import gradio as gr
import matplotlib
import matplotlib.pyplot as plt
import numpy as np
from sklearn.kernel_approximation import Nystroem
from sklearn.linear_model import SGDOneClassSVM
from sklearn.pipeline import make_pipeline
from sklearn.svm import OneClassSVM

md_description = """
This example shows how to approximate the solution of [sklearn.svm.OneClassSVM](https://scikit-learn.org/stable/modules/generated/sklearn.svm.OneClassSVM.html#sklearn.svm.OneClassSVM) in the case of an RBF kernel with [sklearn.linear_model.SGDOneClassSVM](https://scikit-learn.org/stable/modules/generated/sklearn.svm.OneClassSVM.html#sklearn.svm.OneClassSVM), a Stochastic Gradient Descent (SGD) version of the One-Class SVM. A kernel approximation is first used in order to apply [sklearn.linear_model.SGDOneClassSVM](https://scikit-learn.org/stable/modules/generated/sklearn.svm.OneClassSVM.html#sklearn.svm.OneClassSVM) which implements a linear One-Class SVM using SGD.
Note that [sklearn.linear_model.SGDOneClassSVM](https://scikit-learn.org/stable/modules/generated/sklearn.svm.OneClassSVM.html#sklearn.svm.OneClassSVM) scales linearly with the number of samples whereas the complexity of a kernelized [sklearn.svm.OneClassSVM](https://scikit-learn.org/stable/modules/generated/sklearn.svm.OneClassSVM.html#sklearn.svm.OneClassSVM) is at best quadratic with respect to the number of samples. It is not the purpose of this example to illustrate the benefits of such an approximation in terms of computation time but rather to show that we obtain similar results on a toy dataset.
"""

font = {"weight": "normal", "size": 15}

matplotlib.rc("font", **font)

random_state = 42
rng = np.random.RandomState(random_state)
# rng = np.random.default_rng(random_state)

# Generate train data
X = 0.3 * rng.randn(500, 2)
X_train = np.r_[X + 2, X - 2]
# Generate some regular novel observations
X = 0.3 * rng.randn(20, 2)
X_test = np.r_[X + 2, X - 2]
# Generate some abnormal novel observations
X_outliers = rng.uniform(low=-4, high=4, size=(20, 2))

xx, yy = np.meshgrid(np.linspace(-4.5, 4.5, 50), np.linspace(-4.5, 4.5, 50))

# OCSVM hyperparameters
# nu = 0.05
# gamma = 2.0


def make_regression(nu, gamma):
    clf = OneClassSVM(gamma=gamma, kernel="rbf", nu=nu)
    clf.fit(X_train)
    y_pred_train = clf.predict(X_train)
    y_pred_test = clf.predict(X_test)
    y_pred_outliers = clf.predict(X_outliers)
    n_error_train = y_pred_train[y_pred_train == -1].size
    n_error_test = y_pred_test[y_pred_test == -1].size
    n_error_outliers = y_pred_outliers[y_pred_outliers == 1].size

    Z = clf.decision_function(np.c_[xx.ravel(), yy.ravel()])
    Z = Z.reshape(xx.shape)

    # Fit the One-Class SVM using a kernel approximation and SGD
    transform = Nystroem(gamma=gamma, random_state=random_state)
    clf_sgd = SGDOneClassSVM(
        nu=nu, shuffle=True, fit_intercept=True, random_state=random_state, tol=1e-4
    )
    pipe_sgd = make_pipeline(transform, clf_sgd)
    pipe_sgd.fit(X_train)
    y_pred_train_sgd = pipe_sgd.predict(X_train)
    y_pred_test_sgd = pipe_sgd.predict(X_test)
    y_pred_outliers_sgd = pipe_sgd.predict(X_outliers)
    n_error_train_sgd = y_pred_train_sgd[y_pred_train_sgd == -1].size
    n_error_test_sgd = y_pred_test_sgd[y_pred_test_sgd == -1].size
    n_error_outliers_sgd = y_pred_outliers_sgd[y_pred_outliers_sgd == 1].size

    Z_sgd = pipe_sgd.decision_function(np.c_[xx.ravel(), yy.ravel()])
    Z_sgd = Z_sgd.reshape(xx.shape)

    def make_plot(title, curr_z):
        fig = plt.figure(figsize=(9, 6))
        ax = fig.add_subplot(111)

        ax.set_title(title)
        ax.contourf(xx, yy, curr_z, levels=np.linspace(curr_z.min(), 0, 7), cmap=plt.cm.PuBu)
        a = ax.contour(xx, yy, curr_z, levels=[0], linewidths=2, colors="darkred")
        ax.contourf(xx, yy, curr_z, levels=[0, curr_z.max()], colors="palevioletred")

        s = 20
        b1 = ax.scatter(X_train[:, 0], X_train[:, 1], c="white", s=s, edgecolors="k")
        b2 = ax.scatter(X_test[:, 0], X_test[:, 1], c="blueviolet", s=s, edgecolors="k")
        c = ax.scatter(X_outliers[:, 0], X_outliers[:, 1], c="gold", s=s, edgecolors="k")
        ax.axis("tight")
        ax.set_xlim((-4.5, 4.5))
        ax.set_ylim((-4.5, 4.5))
        ax.legend(
            [a.collections[0], b1, b2, c],
            [
                "learned frontier",
                "training observations",
                "new regular observations",
                "new abnormal observations",
            ],
            loc="upper left",
        )
        ax.set_xlabel(
            "error train: %d/%d; errors novel regular: %d/%d; errors novel abnormal: %d/%d"
            % (
                n_error_train_sgd,
                X_train.shape[0],
                n_error_test_sgd,
                X_test.shape[0],
                n_error_outliers_sgd,
                X_outliers.shape[0],
            )
        )

        return fig

    return (
        make_plot("One Class SVM", Z),
        make_plot("Online One-Class SVM", Z_sgd),
        make_example(nu, gamma),
    )


def make_example(nu, gamma):
    return f"""
    With the following code you can reproduce this example with the current values of the sliders and the same data in a notebook:

    ```python
    import numpy as np
    import matplotlib.pyplot as plt
    import matplotlib
    from sklearn.svm import OneClassSVM
    from sklearn.linear_model import SGDOneClassSVM
    from sklearn.kernel_approximation import Nystroem
    from sklearn.pipeline import make_pipeline

    font = {{"weight": "normal", "size": 15}}

    matplotlib.rc("font", **font)

    rng = np.random.RandomState(random_state)

    # Generate train data
    X = 0.3 * rng.randn(500, 2)
    X_train = np.r_[X + 2, X - 2]
    # Generate some regular novel observations
    X = 0.3 * rng.randn(20, 2)
    X_test = np.r_[X + 2, X - 2]
    # Generate some abnormal novel observations
    X_outliers = rng.uniform(low=-4, high=4, size=(20, 2))

    xx, yy = np.meshgrid(np.linspace(-4.5, 4.5, 50), np.linspace(-4.5, 4.5, 50))

    # OCSVM hyperparameters
    nu = {nu}
    gamma = {gamma}

    # Fit the One-Class SVM
    clf = OneClassSVM(gamma=gamma, kernel="rbf", nu=nu)
    clf.fit(X_train)
    y_pred_train = clf.predict(X_train)
    y_pred_test = clf.predict(X_test)
    y_pred_outliers = clf.predict(X_outliers)
    n_error_train = y_pred_train[y_pred_train == -1].size
    n_error_test = y_pred_test[y_pred_test == -1].size
    n_error_outliers = y_pred_outliers[y_pred_outliers == 1].size

    Z = clf.decision_function(np.c_[xx.ravel(), yy.ravel()])
    Z = Z.reshape(xx.shape)


    # Fit the One-Class SVM using a kernel approximation and SGD
    transform = Nystroem(gamma=gamma, random_state=random_state)
    clf_sgd = SGDOneClassSVM(
        nu=nu, shuffle=True, fit_intercept=True, random_state=random_state, tol=1e-4
    )
    pipe_sgd = make_pipeline(transform, clf_sgd)
    pipe_sgd.fit(X_train)
    y_pred_train_sgd = pipe_sgd.predict(X_train)
    y_pred_test_sgd = pipe_sgd.predict(X_test)
    y_pred_outliers_sgd = pipe_sgd.predict(X_outliers)
    n_error_train_sgd = y_pred_train_sgd[y_pred_train_sgd == -1].size
    n_error_test_sgd = y_pred_test_sgd[y_pred_test_sgd == -1].size
    n_error_outliers_sgd = y_pred_outliers_sgd[y_pred_outliers_sgd == 1].size

    Z_sgd = pipe_sgd.decision_function(np.c_[xx.ravel(), yy.ravel()])
    Z_sgd = Z_sgd.reshape(xx.shape)


    # plot the level sets of the decision function
    def make_plot(Z_curr, title):
        plt.figure(figsize=(9, 6))
        plt.title(title)
        plt.contourf(xx, yy, Z_curr, levels=np.linspace(Z_curr.min(), 0, 7), cmap=plt.cm.PuBu)
        a = plt.contour(xx, yy, Z_curr, levels=[0], linewidths=2, colors="darkred")
        plt.contourf(xx, yy, Z_curr, levels=[0, Z_curr.max()], colors="palevioletred")

        s = 20
        b1 = plt.scatter(X_train[:, 0], X_train[:, 1], c="white", s=s, edgecolors="k")
        b2 = plt.scatter(X_test[:, 0], X_test[:, 1], c="blueviolet", s=s, edgecolors="k")
        c = plt.scatter(X_outliers[:, 0], X_outliers[:, 1], c="gold", s=s, edgecolors="k")
        plt.axis("tight")
        plt.xlim((-4.5, 4.5))
        plt.ylim((-4.5, 4.5))
        plt.legend(
            [a.collections[0], b1, b2, c],
            [
                "learned frontier",
                "training observations",
                "new regular observations",
                "new abnormal observations",
            ],
            loc="upper left",
        )
        plt.xlabel(
            "error train: %d/%d; errors novel regular: %d/%d; errors novel abnormal: %d/%d"
            % (
                n_error_train,
                X_train.shape[0],
                n_error_test,
                X_test.shape[0],
                n_error_outliers,
                X_outliers.shape[0],
            )
        )
        plt.show()


    make_plot(Z, "One-Class SVM")
    make_plot(Z_sgd, "Online One-Class SVM")

    ```
    """


with gr.Blocks() as demo:
    with gr.Row():
        gr.Markdown("# One-Class SVM versus One-Class SVM using Stochastic Gradient Descent")
    with gr.Row():
        with gr.Column():
            gr.Markdown(md_description)
        with gr.Column():
            slider_nu = gr.Slider(minimum=0.01, maximum=1, label="Nu", step=0.025, value=0.05)
            slider_gamma = gr.Slider(minimum=0.1, maximum=3, label="Gamma", step=0.1, value=2.0)
            button = gr.Button("Generate")
    with gr.Row():
        with gr.Column():
            plot1 = gr.Plot(label="Output")
        with gr.Column():
            plot2 = gr.Plot(label="Output")
    with gr.Row():
        example = gr.Markdown("")

    slider_nu.change(
        fn=make_regression, inputs=[slider_nu, slider_gamma], outputs=[plot1, plot2, example]
    )
    slider_gamma.change(
        fn=make_regression, inputs=[slider_nu, slider_gamma], outputs=[plot1, plot2, example]
    )
    button.click(make_regression, inputs=[slider_nu, slider_gamma], outputs=[plot1, plot2, example])
    demo.load(make_regression, inputs=[slider_nu, slider_gamma], outputs=[plot1, plot2, example])

demo.launch()