[Python] rewrite function from Pandas to Polars

Please help rewrite to polars:

df = pd.DataFrame({
"fructs": ["apple", "apple", "apple", "banana", "banana", "banana", "banana"],
"g": [1, 1, 1, 2, 2, 2, 5],
"a": [2, 4, 5, 190, 1, 4, 1],
"b": [1, 3, 2, 1, 43, 3, 1]
})

def _calculate_coefficient(group: pd.DataFrame) -> float:
model = smf.ols("a ~ g + b", data=group).fit() # this line might be in pandas
coef: float = model.params["g"]
return coef


markdown_coefficients = (
df.groupby(["fructs"], as_index=False)
.apply(_calculate_coefficient)
.rename(columns={None: "elast_discount_coef"})
)


Expected result:

result = pd.DataFrame({
"fructs": ["apple", "banana"],
"elast_discount_coef": [0.833333, -33.570071],
})


I need to calculate coefficient for each group

Continue reading...