BUG: Cannot use pd.NA in a series with float dtype

[thehomebrewnerd]

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pandas.

• (optional) I have confirmed this bug exists on the master branch of pandas.

---

Note: Please read this guide detailing how to provide the necessary information for us to reproduce your bug.

Code Sample, a copy-pastable example

import pandas as pd
pd.Series([1.1, pd.NA], dtype='float64')

Problem description

Similar to the issue documented in #36586, if the goal of pd.NA is to provide a missing indicator that can be used across datatypes, it should also work with a float datatype. Currently, if a user attempts to use a pd.NA value in a series with dtype of float, an error is raised: TypeError: float() argument must be a string or a number, not 'NAType'.

This also appears related to the issue mentioned in #36585

Expected Output

import pandas as pd
pd.Series([1.1, pd.NA], dtype='float64')

0    1.1
1    <NA>
dtype: float64

Output of pd.show_versions()

INSTALLED VERSIONS

commit : f2ca0a2
python : 3.7.4.final.0
python-bits : 64
OS : Darwin
OS-release : 18.7.0
Version : Darwin Kernel Version 18.7.0: Mon Feb 10 21:08:45 PST 2020; root:xnu-4903.278.28~1/RELEASE_X86_64
machine : x86_64
processor : i386
byteorder : little
LC_ALL : None
LANG : en_US.UTF-8
LOCALE : en_US.UTF-8

pandas : 1.1.1
numpy : 1.19.1
pytz : 2020.1
dateutil : 2.8.1
pip : 20.2.2
setuptools : 40.8.0
Cython : None
pytest : 6.0.1
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : None
IPython : None
pandas_datareader: None
bs4 : None
bottleneck : None
fsspec : None
fastparquet : None
gcsfs : None
matplotlib : None
numexpr : None
odfpy : None
openpyxl : None
pandas_gbq : None
pyarrow : None
pytables : None
pyxlsb : None
s3fs : None
scipy : None
sqlalchemy : None
tables : None
tabulate : None
xarray : None
xlrd : None
xlwt : None
numba : None

[dsaxton]

Nullable float dtypes don't exist yet but are being worked on, see #34307.

[TomAugspurger]

Closing this since it's a duplicate of #32265 (comment).

FYI @thehomebrewnerd, your "expected output" section isn't possible. pd.NA is not a float, so it can't be stored in a NumPy float64 array. That should raise.

[thehomebrewnerd]

@TomAugspurger I guess I didn't search deep enough to see this was being worked in other issues. In any case, I understand your point. As long as there is a float dtype of some sort that can handle pd.NA, analogous to Int64, that seems fine to me.