Merge pull request #113 from wichmann-lab/fixplotclean

Fixes two plots

psignifit/psigniplot.py

@@ -21,7 +21,7 @@ def plot_psychometric_function(result: Result,  # noqa: C901, this function is t
                                extrapolate_stimulus: float = 0.2,
                                x_label='Stimulus Level',
                                y_label='Proportion Correct'):
-    """ Plot oted psychometric function with the data.
+    """ Plot psychometric function fit together with the data.
     """
     if ax is None:
         ax = plt.gca()
@@ -74,8 +74,8 @@ def plot_psychometric_function(result: Result,  # noqa: C901, this function is t
 
     # AXIS SETTINGS
     plt.axis('tight')
-    plt.xlabel(x_label)
-    plt.ylabel(y_label)
+    plt.xlabel(x_label, fontsize=14)
+    plt.ylabel(y_label, fontsize=14)
     plt.ylim([ymin, 1])
     return ax
 
@@ -89,7 +89,7 @@ def plot_stimulus_residuals(result: Result, ax: matplotlib.axes.Axes = None) ->
 def plot_block_residuals(result: Result, ax: matplotlib.axes.Axes = None) -> matplotlib.axes.Axes:
     if ax is None:
         ax = plt.gca()
-    return _plot_residuals(range(result.data.shape[0]), 'Block Number', result, ax)
+    return _plot_residuals(result.data[:, 0], 'Block Number', result, ax)
 
 
 def _plot_residuals(x_values: np.ndarray, x_label: str, result: Result, ax: matplotlib.axes.Axes = None):
@@ -114,8 +114,8 @@ def _plot_residuals(x_values: np.ndarray, x_label: str, result: Result, ax: matp
     linefit = np.polyfit(x_values, deviance, 3)
     ax.plot(x, np.polyval(linefit, x), 'k:', clip_on=False)
 
-    ax.xlabel(x_label, fontsize=14)
-    ax.ylabel('Deviance', fontsize=14)
+    ax.set_xlabel(x_label, fontsize=14)
+    ax.set_ylabel('Deviance', fontsize=14)
     return ax
 
 
@@ -167,7 +167,7 @@ def plot_marginal(result: Result,
 
     Args:
         result: should be a result struct from the main psignifit routine
-        dim: The parameter to plot. 1=threshold, 2=width, 3=lambda, 4=gamma, 5=sigma
+        dim: The parameter to plot. 'threshold', 'width', 'lambda', 'gamma', 'eta'
     """
     if ax is None:
         ax = plt.gca()
@@ -188,14 +188,15 @@ def plot_marginal(result: Result,
             ax.fill_between(ci_x, np.zeros_like(ci_x), np.interp(ci_x, x, marginal), color=line_color, alpha=0.5)
 
         param_value = result.parameter_estimate[parameter]
-        ax.plot([param_value] * 2, [0, np.interp(param_value, x, marginal)], color=line_color)
+        ax.plot([param_value] * 2, [0, np.interp(param_value, x, marginal)], color='#000000')
 
     if plot_prior:
         ax.plot(x, result.prior_values[parameter], ls='--', color=prior_color, clip_on=False)
 
     ax.plot(x, marginal, lw=line_width, c=line_color, clip_on=False)
     ax.set_xlabel(x_label)
     ax.set_ylabel(y_label)
+    ax.spines[['top', 'right']].set_visible(False)
 
     return ax
 
@@ -242,7 +243,7 @@ def plot_prior(result: Result,
                          prior_x[-cumprior[cumprior >= .75].size], max(prior_x)]
         plt.subplot(2, 3, i + 1)
         plt.plot(params[param], priors[param], lw=line_width, c=line_color)
-        plt.scatter(x_percentiles, np.interp(x_percentiles, prior_x, priors[param]), ms=marker_size, c=colors)
+        plt.scatter(x_percentiles, np.interp(x_percentiles, prior_x, priors[param]), s=marker_size, c=colors)
         plt.xlabel('Stimulus Level')
         plt.ylabel('Density')
         plt.title(titles[param])
@@ -252,7 +253,7 @@ def plot_prior(result: Result,
             this_sigmoid_params = dict(sigmoid_params)
             this_sigmoid_params[param] = param_value
             plt.plot(sigmoid_x, sigmoid(sigmoid_x, **this_sigmoid_params), line_width=line_width, color=color)
-        plt.plot(data[:, 0], np.zeros(data[:, 0].shape), 'k.', ms=marker_size * .75)
+        plt.plot(data[:, 0], np.zeros(data[:, 0].shape), 'k.', s=marker_size * .75)
         plt.xlabel('Stimulus Level')
         plt.ylabel('Percent Correct')