Matplotlib scatter plot doesn't take strings on x-axis?

Question

I have data in this format. I am able to get a bar plot out of it but not a scatter plot because the x-axis has strings. I looked into a couple of other posts from where I couldn't gather much info.

  import pandas as pd
    import numpy as np

    df = pd.DataFrame({"id":["ssa", "ssb", "ssc", "xxa", "xxb", "xxc"], "mean":[1.3,1.5,5.2,3.1,2.1,3.2], "sd":[0.9,0.5,0.3,0.1,0.2,0.3]})
    df

I get a bar plot with error bars using the following command:

import matplotlib.pyplot as plt
ax = plt.figure()
ax = df.plot(kind='bar',x='id', y='mean',figsize=[15,6], yerr='sd')
ax.set_xlabel("id")
ax.set_ylabel("mean")
ax = plt.tight_layout()
ax = plt.show()

But I get an error when I try to do a scatter plot of the same df.

ax = plt.figure()
ax = df.plot(kind='scatter',x='id', y='mean',figsize=[15,6], yerr='sd')
ax.set_xlabel("id")
ax.set_ylabel("mean")
ax = plt.tight_layout()
ax = plt.show()

Error traceback:

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
<ipython-input-10-b3ab7237d4f1> in <module>()
      1 ax = plt.figure()
----> 2 ax = df.plot(kind='scatter',x='id', y='mean',figsize=[15,6], yerr='sd', style='.')
      3 ax.set_xlabel("id")
      4 ax.set_ylabel("mean")
      5 ax = plt.tight_layout()

C:\Users\AppData\Local\Continuum\Anaconda2\lib\site-packages\pandas\plotting\_core.pyc in __call__(self, x, y, kind, ax, subplots, sharex, sharey, layout, figsize, use_index, title, grid, legend, style, logx, logy, loglog, xticks, yticks, xlim, ylim, rot, fontsize, colormap, table, yerr, xerr, secondary_y, sort_columns, **kwds)
   2618                           fontsize=fontsize, colormap=colormap, table=table,
   2619                           yerr=yerr, xerr=xerr, secondary_y=secondary_y,
-> 2620                           sort_columns=sort_columns, **kwds)
   2621     __call__.__doc__ = plot_frame.__doc__
   2622 

C:\Users\AppData\Local\Continuum\Anaconda2\lib\site-packages\pandas\plotting\_core.pyc in plot_frame(data, x, y, kind, ax, subplots, sharex, sharey, layout, figsize, use_index, title, grid, legend, style, logx, logy, loglog, xticks, yticks, xlim, ylim, rot, fontsize, colormap, table, yerr, xerr, secondary_y, sort_columns, **kwds)
   1855                  yerr=yerr, xerr=xerr,
   1856                  secondary_y=secondary_y, sort_columns=sort_columns,
-> 1857                  **kwds)
   1858 
   1859 

C:\Users\AppData\Local\Continuum\Anaconda2\lib\site-packages\pandas\plotting\_core.pyc in _plot(data, x, y, subplots, ax, kind, **kwds)
   1680         plot_obj = klass(data, subplots=subplots, ax=ax, kind=kind, **kwds)
   1681 
-> 1682     plot_obj.generate()
   1683     plot_obj.draw()
   1684     return plot_obj.result

C:\Users\AppData\Local\Continuum\Anaconda2\lib\site-packages\pandas\plotting\_core.pyc in generate(self)
    236         self._compute_plot_data()
    237         self._setup_subplots()
--> 238         self._make_plot()
    239         self._add_table()
    240         self._make_legend()

C:\Users\AppData\Local\Continuum\Anaconda2\lib\site-packages\pandas\plotting\_core.pyc in _make_plot(self)
    829         else:
    830             label = None
--> 831         scatter = ax.scatter(data[x].values, data[y].values, c=c_values,
    832                              label=label, cmap=cmap, **self.kwds)
    833         if cb:

C:\Users\AppData\Local\Continuum\Anaconda2\lib\site-packages\pandas\core\frame.pyc in __getitem__(self, key)
   2060             return self._getitem_multilevel(key)
   2061         else:
-> 2062             return self._getitem_column(key)
   2063 
   2064     def _getitem_column(self, key):

C:\Users\AppData\Local\Continuum\Anaconda2\lib\site-packages\pandas\core\frame.pyc in _getitem_column(self, key)
   2067         # get column
   2068         if self.columns.is_unique:
-> 2069             return self._get_item_cache(key)
   2070 
   2071         # duplicate columns & possible reduce dimensionality

C:\Users\AppData\Local\Continuum\Anaconda2\lib\site-packages\pandas\core\generic.pyc in _get_item_cache(self, item)
   1532         res = cache.get(item)
   1533         if res is None:
-> 1534             values = self._data.get(item)
   1535             res = self._box_item_values(item, values)
   1536             cache[item] = res

C:\Users\AppData\Local\Continuum\Anaconda2\lib\site-packages\pandas\core\internals.pyc in get(self, item, fastpath)
   3588 
   3589             if not isnull(item):
-> 3590                 loc = self.items.get_loc(item)
   3591             else:
   3592                 indexer = np.arange(len(self.items))[isnull(self.items)]

C:\Users\AppData\Local\Continuum\Anaconda2\lib\site-packages\pandas\core\indexes\base.pyc in get_loc(self, key, method, tolerance)
   2393                 return self._engine.get_loc(key)
   2394             except KeyError:
-> 2395                 return self._engine.get_loc(self._maybe_cast_indexer(key))
   2396 
   2397         indexer = self.get_indexer([key], method=method, tolerance=tolerance)

pandas\_libs\index.pyx in pandas._libs.index.IndexEngine.get_loc (pandas\_libs\index.c:5239)()

pandas\_libs\index.pyx in pandas._libs.index.IndexEngine.get_loc (pandas\_libs\index.c:5085)()

pandas\_libs\hashtable_class_helper.pxi in pandas._libs.hashtable.PyObjectHashTable.get_item (pandas\_libs\hashtable.c:20405)()

pandas\_libs\hashtable_class_helper.pxi in pandas._libs.hashtable.PyObjectHashTable.get_item (pandas\_libs\hashtable.c:20359)()

KeyError: 'id'

So, instead I used seaborn to plot and it plots perfectly. However, I am not sure how to pass "sd" column to plot error bars in it.

fig, ax = plt.subplots(figsize=(5,3))
ax = sns.pointplot(x="id", y="mean",  data=df, join=False)
ax = plt.xticks(rotation=90)
ax = plt.tight_layout()
ax = plt.show()

fig, ax = plt.subplots(figsize=(25,5))
ax = sns.pointplot(x="id", y="mean", data=df, join=False)
ax.map(plt.errorbar, "id", "mean", "sd", marker="o")
ax = plt.xticks(rotation=90)
ax = plt.tight_layout()
ax = plt.show()

The above code throws the following error:

AttributeError                            Traceback (most recent call last)
<ipython-input-21-18652e3e8b12> in <module>()
      1 fig, ax = plt.subplots(figsize=(25,5))
      2 ax = sns.pointplot(x="id", y="mean", data=df, join=False)
----> 3 ax.map(plt.errorbar, "id", "mean", "sd", marker="o")
      4 ax = plt.xticks(rotation=90)
      5 ax = plt.tight_layout()

AttributeError: 'AxesSubplot' object has no attribute 'map'

What I would ideally like to have is a plot similar to the pointplot but with each point having a different size (as specified by its corresponding sd) or with each point having an error bar (given by sd). Can someone tell me how to do this?

Answer 1

Just use matplotlib.axes.Axes.errorbar() as shown in this example by adding this line:

ax.errorbar(np.arange(len(df['id'])), df['mean'], yerr=df['sd'], ls='None')

But I don't think you have to use seaborn:

import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

df = pd.DataFrame({"id":["ssa", "ssb", "ssc", "xxa", "xxb", "xxc"], 
                   "mean":[1.3,1.5,5.2,3.1,2.1,3.2], 
                   "sd":[0.9,0.5,0.3,0.1,0.2,0.3]})
plt.errorbar(np.arange(len(df['id'])), df['mean'], yerr=df['sd'], ls='None', marker='o')
ax = plt.gca()
ax.xaxis.set_ticks(np.arange(len(df['id'])))
ax.xaxis.set_ticklabels(df['id'], rotation=90)
plt.xlabel("id")
plt.ylabel("mean")

plt.show()

Edit: OP now wants "to remove error bars and draw your point with size proportional to sd".

import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

df = pd.DataFrame({"id":["ssa", "ssb", "ssc", "xxa", "xxb", "xxc"], 
                   "mean":[1.3,1.5,5.2,3.1,2.1,3.2], 
                   "sd":[0.9,0.5,0.3,0.1,0.2,0.3]})
fig, ax = plt.subplots()
size_scaler = 300 # Your points will be too small if you just use sd
ax.scatter(np.arange(len(df['id'])), df['mean'], s=df['sd']*size_scaler, marker='o')
ax.xaxis.set_ticks(np.arange(len(df['id'])))
ax.xaxis.set_ticklabels(df['id'], rotation=90)
plt.xlabel("id")
plt.ylabel("mean")

plt.show()