Skip to content

Get Accuracy Method

Functions

get_accuracy(model, X, Y)

Calculate the accuracy of the model's predictions.

Parameters:

Name Type Description Default
model Sequential

The neural network model.

 required
X ndarray

Input data, shape (n_features, n_samples).

 required
Y ndarray

True labels in one-hot encoded format, shape (n_classes, n_samples).

 required

Returns:

Name Type Description
float

The accuracy of the model's predictions, as a value between 0 and 1.
Example 
model = Sequential([
    Dense(64, activation='relu', input_shape=(784,)),
    Dense(10, activation='softmax')
])
X_test = np.random.randn(784, 100)
Y_test = np.eye(10)[np.random.randint(0, 10, 100)].T
accuracy = get_accuracy(model, X_test, Y_test)
print(f"Model accuracy: {accuracy}")
Note
  • This function assumes that the model's output and true labels are in one-hot encoded format.
  • The accuracy is calculated as the ratio of correct predictions to total predictions.
Source code in microkeras/operations/metrics/get_accuracy.py 
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
def get_accuracy(model, X, Y):
    """
    Calculate the accuracy of the model's predictions.

    Args:
        model (Sequential): The neural network model.
        X (numpy.ndarray): Input data, shape (n_features, n_samples).
        Y (numpy.ndarray): True labels in one-hot encoded format, shape (n_classes, n_samples).

    Returns:
        float: The accuracy of the model's predictions, as a value between 0 and 1.

    Example:
        ```python
        model = Sequential([
            Dense(64, activation='relu', input_shape=(784,)),
            Dense(10, activation='softmax')
        ])
        X_test = np.random.randn(784, 100)
        Y_test = np.eye(10)[np.random.randint(0, 10, 100)].T
        accuracy = get_accuracy(model, X_test, Y_test)
        print(f"Model accuracy: {accuracy}")
        ```

    Note:
        - This function assumes that the model's output and true labels are in one-hot encoded format.
        - The accuracy is calculated as the ratio of correct predictions to total predictions.
    """
    forward(model, X)
    predictions = np.argmax(model.layers[-1].A, axis=0)
    Y_decoded = np.argmax(Y, axis=0)
    return np.sum(predictions == Y_decoded) / Y_decoded.size