plot pandas data frame but most columns have zeros - matplotlib

Question

I am new to pandas and ipython I just setup everything and currently playing around. I have following data frame:
Field 10 20 30 40 50 60 70 80 90 95
0 A 0 0 0 0 0 0 0 0 1 3
1 B 0 0 0 0 0 0 0 1 4 14
2 C 0 0 0 0 0 0 0 1 2 7
3 D 0 0 0 0 0 0 0 1 5 15
4 u 0 0 0 0 0 0 0 1 5 14
5 K 0 0 0 0 0 0 1 2 7 21
6 S 0 0 0 0 0 0 0 1 3 8
7 E 0 0 0 0 0 0 0 1 3 8
8 F 0 0 0 0 0 0 0 1 6 16
I used a csv file to import this data:
df = pd.read_csv('/mycsvfile.csv',
index_col=False, header=0)
As you can see post of the columns are zero this data frame has large number of rows but there is possibility that in column most of the rows can be zero while one or two remaining with a value like "70".
I wounder how can I get this to nice graph where I can show 70, 80, 95 columns with the emphasis.
I found following tutorial: [http://pandas.pydata.org/pandas-docs/version/0.9.1/visualization.html][1] but still I am unable to get a good figure.

Answer 1

It depends a bit on how you want to handle the zero values, but here is an approach:
df = pd.DataFrame({'a': [0,0,0,0,70,0,0,90,0,0,80,0,0],
'b': [0,0,0,50,0,60,0,90,0,80,0,0,0]})
fig, axs = plt.subplots(1,2,figsize=(10,4))
# plot the original, for comparison
df.plot(ax=axs[0])
for name, col in df.iteritems():
col[col != 0].plot(ax=axs[1], label=name)
axs[1].set_xlim(df.index[0],df.index[-1])
axs[1].set_ylim(bottom=0)
axs[1].legend(loc=0)
You could also go for something with .replace(0,np.nan), but matplotlib doesnt draw lines if there are nan's in between. So you probably end up with looping over the columns anyway (and then using dropna().plot() for example).