Python pandas plotting multiple graphs on one plot

Question

I have a dataset that looks like this.

Date Name High Value
2017-12-31 Bitcoin 14377.40 18723.76 2017-12-30 Bitcoin 14681.90 18766.88 2017-12-29 Bitcoin 15279.00 18755.70 2017-12-28 Bitcoin 15888.40 18820.54 ... ... ... ... 2017-01-08 CannaCoin 0.01 0.02 2017-01-07 CannaCoin 0.01 0.02 2017-01-06 CannaCoin 0.01 0.02 2017-01-05 CannaCoin 0.02 0.01

Date is the index column and is in the datetime format. My dataset is big and has more than one item in the Namecolumn. The dates range from the beginning of the year till the end. Also, not all items are of the same length. Most should finish at the end of the year but they don't necessarily start in the beginning of it, they can start later.

What I would like to do is, group by Namevalues, and for each create a separate line on the same graph/plot. Value should be on the y axis.

Because I am used to R, what I did was: df.groupby("Name")["Value"].plot()

What I got is a warning:

UserWarning: Attempting to set identical left==right results in singular transformations; automatically expanding. left=17531.0, right=17531.0 'left=%s, right=%s') % (left, right))

Also, the plot looked like this:

As observable, half of the values are missing, as they are outside the plotting area, dates are in a descending instead of an ascending order and half of the plot is empty.

How can I fix this, so that the entire plot will be visible, with the dates in a correct order?

Answer 1

you can pivot on the "Name" and then pass the dataframe to the plot.

pd.pivot_table(data=df, values="Value",columns="Name",index="Date").plot()

Example:

In[]:
idx = ["2017-12-28","2017-12-29","2017-12-30","2017-12-31"] * 2
name = ['Bitcoin'] * 4 + ['CannaCoin'] * 4
vals = np.random.rand(8) * 1000
df = pd.DataFrame({"Date":idx, "Name":name, "Value":vals})
print(df)

Out[]:
         Date       Name       Value
0  2017-12-28    Bitcoin  788.547631
1  2017-12-29    Bitcoin  572.695484
2  2017-12-30    Bitcoin  661.859195
3  2017-12-31    Bitcoin  205.473883
4  2017-12-28  CannaCoin  270.291858
5  2017-12-29  CannaCoin  683.827404
6  2017-12-30  CannaCoin  447.808772
7  2017-12-31  CannaCoin  616.927833

In[]:
pd.pivot_table(data=df, values="Value",columns="Name",index="Date").plot()