BUG: Don't raise in DataFrame.corr with pd.NA #33809
Conversation
pandas/core/frame.py
Outdated
| mat = numeric_df.to_numpy() | ||
| if is_object_dtype(mat.dtype): | ||
| # We end up with an object array if pd.NA is present | ||
| mat[isna(mat)] = np.nan |
There was a problem hiding this comment.
This feels like a hack, and it seems it should somehow be possible to cast this whole DataFrame to float using to_numpy but that doesn't seem to work
There was a problem hiding this comment.
I was trying that originally but it turns out DataFrame.to_numpy doesn't have an na_value argument (seems only Series and EAs do)
There was a problem hiding this comment.
We'll want to avoid going through object dtype at all. Can you specify dtype="float"? Or does that break other things?
There was a problem hiding this comment.
That raises an error:
TypeError: float() argument must be a string or a number, not 'NAType'There was a problem hiding this comment.
That matches the behavior of Series.to_numpy(). We need to have the NA value passed through as np.nan I think.
There was a problem hiding this comment.
I think as a workaround astyping to float then calling to_numpy works, then later could use na_value for DataFrame.to_numpy directly?
dsaxton
added
NA - MaskedArrays
jreback
left a comment
jreback
left a comment
There was a problem hiding this comment.
can you also use .convert_types() (e.g. like the OP) in the tests
pandas/core/frame.py
Outdated
| mat = numeric_df.to_numpy() | ||
| if is_object_dtype(mat.dtype): | ||
| # We end up with an object array if pd.NA is present | ||
| mat[isna(mat)] = np.nan |
Hmm, I think we wouldn't want that in the test because the functionality we're testing is corr / cov rather than convert_dtypes (the original bug report wasn't minimal, convert_dtypes was working okay)? |
pandas/core/frame.py
Outdated
| cols = numeric_df.columns | ||
| idx = cols.copy() | ||
| mat = numeric_df.values | ||
| mat = numeric_df.astype(float).to_numpy() |
There was a problem hiding this comment.
can you add copy=False here
|
thanks @dsaxton |
3 participants
black pandasgit diff upstream/master -u -- "*.py" | flake8 --diff