Problems when calling Python from Stata 17

Question

The first code below runs without problems in python, but when using the same lines in a Stata do-file, a message appears suggesting a Python coding error. Any idea what could cause this problem? (I am using Stata 17 on Windows.)

# PYTHON CODE: 
import pandas as pd
df = pd.read_stata('data.dta')  
cut_bins = [-5, -4, -3, -2, -1, 0, 1, 2, 3]
df['cut_bins'] = pd.cut(df['v'], bins=cut_bins)

* STATA CODE: 
python 
import pandas as pd
from sfi import Data
cut_bins = [-5, -4, -3, -2, -1, 0, 1, 2, 3]
df = Data.get('id bin v')
df["cut_bins"] = pd.cut(df['v'], bins=cut_bins)
end 

When running the Stata code in a do-file, the following message appears:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: list indices must be integers or slices, not str

This suggests a Python coding error but since the identical lines run in Python without error, it may as well be that there is a problem with Stata.

Answer 1

df = Data.get('id bin v') will return the variables id, bin, and v in a list of lists. ie df[0] hold the first row of values of id, bin, and v.

Try this

* STATA CODE: 
python 
import pandas as pd
from sfi import Data
cut_bins = range(-5,4)
df = pd.DataFrame(Data.getAsDict('id bin v'))
df["cut_bins"] = pd.cut(df['v'], bins=cut_bins)
end

Now, df.head() would produce (using my fake data for id,bin and v):

    id  bin         v  cut_bins
0  1.0  1.0  2.535117    (2, 3]
1  2.0  2.0  0.161955    (0, 1]
2  3.0  3.0 -0.397220   (-1, 0]
3  4.0  4.0  0.003145    (0, 1]
4  5.0  5.0 -2.985772  (-3, -2]