def test_lazy_functions():
df = pl.DataFrame({
"a": ["foo", "bar", "2"],
"b": [1, 2, 3],
"c": [1.0, 2.0, 3.0]
})
out = df[[pl.count("a")]]
assert out[0] == 3
assert pl.count(df["a"]) == 3
out = df[[
pl.var("b"),
pl.std("b"),
pl.max("b"),
pl.min("b"),
pl.sum("b"),
pl.mean("b"),
pl.median("b"),
pl.n_unique("b"),
pl.first("b"),
pl.last("b"),
]]
expected = 1.0
assert np.isclose(out[0], expected)
assert np.isclose(pl.var(df["b"]), expected)
expected = 1.0
assert np.isclose(out[1], expected)
assert np.isclose(pl.std(df["b"]), expected)
expected = 3
assert np.isclose(out[2], expected)
assert np.isclose(pl.max(df["b"]), expected)
expected = 1
assert np.isclose(out[3], expected)
assert np.isclose(pl.min(df["b"]), expected)
expected = 6
assert np.isclose(out[4], expected)
assert np.isclose(pl.sum(df["b"]), expected)
expected = 2
assert np.isclose(out[5], expected)
assert np.isclose(pl.mean(df["b"]), expected)
expected = 2
assert np.isclose(out[6], expected)
assert np.isclose(pl.median(df["b"]), expected)
expected = 3
assert np.isclose(out[7], expected)
assert np.isclose(pl.n_unique(df["b"]), expected)
expected = 1
assert np.isclose(out[8], expected)
assert np.isclose(pl.first(df["b"]), expected)
expected = 3
assert np.isclose(out[9], expected)
assert np.isclose(pl.last(df["b"]), expected)
expected = 3
assert np.isclose(out[9], expected)
assert np.isclose(pl.last(df["b"]), expected)
def test_lazy_functions():
df = pl.DataFrame({
"a": ["foo", "bar", "2"],
"b": [1, 2, 3],
"c": [1.0, 2.0, 3.0]
})
out = df[[pl.count("a")]]
assert out["a"] == 3
assert pl.count(df["a"]) == 3
out = df[[
pl.var("b"),
pl.std("b"),
pl.max("b"),
pl.min("b"),
pl.sum("b"),
pl.mean("b"),
pl.median("b"),
pl.n_unique("b"),
pl.first("b"),
pl.last("b"),
]]
expected = 1.0
assert np.isclose(out.select_at_idx(0), expected)
assert np.isclose(pl.var(df["b"]), expected)
expected = 1.0
assert np.isclose(out.select_at_idx(1), expected)
assert np.isclose(pl.std(df["b"]), expected)
expected = 3
assert np.isclose(out.select_at_idx(2), expected)
assert np.isclose(pl.max(df["b"]), expected)
expected = 1
assert np.isclose(out.select_at_idx(3), expected)
assert np.isclose(pl.min(df["b"]), expected)
expected = 6
assert np.isclose(out.select_at_idx(4), expected)
assert np.isclose(pl.sum(df["b"]), expected)
expected = 2
assert np.isclose(out.select_at_idx(5), expected)
assert np.isclose(pl.mean(df["b"]), expected)
expected = 2
assert np.isclose(out.select_at_idx(6), expected)
assert np.isclose(pl.median(df["b"]), expected)
expected = 3
assert np.isclose(out.select_at_idx(7), expected)
assert np.isclose(pl.n_unique(df["b"]), expected)
expected = 1
assert np.isclose(out.select_at_idx(8), expected)
assert np.isclose(pl.first(df["b"]), expected)
expected = 3
assert np.isclose(out.select_at_idx(9), expected)
assert np.isclose(pl.last(df["b"]), expected)
expected = 3
assert np.isclose(out.select_at_idx(9), expected)
assert np.isclose(pl.last(df["b"]), expected)
print("q5")
out = x.groupby("id6").agg([pl.sum("v1"),
pl.sum("v2"),
pl.sum("v3")]).collect()
print(time.time() - t0)
print("out.shape", out.shape)
print('out["v1_sum"].sum()', out["v1_sum"].sum())
print('out["v2_sum"].sum()', out["v2_sum"].sum())
easy_time = time.time() - t0easy
t0advanced = time.time()
t0 = time.time()
print("q6")
out = (x.groupby(["id4", "id5"]).agg(
[pl.median("v3").alias("v3_median"),
pl.std("v3").alias("v3_std")]).collect())
print(time.time() - t0)
print("out.shape", out.shape)
print('out["v3_median"].sum()', out["v3_median"].sum())
print('out["v3_std"].sum()', out["v3_std"].sum())
t0 = time.time()
print("q7")
out = (x.groupby("id3").agg([(pl.max("v1") - pl.min("v2")).alias("range_v1_v2")
]).collect())
print(time.time() - t0)
print("out.shape", out.shape)
print('out["range_v1_v2"].sum()', out["range_v1_v2"].sum())
t0 = time.time()
print("q8")
def stdize_out(value: str, control_for: str) -> pl.Expr:
return (pl.col(value) - pl.mean(value).over(control_for)) / pl.std(value).over(
control_for
)
from .dataset import df
import polars as pl
from polars import col
df = df[[
pl.sum("random").alias("sum"),
pl.min("random").alias("min"),
pl.max("random").alias("max"),
col("random").max().alias("other_max"),
pl.std("random").alias("std dev"),
pl.var("random").alias("variance"),
]]