How do I create a confusion matrix to evaluate the model?

Question

I am following This Tutorial for Music Genre Classifier. And now I want to visualize the predictions made on test set using confusion matrix. How do I go about creating it? Do I just predict using model.predict() command and add it in the confusion matrix? If so, what should be my other parameter? Please explain as I am not understanding how to properly visualize the results. Below is the code:

import json
import numpy as np
from sklearn import metrics
from sklearn.model_selection import train_test_split
import tensorflow.keras as keras
import matplotlib.pyplot as plt

# path to json file that stores MFCCs and genre labels for each processed segment
DATA_PATH = "data_10.json"

def load_data(data_path):

    with open(data_path, "r") as fp:
        data = json.load(fp)

    # convert lists to numpy arrays
    X = np.array(data["mfcc"])
    y = np.array(data["labels"])

    print("Data succesfully loaded!")

    return  X, y

def plot_history(history):
    """Plots accuracy/loss for training/validation set as a function of the epochs
        :param history: Training history of model
        :return:
    """

    fig, axs = plt.subplots(2)

    # create accuracy sublpot
    axs[0].plot(history.history["accuracy"], label="train accuracy")
    axs[0].plot(history.history["val_accuracy"], label="test accuracy")
    axs[0].set_ylabel("Accuracy")
    axs[0].legend(loc="lower right")
    axs[0].set_title("Accuracy eval")

    # create error sublpot
    axs[1].plot(history.history["loss"], label="train error")
    axs[1].plot(history.history["val_loss"], label="test error")
    axs[1].set_ylabel("Error")
    axs[1].set_xlabel("Epoch")
    axs[1].legend(loc="upper right")
    axs[1].set_title("Error eval")

    plt.show()

if __name__ == "__main__":

    # load data
    X, y = load_data(DATA_PATH)

    # create train/test split
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3)

    # build network topology
    model = keras.Sequential([

        # input layer
        keras.layers.Flatten(input_shape=(X.shape[1], X.shape[2])),

        # 1st dense layer
        keras.layers.Dense(512, activation='relu', kernel_regularizer=keras.regularizers.l2(0.001)),

        # 2nd dense layer
        keras.layers.Dense(256, activation='relu', kernel_regularizer=keras.regularizers.l2(0.001)),

        # 3rd dense layer
        keras.layers.Dense(64, activation='relu', kernel_regularizer=keras.regularizers.l2(0.001)),

        # output layer
        keras.layers.Dense(10, activation='softmax')
    ])

    # compile model
    optimiser = keras.optimizers.Adam(learning_rate=0.0001)
    model.compile(optimizer=optimiser,
                  loss='sparse_categorical_crossentropy',
                  metrics=['accuracy'])

    model.summary()

    # train model
    history = model.fit(X_train, y_train, validation_data=(X_test, y_test), batch_size=32, epochs=100)

    plot_history(history)

Answer 1

Given that the model was trained properly you need to do the following:

from sklearn.metrics import confusion_matrix

y_pred = model.predict(X_test)
y_pred = np.argmax(y_pred, axis=1)

conf_mat = confusion_matrix(y_test, y_pred)

First, get the prediction (probability vector because you are using Softmax), then use np.argmax to get the most probable class. Once you have that just pass it to confusion_matrix.