excaliburzero / blendplot Goto Github PK

When attempting to install I get this Error:

error in blendplot setup command: "values of 'package_data' dict" must be a list of strings (got '[LICENSE]')

Able to fix it with this quick change to setup.py:

Change Line 21 from

'': 'LICENSE'

to

'': ['LICENSE']

Not sure if this is the best fix, or something else can solve the problem.

P.S. - this is a really fun, easy to use tool!

When the user supplies a column either for the (x,y,z) axes or for the category column, then they should be validated against the dataframe created from the data file.

If one or more of the columns are invalid, then the program should print an error message to stderr indicating this and then exit with an error status.

Currently the program uses sklearn's preprocessing.scale function to scale the dataset. There should be a flag that allows the user to specify a different scaling function.

Ideally it should allow for several of the scaling functions supplied by sklearn as well as an option for no scaling at all. Note that the --spacing flag can already be used to do some basic custom scaling.