0
    X_val *= 2.0

    X_test = numpy.float32(X_test)
    X_test /= 255.0
    X_test *= 2.0

    val_data_container = SupervisedDataContainer(X_val, y_val)
    test_data_container = SupervisedDataContainer(X_test, y_test)

    # Construct evaluator
    preprocessor = [util.Normer3(filter_size=5, num_channels=1)]

    checkpoint_file_list = sorted(
        glob.glob(os.path.join(checkpoint_dir, '*.pkl')))
    val_evaluator = util.Evaluator(model, val_data_container,
                                   checkpoint_file_list[0], preprocessor)
    test_evaluator = util.Evaluator(model, test_data_container,
                                    checkpoint_file_list[0], preprocessor)

    # For each checkpoint, compute the overall val accuracy
    val_accuracies = []
    for checkpoint in checkpoint_file_list:
        print 'Checkpoint: %s' % os.path.split(checkpoint)[1]
        val_evaluator.set_checkpoint(checkpoint)

ifp-uiuc / do-neural-networks-learn-faus-iccvw-2015 / ck_plus_six_class / ck_plus_single_checkpoint_evaluator.py View on Github

mask_test = numpy.logical_and(y_test != 0, y_test != 2)
    X_test = X_test[mask_test, :, :, :]
    y_test = y_test[mask_test]
    y_test = reindex_labels(y_test)
    num_test_samples = len(y_test)

    print 'Reduced Val Data: ', X_val.shape, y_val.shape
    print 'Reduced Test Data: ', X_test.shape, y_test.shape

    if test_split == 9:
        X_test, y_test = add_padding(X_test, y_test)
    elif test_split == 8:
        X_val, y_val = add_padding(X_val, y_val)

    val_data_container = SupervisedDataContainer(X_val, y_val)
    test_data_container = SupervisedDataContainer(X_test, y_test)

    # Construct evaluator
    preprocessor = [util.Normer3(filter_size=5, num_channels=1)]

    val_evaluator = util.Evaluator(model, val_data_container,
                                   checkpoint_file, preprocessor)
    test_evaluator = util.Evaluator(model, test_data_container,
                                    checkpoint_file, preprocessor)

    # For the inputted checkpoint, compute the overall test accuracy
    #accuracies = []
    print 'Checkpoint: %s' % os.path.split(checkpoint_file)[1]
    val_evaluator.set_checkpoint(checkpoint_file)
 
    if test_split != 8:
        val_accuracy = val_evaluator.run()

ifp-uiuc / do-neural-networks-learn-faus-iccvw-2015 / ck_plus_six_class / ck_plus_single_checkpoint_evaluator.py View on Github

train_folds, val_fold, _ = data_fold_loader.load_fold_assignment(test_fold=test_split)
    X_val, y_val = data_fold_loader.load_folds(data_paths.ck_plus_data_path, [val_fold])
    X_test, y_test = data_fold_loader.load_folds(data_paths.ck_plus_data_path, [test_split])
    print 'Val Data: ', X_val.shape, y_val.shape
    print 'Test Data: ', X_test.shape, y_test.shape

    X_val = numpy.float32(X_val)
    X_val /= 255.0
    X_val *= 2.0

    X_test = numpy.float32(X_test)
    X_test /= 255.0
    X_test *= 2.0

    val_data_container = SupervisedDataContainer(X_val, y_val)
    test_data_container = SupervisedDataContainer(X_test, y_test)


    # Remove samples with neutral and contempt labels
    val_mask = numpy.logical_and(y_val != 0, y_val != 2)
    X_val = X_val[val_mask, :, :, :]
    y_val = y_val[val_mask]
    y_val = reindex_labels(y_val)
    num_val_samples = len(y_val)

    mask_test = numpy.logical_and(y_test != 0, y_test != 2)
    X_test = X_test[mask_test, :, :, :]
    y_test = y_test[mask_test]
    y_test = reindex_labels(y_test)
    num_test_samples = len(y_test)

ifp-uiuc / do-neural-networks-learn-faus-iccvw-2015 / ck_plus_six_class / ck_plus_checkpoint_checker.py View on Github

train_folds, val_fold, _ = data_fold_loader.load_fold_assignment(test_fold=test_split)
    X_val, y_val = data_fold_loader.load_folds(data_paths.ck_plus_data_path, [val_fold])
    X_test, y_test = data_fold_loader.load_folds(data_paths.ck_plus_data_path, [test_split])
    print 'Val Data: ', X_val.shape, y_val.shape
    print 'Test Data: ', X_test.shape, y_test.shape

    X_val = numpy.float32(X_val)
    X_val /= 255.0
    X_val *= 2.0

    X_test = numpy.float32(X_test)
    X_test /= 255.0
    X_test *= 2.0

    val_data_container = SupervisedDataContainer(X_val, y_val)
    test_data_container = SupervisedDataContainer(X_test, y_test)


    # Remove samples with neutral and contempt labels
    val_mask = numpy.logical_and(y_val != 0, y_val != 2)
    X_val = X_val[val_mask, :, :, :]
    y_val = y_val[val_mask]
    y_val = reindex_labels(y_val)
    num_val_samples = len(y_val)

    test_mask = numpy.logical_and(y_test != 0, y_test != 2)
    X_test = X_test[test_mask, :, :, :]
    y_test = y_test[test_mask]
    y_test = reindex_labels(y_test)
    num_test_samples = len(y_test)

    print 'Reduced Val Data: ', X_val.shape, y_val.shape

ifp-uiuc / do-neural-networks-learn-faus-iccvw-2015 / ck_plus_six_class / ck_plus_checkpoint_checker.py View on Github

train_folds, val_fold, _ = data_fold_loader.load_fold_assignment(test_fold=test_split)
    X_val, y_val = data_fold_loader.load_folds(data_paths.ck_plus_data_path, [val_fold])
    X_test, y_test = data_fold_loader.load_folds(data_paths.ck_plus_data_path, [test_split])
    print 'Val Data: ', X_val.shape, y_val.shape
    print 'Test Data: ', X_test.shape, y_test.shape

    X_val = numpy.float32(X_val)
    X_val /= 255.0
    X_val *= 2.0

    X_test = numpy.float32(X_test)
    X_test /= 255.0
    X_test *= 2.0

    val_data_container = SupervisedDataContainer(X_val, y_val)
    test_data_container = SupervisedDataContainer(X_test, y_test)


    # Remove samples with neutral and contempt labels
    val_mask = numpy.logical_and(y_val != 0, y_val != 2)
    X_val = X_val[val_mask, :, :, :]
    y_val = y_val[val_mask]
    y_val = reindex_labels(y_val)
    num_val_samples = len(y_val)

    test_mask = numpy.logical_and(y_test != 0, y_test != 2)
    X_test = X_test[test_mask, :, :, :]
    y_test = y_test[test_mask]
    y_test = reindex_labels(y_test)
    num_test_samples = len(y_test)

ifp-uiuc / do-neural-networks-learn-faus-iccvw-2015 / ck_plus_six_class / ck_plus_single_checkpoint_evaluator.py View on Github

train_folds, val_fold, _ = data_fold_loader.load_fold_assignment(test_fold=test_split)
    X_val, y_val = data_fold_loader.load_folds(data_paths.ck_plus_data_path, [val_fold])
    X_test, y_test = data_fold_loader.load_folds(data_paths.ck_plus_data_path, [test_split])
    print 'Val Data: ', X_val.shape, y_val.shape
    print 'Test Data: ', X_test.shape, y_test.shape

    X_val = numpy.float32(X_val)
    X_val /= 255.0
    X_val *= 2.0

    X_test = numpy.float32(X_test)
    X_test /= 255.0
    X_test *= 2.0

    val_data_container = SupervisedDataContainer(X_val, y_val)
    test_data_container = SupervisedDataContainer(X_test, y_test)


    # Remove samples with neutral and contempt labels
    val_mask = numpy.logical_and(y_val != 0, y_val != 2)
    X_val = X_val[val_mask, :, :, :]
    y_val = y_val[val_mask]
    y_val = reindex_labels(y_val)
    num_val_samples = len(y_val)

    mask_test = numpy.logical_and(y_test != 0, y_test != 2)
    X_test = X_test[mask_test, :, :, :]
    y_test = y_test[mask_test]
    y_test = reindex_labels(y_test)
    num_test_samples = len(y_test)

    print 'Reduced Val Data: ', X_val.shape, y_val.shape

How to use the anna.datasets.supervised_data_loader.SupervisedDataContainer function in anna

To help you get started, we’ve selected a few anna examples, based on popular ways it is used in public projects.

anna

Package Health Score

Popular anna functions

Similar packages