Remove FutureWarnings due to pandas 2.0

[patrickleonardy]

Story Title

Fixing FutureWarnings (#156)

Changes made

• Changes the syntax of assigning a new column from df.loc[:,"col_name"] = something to df["col_name"] = something
• Replaces the depreciated inplace argument for some methods

How does the solution address the problem

This PR will update the syntax

Comment

No test were written to automatically catch FutureWarnings. FutureWarnings are displayed by default so they are already shown when running the tests.

Linked issues

Resolves #156
Linked to #160

[pietrodantuono]

The black formatting of tests/preprocessing/test_target_encoder.py is out of the scope of this pull request. I suggest to undo this formatting before going further with the PR.

Also, sometimes, specially in testing or when defining 2D arrays, you may want to avoid black formatting to compromise readability and conciseness, e.g.

# yapf (Google-like option)
df = pd.DataFrame({'variable': ['positive', 'positive', 'negative',
                                'neutral', 'negative', 'positive',
                                'negative', 'neutral', 'neutral',
                                'neutral', 'positive'],
                   'target': [5, 4, -5, 0, -4, 5, -5, 0, 1, 0, 4]})

VS

# black
df = pd.DataFrame(
    {
        "variable": [
            "positive",
            "positive",
            "negative",
            "neutral",
            "negative",
            "positive",
            "negative",
            "neutral",
            "neutral",
            "neutral",
            "positive",
        ],
        "target": [5, 4, -5, 0, -4, 5, -5, 0, 1, 0, 4],
    }
)

To avoid for specific code chunks being automatically formatted by black, you can set:

# fmt: off
df = pd.DataFrame({'variable': ['positive', 'positive', 'negative',
                                'neutral', 'negative', 'positive',
                                'negative', 'neutral', 'neutral',
                                'neutral', 'positive'],
                   'target': [5, 4, -5, 0, -4, 5, -5, 0, 1, 0, 4]})
# fmt: on

Also, I see that there are more than one df.append commands in the tests. All of these will raise a FutureWarning, and if your intention is to get rid of them, these should be changed to pd.concat.

[patrickleonardy]

@pietrodantuono
I changed back the formatting in the test file and also adapted code that outputted the FutureWarnings append and .loc[:, "col_name"] =).