ea5d191f7f71b6fa927a96ef4e785c762a54d5db,examples/decoding/plot_oasis_vbm_space_net.py,,,#,14

Before Change

////// Fit and predict //////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
from nilearn.decoding import SpaceNet
decoder = SpaceNet(memory=memory, screening_percentile=10, verbose=1,
                   mask=nifti_masker, n_jobs=14)

////// Fit and predict
decoder.fit(new_images, age)
coef_niimg = decoder.coef_img_
age_pred = decoder.predict(new_images).ravel()

////// Visualization //////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
import matplotlib.pyplot as plt
from nilearn.plotting import plot_stat_map

// weights map
background_img = dataset_files.gray_matter_maps[0]
plot_stat_map(coef_img, background_img, title="Graph-Net weights",
              display_mode="z")

// quality of predictions
plt.figure()
linewidth = 3
ax1 = plt.subplot("211")
ax1.plot(age, label="True age", linewidth=linewidth)
ax1.plot(age_pred, "--", c="g", label="Fitted age", linewidth=linewidth)
ax1.set_ylabel("age")
plt.legend(loc="best")
ax2 = plt.subplot("212")
ax2.plot(age - age_pred, label="True age - fitted age", linewidth=linewidth)
ax2.set_xlabel("subject")
plt.legend(loc="best")

plt.show()

After Change

from nilearn.decoding import SpaceNetRegressor
import matplotlib.pyplot as plt
from nilearn.plotting import plot_stat_map
for penalty in ["tv-l1", "smooth-lasso"]:
    decoder = SpaceNetRegressor(memory="cache", penalty=penalty, verbose=2)
    decoder.fit(X_train, y_train)  // fit
    coef_img = decoder.coef_img_
    y_pred = decoder.predict(X_test).ravel()  // predict
    mse = np.mean(np.abs(y_test - y_pred))

    ////// Visualization //////////////////////////////////////////////////////////////////////////////////////////////////////////////////
    // weights map
    background_img = X[0]
    plot_stat_map(coef_img, background_img, title="%s weights" % penalty,
                  display_mode="z", cut_coords=3)

    // quality of predictions
    plt.figure()
    plt.suptitle("%s: Mean Absolute Error %.2f years" % (penalty, mse))
    linewidth = 3
    ax1 = plt.subplot("211")
    ax1.plot(y_test, label="True age", linewidth=linewidth)
    ax1.plot(y_pred, "--", c="g", label="Predicted age", linewidth=linewidth)
    ax1.set_ylabel("age")
    plt.legend(loc="best")
    ax2 = plt.subplot("212")
    ax2.plot(y_test - y_pred, label="True age - predicted age",
             linewidth=linewidth)
    ax2.set_xlabel("subject")
    plt.legend(loc="best")

plt.show()

Instances