Question

I have a basic (almost naive) question for plotting on top of a pandas df. Given the df below I am trying to do a stack bar plot for ‘stats_value’ and ‘read1_length’ v/s ‘lib_name’.

    temp1=

               parent_library_name lib_name stats_value  read1_length
    58                  None     CXYY         106           150
    311                 CXYY     CSGW         128           150
    432                 CXYY     CSNS         109           150
    552                 CXYY     CXPS         125           150
    671                 CXYY     CXOA         123           150
    1113                CXYY     CXOC         108           150
    1394                CXYY     CXOO         129           150
    1675                CXYY     CXOP         101           150
    1794                CXYY     CXSP         132           150
    1914                CXYY     CXOY         116           150
    2356                CXYY     CXSO          69           150
    2635                CXYY     CSHT          77           150
    2914                CXYY     CXSU          76           150

Tried the following things:

c=temp1.set_index('lib_name')
c[['stats_value','read1_length']].plot(kind='bar',stacked=True)

Error:
TypeError: unsupported operand type(s) for +: 'numpy.float64' and 'str'

Tried something simple just to test:

c=temp1.set_index('lib_name')
c[['stats_value']].plot()

Error:
AttributeError: 'numpy.ndarray' object has no attribute 'find'

So I think I am missing some trick here.

Best,
-Abhi

Answer 1

dtype of the stats_value was not correct …plotting works again..

-A