Subsetting

[versipellis]

It's not immediately apparent to me how to subset in Pandas.jl compared to Python's pandas:

Where in Python pandas:

dftop30 = df[df['ORIGIN'].isin(top30)]

I roughly expect the Pandas.jl equivalent to be:

datatop30 = data[isin(data[:ORIGIN],top30),:]

This, however, results in:

PyError ($(Expr(:escape, :(ccall(#= /home/bversipellis/.julia/packages/PyCall/0jMpb/src/pyfncall.jl:44 =# @pysym(:PyObject_Call), PyPtr, (PyPtr, PyPtr, PyPtr), o, pyargsptr, kw))))) <class 'TypeError'>
TypeError("'Series' objects are mutable, thus they cannot be hashed",)

[malmaud]

Can you post complete code that reproduces the problem? Then I can take a look.

[versipellis]

Sure.

using Pandas
df = DataFrame(Dict(:Group=>["Alpha", "Beta", "Alpha", "Charlie", "Beta", "Delta"], :name=>["James", "Jill", "Jake", "A name", "Yet another name", "Some person"]))
subset = ["Alpha","Delta"]

testsubset = df[isin(df[:Group],subset),:]

All I'm trying to do is subset a dataframe based on another object, which is trivial in Python pandas. The above code in Python for the last line:

testsubset = df[df['Group'].isin(subset)]

[malmaud]

You should index in Julia just like in Python:

df[isin(...)] without the extra , :] at the end. If you try doing that in Python, you get the same error message.

[versipellis]

Huh, weird, I had tried that before and run into an error. Looks like I'm doing it right now, sorry and thanks!

[malmaud]

No problem and thanks for checking out the package.

…

On Wed, Dec 12, 2018 at 11:10 AM versipellis ***@***.***> wrote: Huh, weird, I had tried that before and run into an error. Looks like I'm doing it right now, sorry and thanks! — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub <#45 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA8SvUO_J5MRt5QAoD0gWYsPXSrNjTd5ks5u4SqQgaJpZM4ZOF0W> .