phobson · phobson · Apr 25, 2016 · Apr 20, 2016 · Apr 20, 2016 · Apr 20, 2016
diff --git a/docs/tutorial/activities.ipynb b/docs/tutorial/activities.ipynb
@@ -26,14 +26,15 @@
    "cell_type": "code",
    "execution_count": null,
    "metadata": {
-    "collapsed": true
+    "collapsed": false
    },
    "outputs": [],
    "source": [
     "import warnings\n",
     "warnings.simplefilter('ignore')\n",
     "\n",
     "from numpy.random import seed\n",
+    "from scipy import stats\n",
     "from matplotlib import pyplot\n",
     "import seaborn\n",
     "\n",
@@ -127,7 +128,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Fitting distribution\n",
+    "### Fitting distributions\n",
     "Fitting distributions to data follows a similar pattern."
    ]
   },
@@ -143,6 +144,25 @@
     "paramnormal.activity.fit('beta', data)"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Equivalent command to perform the same fits in raw scipy is shown below:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "# constrained loc and scale\n",
+    "stats.beta.fit(data, floc=0, fscale=1)"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -165,8 +185,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Plotting\n",
-    "There is very limited plotting functionality built into paramnormal. For the time being, we just plot the probability distribution function (PDF), and offload everything else to the seaborn module if it's available."
+    "And again in raw scipy:"
    ]
   },
   {
@@ -177,15 +196,17 @@
    },
    "outputs": [],
    "source": [
-    "ax = paramnormal.activity.plot('beta', α=3, β=2)\n",
-    "ax.figure"
+    "# constrained beta and scale, unconstrained loc\n",
+    "stats.beta.fit(data, f1=2, fscale=1)"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "You can plot on an existing figure through the `ax` argument and control the line style through `line_opts`."
+    "### Plotting\n",
+    "There is very limited plotting functionality built into paramnormal.\n",
+    "The probability distribution function (PDF) is plotted by default, but any other method of the distributions can be plotted by specifying the ``which`` parameters."
    ]
   },
   {
@@ -196,21 +217,16 @@
    },
    "outputs": [],
    "source": [
-    "seaborn.set(style='ticks', rc=clean_bkgd)\n",
-    "fig, (ax1, ax2) = pyplot.subplots(nrows=2, sharex=True, sharey=True)\n",
-    "ax1 = paramnormal.activity.plot('beta', α=6, β=2, ax=ax1, line_opts=dict(color='firebrick', lw=3))\n",
-    "ax2 = paramnormal.activity.plot('beta', α=2, β=6, ax=ax2, line_opts=dict(color='forestgreen', lw=1.25))\n",
-    "ax1.set_ylabel('α=6, β=2')\n",
-    "ax2.set_ylabel('α=2, β=6')\n",
-    "seaborn.despine(fig)\n",
-    "fig"
+    "ax = paramnormal.activity.plot('beta', α=3, β=2)\n",
+    "paramnormal.activity.plot('beta', α=3, β=2, ax=ax, which='CDF')\n",
+    "ax.legend()\n"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Of course, you can create a fully-specified distribtion and omit the distribution parameters."
+    "You can plot on an existing figure through the `ax` argument and control the line style through `line_opts`."
    ]
   },
   {
@@ -221,16 +237,19 @@
    },
    "outputs": [],
    "source": [
-    "beta = paramnormal.beta(α=3, β=2)\n",
-    "ax = paramnormal.activity.plot(beta)\n",
-    "ax.figure"
+    "fig, (ax, ax2) = pyplot.subplots(nrows=2, sharex=True, sharey=True)\n",
+    "paramnormal.activity.plot('beta', α=6, β=2, ax=ax, line_opts=dict(color='firebrick', lw=3))\n",
+    "paramnormal.activity.plot('beta', α=2, β=6, ax=ax2, line_opts=dict(color='forestgreen', lw=1.25))\n",
+    "ax.set_ylabel('α=6, β=2')\n",
+    "ax2.set_ylabel('α=2, β=6')\n",
+    "seaborn.despine(fig)"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Passing an array of sample data will fit the distribution to that data and allow you to use `seaborn.distplot`."
+    "Of course, you can create a fully-specified distribtion and omit the distribution parameters."
    ]
   },
   {
@@ -241,18 +260,15 @@
    },
    "outputs": [],
    "source": [
-    "data = paramnormal.activity.random('beta', α=3, β=2, shape=125)\n",
-    "ax = paramnormal.activity.plot('beta', data, distplot=True)\n",
-    "seaborn.despine()\n",
-    "ax.legend()\n",
-    "ax.figure"
+    "beta = paramnormal.beta(α=3, β=2)\n",
+    "ax = paramnormal.activity.plot(beta)"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "And you can control `seaborn.distplot` through `displot_opts`."
+    "And finally, you can pass an array of data and an unfrozen distribution, and a new distribution will be fit to your data."
    ]
   },
   {
@@ -263,12 +279,11 @@
    },
    "outputs": [],
    "source": [
-    "data = paramnormal.activity.random('normal', μ=0.75, σ=1.25, shape=125)\n",
-    "ax = paramnormal.activity.plot('normal', data, distplot=True,\n",
-    "                               distplot_opts=dict(hist=True, rug=True, kde=False, norm_hist=True))\n",
-    "seaborn.despine()\n",
-    "ax.legend()\n",
-    "ax.figure"
+    "data = paramnormal.activity.random('beta', α=2, β=6, shape=37) + \\\n",
+    "       paramnormal.activity.random('normal', μ=5, σ=1, shape=37)\n",
+    "ax = paramnormal.activity.plot('normal', data=data, line_opts=dict(label='Emperical Fit'))\n",
+    "ax = paramnormal.activity.plot('normal', μ=5, σ=1, line_opts=dict(label='Theoretical'))\n",
+    "ax.legend()"
    ]
   }
  ],
@@ -293,4 +308,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 0
-}
+}
diff --git a/paramnormal/activity.py b/paramnormal/activity.py
@@ -27,10 +27,10 @@ def _check_distro(distro, **params):
 
 def _check_ax(ax):
     if ax is None:
-        fig, ax = pyplot.subplots()
-    else:
         ax = pyplot.gca()
         fig = ax.figure
+    else:
+        fig = ax.figure
 
     return fig, ax
 
@@ -121,22 +121,8 @@ def fit(distro, data, as_params=True, **guesses):
         return distro.from_params(params)
 
 
-def _plot_pdf(distro, xlimits, ax=None, xscale='linear', **line_opts):
-    fig, ax = _check_ax(ax)
-    if xscale == 'log':
-        xlimits = numpy.log10(xlimits)
-        spacer = numpy.logspace
-    else:
-        spacer = numpy.linspace
-
-    x_hat = spacer(xlimits[0], xlimits[1], num=100)
-    y_hat = distro.pdf(x_hat)
-    ax.plot(x_hat, y_hat, **line_opts)
-    return ax
-
-
-def plot(distro, data=None, fit_dist=True, ax=None, pad=0.1,
-         xscale='linear', line_opts=None, **guesses):
+def plot(distro, which='PDF', data=None, fit_dist=True, ax=None,
+         pad=0.05, xscale='linear', line_opts=None, **guesses):
     """
     Plot the PDF of a dataset and other representations of the
     distribution (histogram, kernel density estimate, and rug plot).
@@ -253,18 +239,31 @@ def plot(distro, data=None, fit_dist=True, ax=None, pad=0.1,
 
     """
 
+    # validate the axes and distribution function (`which`)
+    fig, ax = _check_ax(ax)
     if data is not None:
         distro = fit(distro, data, as_params=False, **guesses)
     else:
         distro = _check_distro(distro, **guesses)
+    fxn = getattr(distro, which.lower())
 
-    fig, ax = _check_ax(ax)
+    # determine and set the xlimits of the plot
+    xlimits = distro.ppf([pad/100, 1 - pad/100])
+
+    # determine the x-values
+    if xscale == 'log':
+        #xlimits = numpy.log10(xlimits)
+        x_hat = numpy.logspace(*numpy.log10(xlimits), num=100)
+    else:
+        x_hat = numpy.linspace(*xlimits, num=100)
+
+    # compute y-values
+    y_hat = fxn(x_hat)
 
     line_opts = dict() if line_opts is None else line_opts
-    line_opts['label'] = line_opts.pop('label', 'PDF')
-    xlimits = distro.ppf([pad/100, 1 - pad/100])
+    line_opts['label'] = line_opts.pop('label', which)
 
-    ax = _plot_pdf(distro, xlimits, ax=ax, xscale=xscale, **line_opts)
+    line, = ax.plot(x_hat, y_hat, **line_opts)
     ax.set_xscale(xscale)
 
     return ax
diff --git a/paramnormal/tests/baseline_images/test_activity/test_plot_basic.png b/paramnormal/tests/baseline_images/test_activity/test_plot_basic.png
diff --git a/paramnormal/tests/baseline_images/test_activity/test_plot_cdf_basic.png b/paramnormal/tests/baseline_images/test_activity/test_plot_cdf_basic.png
diff --git a/paramnormal/tests/baseline_images/test_activity/test_plot_cdf_fit.png b/paramnormal/tests/baseline_images/test_activity/test_plot_cdf_fit.png
diff --git a/paramnormal/tests/baseline_images/test_activity/test_plot_cdf_xlog.png b/paramnormal/tests/baseline_images/test_activity/test_plot_cdf_xlog.png
diff --git a/paramnormal/tests/baseline_images/test_activity/test_plot_distplot.png b/paramnormal/tests/baseline_images/test_activity/test_plot_distplot.png
diff --git a/paramnormal/tests/baseline_images/test_activity/test_plot_distplot_lognormal.png b/paramnormal/tests/baseline_images/test_activity/test_plot_distplot_lognormal.png
diff --git a/paramnormal/tests/baseline_images/test_activity/test_plot_fit.png b/paramnormal/tests/baseline_images/test_activity/test_plot_fit.png
diff --git a/paramnormal/tests/baseline_images/test_activity/test_plot_pdf_basic.png b/paramnormal/tests/baseline_images/test_activity/test_plot_pdf_basic.png
diff --git a/paramnormal/tests/baseline_images/test_activity/test_plot_pdf_fit.png b/paramnormal/tests/baseline_images/test_activity/test_plot_pdf_fit.png
diff --git a/paramnormal/tests/baseline_images/test_activity/test_plot_pdf_xlog.png b/paramnormal/tests/baseline_images/test_activity/test_plot_pdf_xlog.png
diff --git a/paramnormal/tests/baseline_images/test_activity/test_plot_sf_basic.png b/paramnormal/tests/baseline_images/test_activity/test_plot_sf_basic.png
diff --git a/paramnormal/tests/baseline_images/test_activity/test_plot_sf_fit.png b/paramnormal/tests/baseline_images/test_activity/test_plot_sf_fit.png
diff --git a/paramnormal/tests/baseline_images/test_activity/test_plot_sf_xlog.png b/paramnormal/tests/baseline_images/test_activity/test_plot_sf_xlog.png
diff --git a/paramnormal/tests/baseline_images/test_activity/test_plot_xlog.png b/paramnormal/tests/baseline_images/test_activity/test_plot_xlog.png
diff --git a/paramnormal/tests/test_activity.py b/paramnormal/tests/test_activity.py
@@ -111,20 +111,20 @@ def test_normal(self):
 
         assert_dists_are_equivalent(dist, stats.norm(params.mu, params.sigma))
 
-@image_comparison(baseline_images=['test_plot_basic'], extensions=['png'])
+@image_comparison(baseline_images=['test_plot_pdf_basic'], extensions=['png'])
 @nptest.dec.skipif(sys.version_info.minor < 4)
 @seed
-def test_plot_basic():
+def test_plot_pdf_basic():
     # first
     fig, ax1 = pyplot.subplots()
     norm_dist = dist.normal(μ=5.4, σ=2.5)
     ax1 = activity.plot(norm_dist, ax=ax1)
 
 
-@image_comparison(baseline_images=['test_plot_fit'], extensions=['png'])
+@image_comparison(baseline_images=['test_plot_pdf_fit'], extensions=['png'])
 @nptest.dec.skipif(sys.version_info.minor < 4)
 @seed
-def test_plot_fit():
+def test_plot_pdf_fit():
     # second
     fig2, ax2 = pyplot.subplots()
     norm_dist = dist.normal(μ=5.4, σ=2.5)
@@ -134,11 +134,85 @@ def test_plot_fit():
     ax2.legend()
 
 
-@image_comparison(baseline_images=['test_plot_xlog'], extensions=['png'])
+@image_comparison(baseline_images=['test_plot_pdf_xlog'], extensions=['png'])
 @nptest.dec.skipif(sys.version_info.minor < 4)
 @seed
-def test_plot_xlog():
+def test_plot_pdf_xlog():
     # first
     fig, ax1 = pyplot.subplots()
     loc_dist = dist.lognormal(μ=1.25, σ=0.75)
     ax1 = activity.plot(loc_dist, ax=ax1, xscale='log')
+
+
+@image_comparison(baseline_images=['test_plot_cdf_basic'], extensions=['png'])
+@nptest.dec.skipif(sys.version_info.minor < 4)
+@seed
+def test_plot_cdf_basic():
+    # first
+    fig, ax1 = pyplot.subplots()
+    norm_dist = dist.normal(μ=5.4, σ=2.5)
+    ax1 = activity.plot(norm_dist, ax=ax1, which='cdf')
+
+
+@image_comparison(baseline_images=['test_plot_cdf_fit'], extensions=['png'])
+@nptest.dec.skipif(sys.version_info.minor < 4)
+@seed
+def test_plot_cdf_fit():
+    # second
+    fig2, ax2 = pyplot.subplots()
+    norm_dist = dist.normal(μ=5.4, σ=2.5)
+    data = activity.random('normal', μ=5.4, σ=2.5, shape=37)
+    ax2 = activity.plot(norm_dist, ax=ax2, line_opts=dict(label='Theoretical CDF'), which='cdf')
+    ax2 = activity.plot('normal', data=data, ax=ax2, line_opts=dict(label='Fit CDF'), which='cdf')
+    ax2.legend()
+
+
+@image_comparison(baseline_images=['test_plot_cdf_xlog'], extensions=['png'])
+@nptest.dec.skipif(sys.version_info.minor < 4)
+@seed
+def test_plot_cdf_xlog():
+    # first
+    fig, ax1 = pyplot.subplots()
+    loc_dist = dist.lognormal(μ=1.25, σ=0.75)
+    ax1 = activity.plot(loc_dist, ax=ax1, xscale='log', which='CDF')
+    ax1.legend()
+
+@image_comparison(baseline_images=['test_plot_sf_basic'], extensions=['png'])
+@nptest.dec.skipif(sys.version_info.minor < 4)
+@seed
+def test_plot_sf_basic():
+    # first
+    fig, ax1 = pyplot.subplots()
+    norm_dist = dist.normal(μ=5.4, σ=2.5)
+    ax1 = activity.plot(norm_dist, ax=ax1, which='sf')
+
+
+@image_comparison(baseline_images=['test_plot_sf_fit'], extensions=['png'])
+@nptest.dec.skipif(sys.version_info.minor < 4)
+@seed
+def test_plot_sf_fit():
+    # second
+    fig2, ax2 = pyplot.subplots()
+    norm_dist = dist.normal(μ=5.4, σ=2.5)
+    data = activity.random('normal', μ=5.4, σ=2.5, shape=37)
+    ax2 = activity.plot(norm_dist, ax=ax2, line_opts=dict(label='Theoretical sf'), which='sf')
+    ax2 = activity.plot('normal', data=data, ax=ax2, line_opts=dict(label='Fit sf'), which='sf')
+    ax2.legend()
+
+
+@image_comparison(baseline_images=['test_plot_sf_xlog'], extensions=['png'])
+@nptest.dec.skipif(sys.version_info.minor < 4)
+@seed
+def test_plot_sf_xlog():
+    # first
+    fig, ax1 = pyplot.subplots()
+    loc_dist = dist.lognormal(μ=1.25, σ=0.75)
+    ax1 = activity.plot(loc_dist, ax=ax1, xscale='log', which='sf')
+    ax1.legend()
+
+
+@cleanup
+@nt.raises(AttributeError)
+def test_plot_bad_attribute():
+    loc_dist = dist.lognormal(μ=1.25, σ=0.75)
+    activity.plot(loc_dist, xscale='log', which='JUNK')