python类clf()的实例源码

def plot_log():
    pylab.clf()
    pylab.figure(num=None, figsize=(6, 5))

    x = np.arange(0.001, 1, 0.001)
    y = np.log(x)

    pylab.title('Relationship between probabilities and their logarithm')
    pylab.plot(x, y)
    pylab.grid(True)
    pylab.xlabel('P')
    pylab.ylabel('log(P)')
    filename = 'log_probs.png'
    pylab.savefig(os.path.join(CHART_DIR, filename), bbox_inches="tight")

def plot_feat_hist(data_name_list, filename=None):
    pylab.clf()
    num_rows = 1 + (len(data_name_list) - 1) / 2
    num_cols = 1 if len(data_name_list) == 1 else 2
    pylab.figure(figsize=(5 * num_cols, 4 * num_rows))

    for i in range(num_rows):
        for j in range(num_cols):
            pylab.subplot(num_rows, num_cols, 1 + i * num_cols + j)
            x, name = data_name_list[i * num_cols + j]
            pylab.title(name)
            pylab.xlabel('Value')
            pylab.ylabel('Density')
            # the histogram of the data
            max_val = np.max(x)
            if max_val <= 1.0:
                bins = 50
            elif max_val > 50:
                bins = 50
            else:
                bins = max_val
            n, bins, patches = pylab.hist(
                x, bins=bins, normed=1, facecolor='green', alpha=0.75)

            pylab.grid(True)

    if not filename:
        filename = "feat_hist_%s.png" % name

    pylab.savefig(os.path.join(CHART_DIR, filename), bbox_inches="tight")

def log_false_positives(clf, X, y, name):
    with open("FP_" + name.replace(" ", "_") + ".tsv", "w") as f:
        false_positive = clf.predict(X) != y
        for tweet, false_class in zip(X[false_positive], y[false_positive]):
            f.write("%s\t%s\n" %
                    (false_class, tweet.encode("ascii", "ignore")))

def plot_pr(auc_score, name, precision, recall, label=None):
    pylab.clf()
    pylab.figure(num=None, figsize=(5, 4))
    pylab.grid(True)
    pylab.fill_between(recall, precision, alpha=0.5)
    pylab.plot(recall, precision, lw=1)
    pylab.xlim([0.0, 1.0])
    pylab.ylim([0.0, 1.0])
    pylab.xlabel('Recall')
    pylab.ylabel('Precision')
    pylab.title('P/R curve (AUC = %0.2f) / %s' % (auc_score, label))
    filename = name.replace(" ", "_")
    pylab.savefig(
        os.path.join(CHART_DIR, "pr_" + filename + ".png"), bbox_inches="tight")

def show_most_informative_features(vectorizer, clf, n=20):
    c_f = sorted(zip(clf.coef_[0], vectorizer.get_feature_names()))
    top = zip(c_f[:n], c_f[:-(n + 1):-1])
    for (c1, f1), (c2, f2) in top:
        print("\t%.4f\t%-15s\t\t%.4f\t%-15s" % (c1, f1, c2, f2))

def plot_feat_hist(data_name_list, filename=None):
    pylab.clf()
    num_rows = 1 + (len(data_name_list) - 1) / 2
    num_cols = 1 if len(data_name_list) == 1 else 2
    pylab.figure(figsize=(5 * num_cols, 4 * num_rows))

    for i in range(num_rows):
        for j in range(num_cols):
            pylab.subplot(num_rows, num_cols, 1 + i * num_cols + j)
            x, name = data_name_list[i * num_cols + j]
            pylab.title(name)
            pylab.xlabel('Value')
            pylab.ylabel('Density')
            # the histogram of the data
            max_val = np.max(x)
            if max_val <= 1.0:
                bins = 50
            elif max_val > 50:
                bins = 50
            else:
                bins = max_val
            n, bins, patches = pylab.hist(
                x, bins=bins, normed=1, facecolor='green', alpha=0.75)

            pylab.grid(True)

    if not filename:
        filename = "feat_hist_%s.png" % name

    pylab.savefig(os.path.join(CHART_DIR, filename), bbox_inches="tight")

def plot_bias_variance(data_sizes, train_errors, test_errors, name):
    pylab.clf()
    pylab.ylim([0.0, 1.0])
    pylab.xlabel('Data set size')
    pylab.ylabel('Error')
    pylab.title("Bias-Variance for '%s'" % name)
    pylab.plot(
        data_sizes, train_errors, "-", data_sizes, test_errors, "--", lw=1)
    pylab.legend(["train error", "test error"], loc="upper right")
    pylab.grid(True)
    pylab.savefig(os.path.join(CHART_DIR, "bv_" + name + ".png"))

def plot_forest_all_proba(y_proba_all, y_gt):
    from matplotlib import pylab
    N = len(y_gt)
    num_tree = len(y_proba_all)
    pylab.clf()
    mat = np.zeros((num_tree, N))
    LOGGER.info('mat.shape={}'.format(mat.shape))
    for i in range(num_tree):
        mat[i,:] = y_proba_all[i][(range(N), y_gt)]
    pylab.matshow(mat, fignum=False, cmap='Blues', vmin=0, vmax=1.0)
    pylab.grid(False)
    pylab.show()

def twohists(x1,x2,xmin,xmax,range=None,x1leg='$x_1$',x2leg='$x_2$',xlabel='',fig=1,sharey=False,fontsize=12,bins1=10,bins2=10):
    """
Script that plots two histograms of quantities x1 and x2
sharing the same X-axis.

:param x1,x2: arrays with data to be plotted
:param xmin,xmax: lower and upper range of plotted values, will be used to set a consistent x-range
    for both histograms.
:param x1leg, x2leg: legends for each histogram 
:param xlabel: self-explanatory.
:param bins1,bins2: number of bins in each histogram
:param fig: which plot window should I use?
:param range: in the form (xmin,xmax), same as range argument for hist and applied to both
    histograms.

Inspired by `Scipy <http://www.scipy.org/Cookbook/Matplotlib/Multiple_Subplots_with_One_Axis_Label>`_.
    """

    pylab.rcParams.update({'font.size': fontsize})
    fig=pylab.figure(fig)
    pylab.clf()

    a=fig.add_subplot(2,1,1)
    if sharey==True:
        b=fig.add_subplot(2,1,2, sharex=a, sharey=a)
    else:
        b=fig.add_subplot(2,1,2, sharex=a)

    a.hist(x1,bins1,label=x1leg,color='b',histtype='stepfilled',range=range)
    a.legend(loc='best',frameon=False)
    a.set_xlim(xmin,xmax)

    b.hist(x2,bins2,label=x2leg,color='r',histtype='stepfilled',range=range)
    b.legend(loc='best',frameon=False)

    pylab.setp(a.get_xticklabels(), visible=False)

    b.set_xlabel(xlabel)
    b.set_ylabel('Number',verticalalignment='bottom')
    pylab.minorticks_on()
    pylab.subplots_adjust(hspace=0.15)
    pylab.draw()
    pylab.show()

def threehists(x1,x2,x3,xmin,xmax,x1leg='$x_1$',x2leg='$x_2$',x3leg='$x_3$',xlabel='',fig=1,sharey=False,fontsize=12):
    """
Script that plots three histograms of quantities x1, x2 and x3 
sharing the same X-axis.

Arguments:
- x1,x2,x3: arrays with data to be plotted
- xmin,xmax: lower and upper range of plotted values, will be used to set a consistent x-range for both histograms.
- x1leg, x2leg, x3leg: legends for each histogram   
- xlabel: self-explanatory.
- sharey: sharing the Y-axis among the histograms?
- fig: which plot window should I use?

Example:
x1=Lbol(AD), x2=Lbol(JD), x3=Lbol(EHF10)

>>> threehists(x1,x2,x3,38,44,'AD','JD','EHF10','$\log L_{\\rm bol}$ (erg s$^{-1}$)',sharey=True)

Inspired by `Scipy <http://www.scipy.org/Cookbook/Matplotlib/Multiple_Subplots_with_One_Axis_Label>`_.
    """
    pylab.rcParams.update({'font.size': fontsize})
    fig=pylab.figure(fig)
    pylab.clf()

    a=fig.add_subplot(3,1,1)
    if sharey==True:
        b=fig.add_subplot(3,1,2, sharex=a, sharey=a)
        c=fig.add_subplot(3,1,3, sharex=a, sharey=a)
    else:
        b=fig.add_subplot(3,1,2, sharex=a)
        c=fig.add_subplot(3,1,3, sharex=a)      

    a.hist(x1,label=x1leg,color='b',histtype='stepfilled')
    a.legend(loc='best',frameon=False)
    a.set_xlim(xmin,xmax)

    b.hist(x2,label=x2leg,color='r',histtype='stepfilled')
    b.legend(loc='best',frameon=False)

    c.hist(x3,label=x3leg,color='y',histtype='stepfilled')
    c.legend(loc='best',frameon=False)

    pylab.setp(a.get_xticklabels(), visible=False)
    pylab.setp(b.get_xticklabels(), visible=False)

    c.set_xlabel(xlabel)
    b.set_ylabel('Number')
    pylab.minorticks_on()
    pylab.subplots_adjust(hspace=0.15)
    pylab.draw()
    pylab.show()

def fourcumplot(x1,x2,x3,x4,xmin,xmax,x1leg='$x_1$',x2leg='$x_2$',x3leg='$x_3$',x4leg='$x_3$',xlabel='',ylabel='$N(x>x\')$',fig=1,sharey=False,fontsize=12,bins1=50,bins2=50,bins3=50,bins4=50):
    """
Script that plots the cumulative histograms of four variables x1, x2, x3 and x4
sharing the same X-axis. For each bin, Y is the fraction of the sample 
with values above X.

Arguments:

- x1,x2,x3,x4: arrays with data to be plotted
- xmin,xmax: lower and upper range of plotted values, will be used to set a consistent x-range
for both histograms.
- x1leg, x2leg, x3leg, x4leg: legends for each histogram    
- xlabel: self-explanatory.
- sharey: sharing the Y-axis among the histograms?
- bins1,bins2,...: number of bins in each histogram
- fig: which plot window should I use?

Inspired by `Scipy <http://www.scipy.org/Cookbook/Matplotlib/Multiple_Subplots_with_One_Axis_Label>`_.

v1 Jun. 2012: inherited from fourhists.
    """
    pylab.rcParams.update({'font.size': fontsize})
    fig=pylab.figure(fig)
    pylab.clf()

    a=fig.add_subplot(4,1,1)
    if sharey==True:
        b=fig.add_subplot(4,1,2, sharex=a, sharey=a)
        c=fig.add_subplot(4,1,3, sharex=a, sharey=a)
        d=fig.add_subplot(4,1,4, sharex=a, sharey=a)
    else:
        b=fig.add_subplot(4,1,2, sharex=a)
        c=fig.add_subplot(4,1,3, sharex=a)      
        d=fig.add_subplot(4,1,4, sharex=a)

    a.hist(x1,bins1,label=x1leg,color='b',cumulative=-True,normed=True,histtype='stepfilled')
    a.legend(loc='best',frameon=False)
    a.set_xlim(xmin,xmax)

    b.hist(x2,bins2,label=x2leg,color='r',cumulative=-True,normed=True,histtype='stepfilled')
    b.legend(loc='best',frameon=False)

    c.hist(x3,bins3,label=x3leg,color='y',cumulative=-True,normed=True,histtype='stepfilled')
    c.legend(loc='best',frameon=False)

    d.hist(x4,bins4,label=x4leg,color='g',cumulative=-True,normed=True,histtype='stepfilled')
    d.legend(loc='best',frameon=False)

    pylab.setp(a.get_xticklabels(), visible=False)
    pylab.setp(b.get_xticklabels(), visible=False)
    pylab.setp(c.get_xticklabels(), visible=False)

    d.set_xlabel(xlabel)
    c.set_ylabel(ylabel)
    pylab.minorticks_on()
    pylab.subplots_adjust(hspace=0.15)
    pylab.draw()
    pylab.show()

def plotPolicy(self, policy, prefix):
        plt.clf()
        for idx in xrange(len(policy)):
            i, j = self.env.getStateXY(idx)

            dx = 0
            dy = 0
            if policy[idx] == 0: # up
                dy = 0.35
            elif policy[idx] == 1: #right
                dx = 0.35
            elif policy[idx] == 2: #down
                dy = -0.35
            elif policy[idx] == 3: #left
                dx = -0.35
            elif self.matrixMDP[i][j] != -1 and policy[idx] == 4: # termination
                circle = plt.Circle(
                    (j + 0.5, self.numRows - i + 0.5 - 1), 0.025, color='k')
                plt.gca().add_artist(circle)

            if self.matrixMDP[i][j] != -1:
                plt.arrow(j + 0.5, self.numRows - i + 0.5 - 1, dx, dy,
                    head_width=0.05, head_length=0.05, fc='k', ec='k')
            else:
                plt.gca().add_patch(
                    patches.Rectangle(
                    (j, self.numRows - i - 1), # (x,y)
                    1.0,                   # width
                    1.0,                   # height
                    facecolor = "gray"
                    )
                )

        plt.xlim([0, self.numCols])
        plt.ylim([0, self.numRows])


        for i in xrange(self.numCols):
            plt.axvline(i, color='k', linestyle=':')
        plt.axvline(self.numCols, color='k', linestyle=':')

        for j in xrange(self.numRows):
            plt.axhline(j, color='k', linestyle=':')
        plt.axhline(self.numRows, color='k', linestyle=':')

        plt.savefig(self.outputPath + prefix + 'policy.png')
        plt.close()

def plot_mi_demo():
    np.random.seed(0)  # to reproduce the data later on
    pylab.clf()
    pylab.figure(num=None, figsize=(8, 8))

    x = np.arange(0, 10, 0.2)

    pylab.subplot(221)
    y = 0.5 * x + norm.rvs(1, scale=.01, size=len(x))
    _plot_mi_func(x, y)

    pylab.subplot(222)
    y = 0.5 * x + norm.rvs(1, scale=.1, size=len(x))
    _plot_mi_func(x, y)

    pylab.subplot(223)
    y = 0.5 * x + norm.rvs(1, scale=1, size=len(x))
    _plot_mi_func(x, y)

    pylab.subplot(224)
    y = norm.rvs(1, scale=10, size=len(x))
    _plot_mi_func(x, y)

    pylab.autoscale(tight=True)
    pylab.grid(True)

    filename = "mi_demo_1.png"
    pylab.savefig(os.path.join(CHART_DIR, filename), bbox_inches="tight")

    pylab.clf()
    pylab.figure(num=None, figsize=(8, 8))

    x = np.arange(-5, 5, 0.2)

    pylab.subplot(221)
    y = 0.5 * x ** 2 + norm.rvs(1, scale=.01, size=len(x))
    _plot_mi_func(x, y)

    pylab.subplot(222)
    y = 0.5 * x ** 2 + norm.rvs(1, scale=.1, size=len(x))
    _plot_mi_func(x, y)

    pylab.subplot(223)
    y = 0.5 * x ** 2 + norm.rvs(1, scale=1, size=len(x))
    _plot_mi_func(x, y)

    pylab.subplot(224)
    y = 0.5 * x ** 2 + norm.rvs(1, scale=10, size=len(x))
    _plot_mi_func(x, y)

    pylab.autoscale(tight=True)
    pylab.grid(True)

    filename = "mi_demo_2.png"
    pylab.savefig(os.path.join(CHART_DIR, filename), bbox_inches="tight")

def plot_correlation_demo():
    np.random.seed(0)  # to reproduce the data later on
    pylab.clf()
    pylab.figure(num=None, figsize=(8, 8))

    x = np.arange(0, 10, 0.2)

    pylab.subplot(221)
    y = 0.5 * x + norm.rvs(1, scale=.01, size=len(x))
    _plot_correlation_func(x, y)

    pylab.subplot(222)
    y = 0.5 * x + norm.rvs(1, scale=.1, size=len(x))
    _plot_correlation_func(x, y)

    pylab.subplot(223)
    y = 0.5 * x + norm.rvs(1, scale=1, size=len(x))
    _plot_correlation_func(x, y)

    pylab.subplot(224)
    y = norm.rvs(1, scale=10, size=len(x))
    _plot_correlation_func(x, y)

    pylab.autoscale(tight=True)
    pylab.grid(True)

    filename = "corr_demo_1.png"
    pylab.savefig(os.path.join(CHART_DIR, filename), bbox_inches="tight")

    pylab.clf()
    pylab.figure(num=None, figsize=(8, 8))

    x = np.arange(-5, 5, 0.2)

    pylab.subplot(221)
    y = 0.5 * x ** 2 + norm.rvs(1, scale=.01, size=len(x))
    _plot_correlation_func(x, y)

    pylab.subplot(222)
    y = 0.5 * x ** 2 + norm.rvs(1, scale=.1, size=len(x))
    _plot_correlation_func(x, y)

    pylab.subplot(223)
    y = 0.5 * x ** 2 + norm.rvs(1, scale=1, size=len(x))
    _plot_correlation_func(x, y)

    pylab.subplot(224)
    y = 0.5 * x ** 2 + norm.rvs(1, scale=10, size=len(x))
    _plot_correlation_func(x, y)

    pylab.autoscale(tight=True)
    pylab.grid(True)

    filename = "corr_demo_2.png"
    pylab.savefig(os.path.join(CHART_DIR, filename), bbox_inches="tight")

def plot_simple_demo_1():
    pylab.clf()
    fig = pylab.figure(num=None, figsize=(10, 4))
    pylab.subplot(121)

    title = "Original feature space"
    pylab.title(title)
    pylab.xlabel("$X_1$")
    pylab.ylabel("$X_2$")

    x1 = np.arange(0, 10, .2)
    x2 = x1 + np.random.normal(scale=1, size=len(x1))

    good = (x1 > 5) | (x2 > 5)
    bad = ~good

    x1g = x1[good]
    x2g = x2[good]
    pylab.scatter(x1g, x2g, edgecolor="blue", facecolor="blue")

    x1b = x1[bad]
    x2b = x2[bad]
    pylab.scatter(x1b, x2b, edgecolor="red", facecolor="white")

    pylab.grid(True)

    pylab.subplot(122)

    X = np.c_[(x1, x2)]

    pca = decomposition.PCA(n_components=1)
    Xtrans = pca.fit_transform(X)

    Xg = Xtrans[good]
    Xb = Xtrans[bad]

    pylab.scatter(
        Xg[:, 0], np.zeros(len(Xg)), edgecolor="blue", facecolor="blue")
    pylab.scatter(
        Xb[:, 0], np.zeros(len(Xb)), edgecolor="red", facecolor="white")
    title = "Transformed feature space"
    pylab.title(title)
    pylab.xlabel("$X'$")
    fig.axes[1].get_yaxis().set_visible(False)

    print(pca.explained_variance_ratio_)

    pylab.grid(True)

    pylab.autoscale(tight=True)
    filename = "pca_demo_1.png"
    pylab.savefig(os.path.join(CHART_DIR, filename), bbox_inches="tight")

def plot_simple_demo_2():
    pylab.clf()
    fig = pylab.figure(num=None, figsize=(10, 4))
    pylab.subplot(121)

    title = "Original feature space"
    pylab.title(title)
    pylab.xlabel("$X_1$")
    pylab.ylabel("$X_2$")

    x1 = np.arange(0, 10, .2)
    x2 = x1 + np.random.normal(scale=1, size=len(x1))

    good = x1 > x2
    bad = ~good

    x1g = x1[good]
    x2g = x2[good]
    pylab.scatter(x1g, x2g, edgecolor="blue", facecolor="blue")

    x1b = x1[bad]
    x2b = x2[bad]
    pylab.scatter(x1b, x2b, edgecolor="red", facecolor="white")

    pylab.grid(True)

    pylab.subplot(122)

    X = np.c_[(x1, x2)]

    pca = decomposition.PCA(n_components=1)
    Xtrans = pca.fit_transform(X)

    Xg = Xtrans[good]
    Xb = Xtrans[bad]

    pylab.scatter(
        Xg[:, 0], np.zeros(len(Xg)), edgecolor="blue", facecolor="blue")
    pylab.scatter(
        Xb[:, 0], np.zeros(len(Xb)), edgecolor="red", facecolor="white")
    title = "Transformed feature space"
    pylab.title(title)
    pylab.xlabel("$X'$")
    fig.axes[1].get_yaxis().set_visible(False)

    print(pca.explained_variance_ratio_)

    pylab.grid(True)

    pylab.autoscale(tight=True)
    filename = "pca_demo_2.png"
    pylab.savefig(os.path.join(CHART_DIR, filename), bbox_inches="tight")