Merge branch 'devel' of github.com:SheffieldML/GPy into devel

GPy/core/gp_base.py

@@ -33,7 +33,7 @@ class GPBase(model.model):
         # All leaf nodes should call self._set_params(self._get_params()) at
         # the end
 
-    def plot_f(self, samples=0, plot_limits=None, which_data='all', which_parts='all', resolution=None, full_cov=False):
+    def plot_f(self, samples=0, plot_limits=None, which_data='all', which_parts='all', resolution=None, full_cov=False, fignum=None, ax=None):
         """
         Plot the GP's view of the world, where the data is normalized and the
         likelihood is Gaussian.
@@ -57,46 +57,55 @@ class GPBase(model.model):
         if which_data == 'all':
             which_data = slice(None)
 
+        if ax is None:
+            fig = pb.figure(num=fignum)
+            ax = fig.add_subplot(111)
+
         if self.X.shape[1] == 1:
             Xnew, xmin, xmax = x_frame1D(self.X, plot_limits=plot_limits)
             if samples == 0:
                 m, v = self._raw_predict(Xnew, which_parts=which_parts)
-                gpplot(Xnew, m, m - 2 * np.sqrt(v), m + 2 * np.sqrt(v))
+                gpplot(Xnew, m, m - 2 * np.sqrt(v), m + 2 * np.sqrt(v), axes=ax)
-                pb.plot(self.X[which_data], self.likelihood.Y[which_data], 'kx', mew=1.5)
+                ax.plot(self.X[which_data], self.likelihood.Y[which_data], 'kx', mew=1.5)
             else:
                 m, v = self._raw_predict(Xnew, which_parts=which_parts, full_cov=True)
                 Ysim = np.random.multivariate_normal(m.flatten(), v, samples)
-                gpplot(Xnew, m, m - 2 * np.sqrt(np.diag(v)[:, None]), m + 2 * np.sqrt(np.diag(v))[:, None])
+                gpplot(Xnew, m, m - 2 * np.sqrt(np.diag(v)[:, None]), m + 2 * np.sqrt(np.diag(v))[:, None,], axes=ax)
                 for i in range(samples):
-                    pb.plot(Xnew, Ysim[i, :], Tango.colorsHex['darkBlue'], linewidth=0.25)
+                    ax.plot(Xnew, Ysim[i, :], Tango.colorsHex['darkBlue'], linewidth=0.25)
-            pb.plot(self.X[which_data], self.likelihood.Y[which_data], 'kx', mew=1.5)
+            ax.plot(self.X[which_data], self.likelihood.Y[which_data], 'kx', mew=1.5)
-            pb.xlim(xmin, xmax)
+            ax.set_xlim(xmin, xmax)
             ymin, ymax = min(np.append(self.likelihood.Y, m - 2 * np.sqrt(np.diag(v)[:, None]))), max(np.append(self.likelihood.Y, m + 2 * np.sqrt(np.diag(v)[:, None])))
             ymin, ymax = ymin - 0.1 * (ymax - ymin), ymax + 0.1 * (ymax - ymin)
-            pb.ylim(ymin, ymax)
+            ax.set_ylim(ymin, ymax)
 
         elif self.X.shape[1] == 2:
             resolution = resolution or 50
             Xnew, xmin, xmax, xx, yy = x_frame2D(self.X, plot_limits, resolution)
             m, v = self._raw_predict(Xnew, which_parts=which_parts)
             m = m.reshape(resolution, resolution).T
-            pb.contour(xx, yy, m, vmin=m.min(), vmax=m.max(), cmap=pb.cm.jet)
+            ax.contour(xx, yy, m, vmin=m.min(), vmax=m.max(), cmap=pb.cm.jet)
-            pb.scatter(self.X[:, 0], self.X[:, 1], 40, self.likelihood.Y, linewidth=0, cmap=pb.cm.jet, vmin=m.min(), vmax=m.max())
+            ax.scatter(self.X[:, 0], self.X[:, 1], 40, self.likelihood.Y, linewidth=0, cmap=pb.cm.jet, vmin=m.min(), vmax=m.max())
-            pb.xlim(xmin[0], xmax[0])
+            ax.set_xlim(xmin[0], xmax[0])
-            pb.ylim(xmin[1], xmax[1])
+            ax.set_ylim(xmin[1], xmax[1])
         else:
             raise NotImplementedError, "Cannot define a frame with more than two input dimensions"
 
-    def plot(self, samples=0, plot_limits=None, which_data='all', which_parts='all', resolution=None, levels=20):
+    def plot(self, plot_limits=None, which_data='all', which_parts='all', resolution=None, levels=20, samples=0, fignum=None, ax=None):
         """
         TODO: Docstrings!
         :param levels: for 2D plotting, the number of contour levels to use
+        is ax is None, create a new figure
 
         """
         # TODO include samples
         if which_data == 'all':
             which_data = slice(None)
 
+        if ax is None:
+            fig = pb.figure(num=fignum)
+            ax = fig.add_subplot(111)
+
         if self.X.shape[1] == 1:
 
             Xu = self.X * self._Xstd + self._Xmean  # NOTE self.X are the normalized values now
@@ -104,12 +113,12 @@ class GPBase(model.model):
             Xnew, xmin, xmax = x_frame1D(Xu, plot_limits=plot_limits)
             m, var, lower, upper = self.predict(Xnew, which_parts=which_parts)
             for d in range(m.shape[1]):
-                gpplot(Xnew, m[:,d], lower[:,d], upper[:,d])
+                gpplot(Xnew, m[:,d], lower[:,d], upper[:,d],axes=ax)
-                pb.plot(Xu[which_data], self.likelihood.data[which_data,d], 'kx', mew=1.5)
+                ax.plot(Xu[which_data], self.likelihood.data[which_data,d], 'kx', mew=1.5)
             ymin, ymax = min(np.append(self.likelihood.data, lower)), max(np.append(self.likelihood.data, upper))
             ymin, ymax = ymin - 0.1 * (ymax - ymin), ymax + 0.1 * (ymax - ymin)
-            pb.xlim(xmin, xmax)
+            ax.set_xlim(xmin, xmax)
-            pb.ylim(ymin, ymax)
+            ax.set_ylim(ymin, ymax)
 
         elif self.X.shape[1] == 2:  # FIXME
             resolution = resolution or 50
@@ -117,11 +126,11 @@ class GPBase(model.model):
             x, y = np.linspace(xmin[0], xmax[0], resolution), np.linspace(xmin[1], xmax[1], resolution)
             m, var, lower, upper = self.predict(Xnew, which_parts=which_parts)
             m = m.reshape(resolution, resolution).T
-            pb.contour(x, y, m, levels, vmin=m.min(), vmax=m.max(), cmap=pb.cm.jet)
+            ax.contour(x, y, m, levels, vmin=m.min(), vmax=m.max(), cmap=pb.cm.jet)
             Yf = self.likelihood.Y.flatten()
-            pb.scatter(self.X[:, 0], self.X[:, 1], 40, Yf, cmap=pb.cm.jet, vmin=m.min(), vmax=m.max(), linewidth=0.)
+            ax.scatter(self.X[:, 0], self.X[:, 1], 40, Yf, cmap=pb.cm.jet, vmin=m.min(), vmax=m.max(), linewidth=0.)
-            pb.xlim(xmin[0], xmax[0])
+            ax.set_xlim(xmin[0], xmax[0])
-            pb.ylim(xmin[1], xmax[1])
+            ax.set_ylim(xmin[1], xmax[1])
 
         else:
             raise NotImplementedError, "Cannot define a frame with more than two input dimensions"

GPy/core/sparse_GP.py

@@ -281,17 +281,21 @@ class sparse_GP(GPBase):
 
         return mean, var, _025pm, _975pm
 
-    def plot(self, samples=0, plot_limits=None, which_data='all', which_parts='all', resolution=None, levels=20):
+    def plot(self, samples=0, plot_limits=None, which_data='all', which_parts='all', resolution=None, levels=20, fignum=None, ax=None):
-        GPBase.plot(self, samples=0, plot_limits=None, which_data='all', which_parts='all', resolution=None, levels=20)
+        if ax is None:
+            fig = pb.figure(num=fignum)
+            ax = fig.add_subplot(111)
+
+        GPBase.plot(self, samples=0, plot_limits=None, which_data='all', which_parts='all', resolution=None, levels=20, ax=ax)
         if self.X.shape[1] == 1:
-            Xu = self.X * self._Xstd + self._Xmean  # NOTE self.X are the normalized values now
             if self.has_uncertain_inputs:
-                pb.errorbar(Xu[which_data, 0], self.likelihood.data[which_data, 0],
+                Xu = self.X * self._Xstd + self._Xmean  # NOTE self.X are the normalized values now
+                ax.errorbar(Xu[which_data, 0], self.likelihood.data[which_data, 0],
                             xerr=2 * np.sqrt(self.X_variance[which_data, 0]),
                             ecolor='k', fmt=None, elinewidth=.5, alpha=.5)
             Zu = self.Z * self._Xstd + self._Xmean
-            pb.plot(Zu, Zu * 0 + pb.ylim()[0], 'r|', mew=1.5, markersize=12)
+            ax.plot(Zu, np.zeros_like(Zu) + ax.get_ylim()[0], 'r|', mew=1.5, markersize=12)
-                # pb.errorbar(self.X[:,0], pb.ylim()[0]+np.zeros(self.N), xerr=2*np.sqrt(self.X_variance.flatten()))
 
-        elif self.X.shape[1] == 2:  # FIXME
+        elif self.X.shape[1] == 2:
-            pb.plot(self.Z[:, 0], self.Z[:, 1], 'wo')
+            Zu = self.Z * self._Xstd + self._Xmean
+            ax.plot(Zu[:, 0], Zu[:, 1], 'wo')

GPy/examples/regression.py

@@ -85,10 +85,10 @@ def coregionalisation_toy2(max_nb_eval_optim=100):
 
     k1 = GPy.kern.rbf(1) + GPy.kern.bias(1)
     k2 = GPy.kern.coregionalise(2,1)
-    k = k1.prod_orthogonal(k2)
+    k = k1.prod(k2,tensor=True)
     m = GPy.models.GP_regression(X,Y,kernel=k)
-    m.constrain_fixed('rbf_var',1.)
+    m.constrain_fixed('.*rbf_var',1.)
-    m.constrain_positive('kappa')
+    #m.constrain_positive('.*kappa')
     m.ensure_default_constraints()
     m.optimize('sim',messages=1,max_f_eval=max_nb_eval_optim)
 
@@ -117,10 +117,10 @@ def coregionalisation_toy(max_nb_eval_optim=100):
 
     k1 = GPy.kern.rbf(1)
     k2 = GPy.kern.coregionalise(2,2)
-    k = k1.prod_orthogonal(k2)
+    k = k1.prod(k2,tensor=True)
     m = GPy.models.GP_regression(X,Y,kernel=k)
-    m.constrain_fixed('rbf_var',1.)
+    m.constrain_fixed('.*rbf_var',1.)
-    m.constrain_positive('kappa')
+    #m.constrain_positive('kappa')
     m.ensure_default_constraints()
     m.optimize(max_f_eval=max_nb_eval_optim)
 
@@ -153,12 +153,12 @@ def coregionalisation_sparse(max_nb_eval_optim=100):
 
     k1 = GPy.kern.rbf(1)
     k2 = GPy.kern.coregionalise(2,2)
-    k = k1.prod_orthogonal(k2) + GPy.kern.white(2,0.001)
+    k = k1.prod(k2,tensor=True) + GPy.kern.white(2,0.001)
 
     m = GPy.models.sparse_GP_regression(X,Y,kernel=k,Z=Z)
     m.scale_factor = 10000.
-    m.constrain_fixed('rbf_var',1.)
+    m.constrain_fixed('.*rbf_var',1.)
-    m.constrain_positive('kappa')
+    #m.constrain_positive('kappa')
     m.constrain_fixed('iip')
     m.ensure_default_constraints()
     m.optimize_restarts(5, robust=True, messages=1, max_f_eval=max_nb_eval_optim)
@@ -293,11 +293,11 @@ def sparse_GP_regression_2D(N = 400, M = 50, max_nb_eval_optim=100):
     kernel = rbf + noise
 
     # create simple GP model
-    m = GPy.models.sparse_GP_regression(X,Y,kernel, M = M, max_nb_eval_optim=100)
+    m = GPy.models.sparse_GP_regression(X,Y,kernel, M = M)
 
     # contrain all parameters to be positive (but not inducing inputs)
-    m.constrain_positive('(variance|lengthscale|precision)')
+    m.ensure_default_constraints()
-    m.set('len',2.)
+    m.set('.*len',2.)
 
     m.checkgrad()
 
@@ -314,16 +314,16 @@ def uncertain_inputs_sparse_regression(max_nb_eval_optim=100):
     S = np.ones((20,1))
     X = np.random.uniform(-3.,3.,(20,1))
     Y = np.sin(X)+np.random.randn(20,1)*0.05
-    likelihood = GPy.likelihoods.Gaussian(Y)
+    #likelihood = GPy.likelihoods.Gaussian(Y)
     Z = np.random.uniform(-3.,3.,(7,1))
 
     k = GPy.kern.rbf(1) + GPy.kern.white(1)
 
     # create simple GP model
-    m = GPy.models.sparse_GP(X, likelihood, kernel=k, Z=Z, X_uncertainty=S)
+    m = GPy.models.sparse_GP_regression(X, Y, kernel=k, Z=Z, X_variance=S)
 
     # contrain all parameters to be positive
-    m.constrain_positive('(variance|prec)')
+    m.ensure_default_constraints()
 
     # optimize and plot
     m.optimize('tnc', messages=1, max_f_eval=max_nb_eval_optim)

GPy/kern/kern.py

@@ -46,10 +46,11 @@ class kern(parameterised):
         parameterised.__init__(self)
 
 
-    def plot_ARD(self, ax=None):
+    def plot_ARD(self, fignum=None, ax=None):
         """If an ARD kernel is present, it bar-plots the ARD parameters"""
         if ax is None:
-            ax = pb.gca()
+            fig = pb.figure(fignum)
+            ax = fig.add_subplot(111)
         for p in self.parts:
             if hasattr(p, 'ARD') and p.ARD:
                 ax.set_title('ARD parameters, %s kernel' % p.name)

GPy/models/Bayesian_GPLVM.py

@@ -218,7 +218,7 @@ class Bayesian_GPLVM(sparse_GP, GPLVM):
         return means, covars
 
 
-    def plot_X_1d(self, ax=None, fignum=None, colors=None):
+    def plot_X_1d(self, fignum=None, ax=None, colors=None):
         """
         Plot latent space X in 1D:
 
@@ -230,7 +230,8 @@ class Bayesian_GPLVM(sparse_GP, GPLVM):
             colors of different latent space dimensions Q
         """
         import pylab
-        fig = pylab.figure(num=fignum, figsize=(8, min(12, (2 * self.X.shape[1]))))
+        if ax is None:
+            fig = pylab.figure(num=fignum, figsize=(8, min(12, (2 * self.X.shape[1]))))
         if colors is None:
             colors = pylab.gca()._get_lines.color_cycle
             pylab.clf()
@@ -241,8 +242,10 @@ class Bayesian_GPLVM(sparse_GP, GPLVM):
         for i in range(self.X.shape[1]):
             if ax is None:
                 ax = fig.add_subplot(self.X.shape[1], 1, i + 1)
-            else:
+            elif isinstance(ax, (tuple, list)):
                 ax = ax[i]
+            else:
+                raise ValueError("Need one ax per latent dimnesion Q")
             ax.plot(self.X, c='k', alpha=.3)
             plots.extend(ax.plot(x, self.X.T[i], c=colors.next(), label=r"$\mathbf{{X_{{{}}}}}$".format(i)))
             ax.fill_between(x,

GPy/models/mrd.py

@@ -256,15 +256,17 @@ class MRD(model):
         self.Z = Z
         return Z
 
-    def _handle_plotting(self, fig_num, axes, plotf):
+    def _handle_plotting(self, fignum, axes, plotf):
         if axes is None:
-            fig = pylab.figure(num=fig_num, figsize=(4 * len(self.bgplvms), 3))
+            fig = pylab.figure(num=fignum, figsize=(4 * len(self.bgplvms), 3))
         for i, g in enumerate(self.bgplvms):
             if axes is None:
-                ax = fig.add_subplot(1, len(self.bgplvms), i + 1)
+                axes = fig.add_subplot(1, len(self.bgplvms), i + 1)
+            elif isinstance(axes, (tuple, list)):
+                axes = axes[i]
             else:
-                ax = axes[i]
+                raise ValueError("Need one axes per latent dimension Q")
-            plotf(i, g, ax)
+            plotf(i, g, axes)
         pylab.draw()
         if axes is None:
             fig.tight_layout()
@@ -275,20 +277,20 @@ class MRD(model):
     def plot_X_1d(self):
         return self.gref.plot_X_1d()
 
-    def plot_X(self, fig_num="MRD Predictions", axes=None):
+    def plot_X(self, fignum="MRD Predictions", ax=None):
-        fig = self._handle_plotting(fig_num, axes, lambda i, g, ax: ax.imshow(g.X))
+        fig = self._handle_plotting(fignum, ax, lambda i, g, ax: ax.imshow(g.X))
         return fig
 
-    def plot_predict(self, fig_num="MRD Predictions", axes=None, **kwargs):
+    def plot_predict(self, fignum="MRD Predictions", ax=None, **kwargs):
-        fig = self._handle_plotting(fig_num, axes, lambda i, g, ax: ax.imshow(g.predict(g.X)[0], **kwargs))
+        fig = self._handle_plotting(fignum, ax, lambda i, g, ax: ax.imshow(g.predict(g.X)[0], **kwargs))
         return fig
 
-    def plot_scales(self, fig_num="MRD Scales", axes=None, *args, **kwargs):
+    def plot_scales(self, fignum="MRD Scales", ax=None, *args, **kwargs):
-        fig = self._handle_plotting(fig_num, axes, lambda i, g, ax: g.kern.plot_ARD(ax=ax, *args, **kwargs))
+        fig = self._handle_plotting(fignum, ax, lambda i, g, ax: g.kern.plot_ARD(axes=ax, *args, **kwargs))
         return fig
 
-    def plot_latent(self, fig_num="MRD Latent Spaces", axes=None, *args, **kwargs):
+    def plot_latent(self, fignum="MRD Latent Spaces", ax=None, *args, **kwargs):
-        fig = self._handle_plotting(fig_num, axes, lambda i, g, ax: g.plot_latent(ax=ax, *args, **kwargs))
+        fig = self._handle_plotting(fignum, ax, lambda i, g, ax: g.plot_latent(axes=ax, *args, **kwargs))
         return fig
 
     def _debug_plot(self):
@@ -296,11 +298,11 @@ class MRD(model):
         fig = pylab.figure("MRD DEBUG PLOT", figsize=(4 * len(self.bgplvms), 9))
         fig.clf()
         axes = [fig.add_subplot(3, len(self.bgplvms), i + 1) for i in range(len(self.bgplvms))]
-        self.plot_X(axes=axes)
+        self.plot_X(ax=axes)
         axes = [fig.add_subplot(3, len(self.bgplvms), i + len(self.bgplvms) + 1) for i in range(len(self.bgplvms))]
-        self.plot_latent(axes=axes)
+        self.plot_latent(ax=axes)
         axes = [fig.add_subplot(3, len(self.bgplvms), i + 2 * len(self.bgplvms) + 1) for i in range(len(self.bgplvms))]
-        self.plot_scales(axes=axes)
+        self.plot_scales(ax=axes)
         pylab.draw()
         fig.tight_layout()
 

GPy/models/sparse_GP_regression.py

@@ -28,7 +28,7 @@ class sparse_GP_regression(sparse_GP):
 
     """
 
-    def __init__(self, X, Y, kernel=None, normalize_X=False, normalize_Y=False, Z=None, M=10):
+    def __init__(self, X, Y, kernel=None, normalize_X=False, normalize_Y=False, Z=None, M=10, X_variance=None):
         #kern defaults to rbf (plus white for stability)
         if kernel is None:
             kernel = kern.rbf(X.shape[1]) + kern.white(X.shape[1],1e-3)
@@ -43,5 +43,5 @@ class sparse_GP_regression(sparse_GP):
         #likelihood defaults to Gaussian
         likelihood = likelihoods.Gaussian(Y,normalize=normalize_Y)
 
-        sparse_GP.__init__(self, X, likelihood, kernel, Z=Z, normalize_X=normalize_X)
+        sparse_GP.__init__(self, X, likelihood, kernel, Z=Z, normalize_X=normalize_X, X_variance=X_variance)
         self._set_params(self._get_params())