Class: Polars::Expr

Inherits:
Object
  • Object
show all
Defined in:
lib/polars/expr.rb

Overview

Expressions that can be used in various contexts.

Instance Method Summary collapse

Instance Method Details

#! ⇒ Expr

Performs boolean not.

Returns:



137
138
139
 
# File 'lib/polars/expr.rb', line 137

def !
  is_not
end

#!=(other) ⇒ Expr

Not equal.

Returns:



116
117
118
 
# File 'lib/polars/expr.rb', line 116

def !=(other)
  wrap_expr(_rbexpr.neq(_to_expr(other)._rbexpr))
end

#%(other) ⇒ Expr

Returns the modulo.

Returns:



81
82
83
 
# File 'lib/polars/expr.rb', line 81

def %(other)
  wrap_expr(_rbexpr % _to_rbexpr(other))
end

#&(other) ⇒ Expr

Bitwise AND.

Returns:



32
33
34
 
# File 'lib/polars/expr.rb', line 32

def &(other)
  wrap_expr(_rbexpr._and(_to_rbexpr(other)))
end

#*(other) ⇒ Expr

Performs multiplication.

Returns:



60
61
62
 
# File 'lib/polars/expr.rb', line 60

def *(other)
  wrap_expr(_rbexpr * _to_rbexpr(other))
end

#**(power) ⇒ Expr

Raises to the power of exponent.

Returns:



88
89
90
 
# File 'lib/polars/expr.rb', line 88

def **(power)
  pow(power)
end

#+(other) ⇒ Expr

Performs addition.

Returns:



46
47
48
 
# File 'lib/polars/expr.rb', line 46

def +(other)
  wrap_expr(_rbexpr + _to_rbexpr(other))
end

#-(other) ⇒ Expr

Performs subtraction.

Returns:



53
54
55
 
# File 'lib/polars/expr.rb', line 53

def -(other)
  wrap_expr(_rbexpr - _to_rbexpr(other))
end

#-@ ⇒ Expr

Performs negation.

Returns:



144
145
146
 
# File 'lib/polars/expr.rb', line 144

def -@
  Utils.lit(0) - self
end

#/(other) ⇒ Expr

Performs division.

Returns:



67
68
69
 
# File 'lib/polars/expr.rb', line 67

def /(other)
  wrap_expr(_rbexpr / _to_rbexpr(other))
end

#<(other) ⇒ Expr

Less than.

Returns:



123
124
125
 
# File 'lib/polars/expr.rb', line 123

def <(other)
  wrap_expr(_rbexpr.lt(_to_expr(other)._rbexpr))
end

#<=(other) ⇒ Expr

Less than or equal.

Returns:



102
103
104
 
# File 'lib/polars/expr.rb', line 102

def <=(other)
  wrap_expr(_rbexpr.lt_eq(_to_expr(other)._rbexpr))
end

#==(other) ⇒ Expr

Equal.

Returns:



109
110
111
 
# File 'lib/polars/expr.rb', line 109

def ==(other)
  wrap_expr(_rbexpr.eq(_to_expr(other)._rbexpr))
end

#>(other) ⇒ Expr

Greater than.

Returns:



130
131
132
 
# File 'lib/polars/expr.rb', line 130

def >(other)
  wrap_expr(_rbexpr.gt(_to_expr(other)._rbexpr))
end

#>=(other) ⇒ Expr

Greater than or equal.

Returns:



95
96
97
 
# File 'lib/polars/expr.rb', line 95

def >=(other)
  wrap_expr(_rbexpr.gt_eq(_to_expr(other)._rbexpr))
end

#^(other) ⇒ Expr

Bitwise XOR.

Returns:



25
26
27
 
# File 'lib/polars/expr.rb', line 25

def ^(other)
  wrap_expr(_rbexpr._xor(_to_rbexpr(other)))
end

#_hash(seed = 0, seed_1 = nil, seed_2 = nil, seed_3 = nil) ⇒ Expr

Hash the elements in the selection.

The hash value is of type :u64.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [1, 2, nil],
    "b" => ["x", nil, "z"]
  }
)
df.with_column(Polars.all._hash(10, 20, 30, 40))
# =>
# shape: (3, 2)
# ┌──────────────────────┬──────────────────────┐
# │ a                    ┆ b                    │
# │ ---                  ┆ ---                  │
# │ u64                  ┆ u64                  │
# ╞══════════════════════╪══════════════════════╡
# │ 4629889412789719550  ┆ 6959506404929392568  │
# │ 16386608652769605760 ┆ 11638928888656214026 │
# │ 11638928888656214026 ┆ 11040941213715918520 │
# └──────────────────────┴──────────────────────┘

Parameters:

  • seed (Integer) (defaults to: 0) —

    Random seed parameter. Defaults to 0.

  • seed_1 (Integer) (defaults to: nil) —

    Random seed parameter. Defaults to seed if not set.

  • seed_2 (Integer) (defaults to: nil) —

    Random seed parameter. Defaults to seed if not set.

  • seed_3 (Integer) (defaults to: nil) —

    Random seed parameter. Defaults to seed if not set.

Returns:



3236
3237
3238
3239
3240
3241
3242
 
# File 'lib/polars/expr.rb', line 3236

def _hash(seed = 0, seed_1 = nil, seed_2 = nil, seed_3 = nil)
  k0 = seed
  k1 = seed_1.nil? ? seed : seed_1
  k2 = seed_2.nil? ? seed : seed_2
  k3 = seed_3.nil? ? seed : seed_3
  wrap_expr(_rbexpr._hash(k0, k1, k2, k3))
end

#abs ⇒ Expr

Compute absolute values.

Examples:

df = Polars::DataFrame.new(
  {
    "A" => [-1.0, 0.0, 1.0, 2.0]
  }
)
df.select(Polars.col("A").abs)
# =>
# shape: (4, 1)
# ┌─────┐
# │ A   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 1.0 │
# │ 0.0 │
# │ 1.0 │
# │ 2.0 │
# └─────┘

Returns:



4156
4157
4158
 
# File 'lib/polars/expr.rb', line 4156

def abs
  wrap_expr(_rbexpr.abs)
end

#agg_groups ⇒ Expr

Get the group indexes of the group by operation.

Should be used in aggregation context only.

Examples:

df = Polars::DataFrame.new(
  {
    "group" => [
      "one",
      "one",
      "one",
      "two",
      "two",
      "two"
    ],
    "value" => [94, 95, 96, 97, 97, 99]
  }
)
df.group_by("group", maintain_order: true).agg(Polars.col("value").agg_groups)
# =>
# shape: (2, 2)
# ┌───────┬───────────┐
# │ group ┆ value     │
# │ ---   ┆ ---       │
# │ str   ┆ list[u32] │
# ╞═══════╪═══════════╡
# │ one   ┆ [0, 1, 2] │
# │ two   ┆ [3, 4, 5] │
# └───────┴───────────┘

Returns:



703
704
705
 
# File 'lib/polars/expr.rb', line 703

def agg_groups
  wrap_expr(_rbexpr.agg_groups)
end

#alias(name) ⇒ Expr

Rename the output of an expression.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [1, 2, 3],
    "b" => ["a", "b", nil]
  }
)
df.select(
  [
    Polars.col("a").alias("bar"),
    Polars.col("b").alias("foo")
  ]
)
# =>
# shape: (3, 2)
# ┌─────┬──────┐
# │ bar ┆ foo  │
# │ --- ┆ ---  │
# │ i64 ┆ str  │
# ╞═════╪══════╡
# │ 1   ┆ a    │
# │ 2   ┆ b    │
# │ 3   ┆ null │
# └─────┴──────┘

Parameters:

Returns:



327
328
329
 
# File 'lib/polars/expr.rb', line 327

def alias(name)
  wrap_expr(_rbexpr._alias(name))
end

#all(drop_nulls: true) ⇒ Boolean

Check if all boolean values in a Boolean column are true.

This method is an expression - not to be confused with Polars.all which is a function to select all columns.

Examples:

df = Polars::DataFrame.new(
  {"TT" => [true, true], "TF" => [true, false], "FF" => [false, false]}
)
df.select(Polars.col("*").all)
# =>
# shape: (1, 3)
# ┌──────┬───────┬───────┐
# │ TT   ┆ TF    ┆ FF    │
# │ ---  ┆ ---   ┆ ---   │
# │ bool ┆ bool  ┆ bool  │
# ╞══════╪═══════╪═══════╡
# │ true ┆ false ┆ false │
# └──────┴───────┴───────┘

Returns:



226
227
228
 
# File 'lib/polars/expr.rb', line 226

def all(drop_nulls: true)
  wrap_expr(_rbexpr.all(drop_nulls))
end

#any(drop_nulls: true) ⇒ Boolean

Check if any boolean value in a Boolean column is true.

Examples:

df = Polars::DataFrame.new({"TF" => [true, false], "FF" => [false, false]})
df.select(Polars.all.any)
# =>
# shape: (1, 2)
# ┌──────┬───────┐
# │ TF   ┆ FF    │
# │ ---  ┆ ---   │
# │ bool ┆ bool  │
# ╞══════╪═══════╡
# │ true ┆ false │
# └──────┴───────┘

Returns:



201
202
203
 
# File 'lib/polars/expr.rb', line 201

def any(drop_nulls: true)
  wrap_expr(_rbexpr.any(drop_nulls))
end

#append(other, upcast: true) ⇒ Expr

Append expressions.

This is done by adding the chunks of other to this Series.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [8, 9, 10],
    "b" => [nil, 4, 4]
  }
)
df.select(Polars.all.head(1).append(Polars.all.tail(1)))
# =>
# shape: (2, 2)
# ┌─────┬──────┐
# │ a   ┆ b    │
# │ --- ┆ ---  │
# │ i64 ┆ i64  │
# ╞═════╪══════╡
# │ 8   ┆ null │
# │ 10  ┆ 4    │
# └─────┴──────┘

Parameters:

  • other (Expr) —

    Expression to append.

  • upcast (Boolean) (defaults to: true) —

    Cast both Series to the same supertype.

Returns:



817
818
819
820
 
# File 'lib/polars/expr.rb', line 817

def append(other, upcast: true)
  other = Utils.expr_to_lit_or_expr(other)
  wrap_expr(_rbexpr.append(other._rbexpr, upcast))
end

#approx_unique ⇒ Expr

Approx count unique values.

This is done using the HyperLogLog++ algorithm for cardinality estimation.

Examples:

df = Polars::DataFrame.new({"a" => [1, 1, 2]})
df.select(Polars.col("a").approx_unique)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ u32 │
# ╞═════╡
# │ 2   │
# └─────┘

Returns:



2086
2087
2088
 
# File 'lib/polars/expr.rb', line 2086

def approx_unique
  wrap_expr(_rbexpr.approx_n_unique)
end

#arccos ⇒ Expr

Compute the element-wise value for the inverse cosine.

Examples:

df = Polars::DataFrame.new({"a" => [0.0]})
df.select(Polars.col("a").arccos)
# =>
# shape: (1, 1)
# ┌──────────┐
# │ a        │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ 1.570796 │
# └──────────┘

Returns:



4649
4650
4651
 
# File 'lib/polars/expr.rb', line 4649

def arccos
  wrap_expr(_rbexpr.arccos)
end

#arccosh ⇒ Expr

Compute the element-wise value for the inverse hyperbolic cosine.

Examples:

df = Polars::DataFrame.new({"a" => [1.0]})
df.select(Polars.col("a").arccosh)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 0.0 │
# └─────┘

Returns:



4769
4770
4771
 
# File 'lib/polars/expr.rb', line 4769

def arccosh
  wrap_expr(_rbexpr.arccosh)
end

#arcsin ⇒ Expr

Compute the element-wise value for the inverse sine.

Examples:

df = Polars::DataFrame.new({"a" => [1.0]})
df.select(Polars.col("a").arcsin)
# =>
# shape: (1, 1)
# ┌──────────┐
# │ a        │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ 1.570796 │
# └──────────┘

Returns:



4629
4630
4631
 
# File 'lib/polars/expr.rb', line 4629

def arcsin
  wrap_expr(_rbexpr.arcsin)
end

#arcsinh ⇒ Expr

Compute the element-wise value for the inverse hyperbolic sine.

Examples:

df = Polars::DataFrame.new({"a" => [1.0]})
df.select(Polars.col("a").arcsinh)
# =>
# shape: (1, 1)
# ┌──────────┐
# │ a        │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ 0.881374 │
# └──────────┘

Returns:



4749
4750
4751
 
# File 'lib/polars/expr.rb', line 4749

def arcsinh
  wrap_expr(_rbexpr.arcsinh)
end

#arctan ⇒ Expr

Compute the element-wise value for the inverse tangent.

Examples:

df = Polars::DataFrame.new({"a" => [1.0]})
df.select(Polars.col("a").arctan)
# =>
# shape: (1, 1)
# ┌──────────┐
# │ a        │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ 0.785398 │
# └──────────┘

Returns:



4669
4670
4671
 
# File 'lib/polars/expr.rb', line 4669

def arctan
  wrap_expr(_rbexpr.arctan)
end

#arctanh ⇒ Expr

Compute the element-wise value for the inverse hyperbolic tangent.

Examples:

df = Polars::DataFrame.new({"a" => [1.0]})
df.select(Polars.col("a").arctanh)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ inf │
# └─────┘

Returns:



4789
4790
4791
 
# File 'lib/polars/expr.rb', line 4789

def arctanh
  wrap_expr(_rbexpr.arctanh)
end

#arg_max ⇒ Expr

Get the index of the maximal value.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [20, 10, 30]
  }
)
df.select(Polars.col("a").arg_max)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ u32 │
# ╞═════╡
# │ 2   │
# └─────┘

Returns:



1447
1448
1449
 
# File 'lib/polars/expr.rb', line 1447

def arg_max
  wrap_expr(_rbexpr.arg_max)
end

#arg_min ⇒ Expr

Get the index of the minimal value.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [20, 10, 30]
  }
)
df.select(Polars.col("a").arg_min)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ u32 │
# ╞═════╡
# │ 1   │
# └─────┘

Returns:



1471
1472
1473
 
# File 'lib/polars/expr.rb', line 1471

def arg_min
  wrap_expr(_rbexpr.arg_min)
end

#arg_sort(reverse: false, nulls_last: false) ⇒ Expr

Get the index values that would sort this column.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [20, 10, 30]
  }
)
df.select(Polars.col("a").arg_sort)
# =>
# shape: (3, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ u32 │
# ╞═════╡
# │ 1   │
# │ 0   │
# │ 2   │
# └─────┘

Parameters:

  • reverse (Boolean) (defaults to: false) —

    Sort in reverse (descending) order.

  • nulls_last (Boolean) (defaults to: false) —

    Place null values last instead of first.

Returns:



1423
1424
1425
 
# File 'lib/polars/expr.rb', line 1423

def arg_sort(reverse: false, nulls_last: false)
  wrap_expr(_rbexpr.arg_sort(reverse, nulls_last))
end

#arg_unique ⇒ Expr

Get index of first unique value.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [8, 9, 10],
    "b" => [nil, 4, 4]
  }
)
df.select(Polars.col("a").arg_unique)
# =>
# shape: (3, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ u32 │
# ╞═════╡
# │ 0   │
# │ 1   │
# │ 2   │
# └─────┘

df.select(Polars.col("b").arg_unique)
# =>
# shape: (2, 1)
# ┌─────┐
# │ b   │
# │ --- │
# │ u32 │
# ╞═════╡
# │ 0   │
# │ 1   │
# └─────┘

Returns:



2151
2152
2153
 
# File 'lib/polars/expr.rb', line 2151

def arg_unique
  wrap_expr(_rbexpr.arg_unique)
end

#argsort(reverse: false, nulls_last: false) ⇒ expr

Get the index values that would sort this column.

Alias for #arg_sort.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [20, 10, 30]
  }
)
df.select(Polars.col("a").argsort)
# =>
# shape: (3, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ u32 │
# ╞═════╡
# │ 1   │
# │ 0   │
# │ 2   │
# └─────┘

Parameters:

  • reverse (Boolean) (defaults to: false) —

    Sort in reverse (descending) order.

  • nulls_last (Boolean) (defaults to: false) —

    Place null values last instead of first.

Returns:

  • (expr) 


4189
4190
4191
 
# File 'lib/polars/expr.rb', line 4189

def argsort(reverse: false, nulls_last: false)
  arg_sort(reverse: reverse, nulls_last: nulls_last)
end

#arr ⇒ ArrayExpr

Create an object namespace of all array related methods.

Returns:



5309
5310
5311
 
# File 'lib/polars/expr.rb', line 5309

def arr
  ArrayExpr.new(self)
end

#backward_fill(limit: nil) ⇒ Expr

Fill missing values with the next to be seen values.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [1, 2, nil],
    "b" => [4, nil, 6]
  }
)
df.select(Polars.all.backward_fill)
# =>
# shape: (3, 2)
# ┌──────┬─────┐
# │ a    ┆ b   │
# │ ---  ┆ --- │
# │ i64  ┆ i64 │
# ╞══════╪═════╡
# │ 1    ┆ 4   │
# │ 2    ┆ 6   │
# │ null ┆ 6   │
# └──────┴─────┘

Parameters:

  • limit (Integer) (defaults to: nil) —

    The number of consecutive null values to backward fill.

Returns:



1827
1828
1829
 
# File 'lib/polars/expr.rb', line 1827

def backward_fill(limit: nil)
  wrap_expr(_rbexpr.backward_fill(limit))
end

#bin ⇒ BinaryExpr

Create an object namespace of all binary related methods.

Returns:



5316
5317
5318
 
# File 'lib/polars/expr.rb', line 5316

def bin
  BinaryExpr.new(self)
end

#bottom_k(k: 5) ⇒ Expr

Return the k smallest elements.

If 'reverse: true` the smallest elements will be given.

Examples:

df = Polars::DataFrame.new(
  {
    "value" => [1, 98, 2, 3, 99, 4]
  }
)
df.select(
  [
    Polars.col("value").top_k.alias("top_k"),
    Polars.col("value").bottom_k.alias("bottom_k")
  ]
)
# =>
# shape: (5, 2)
# ┌───────┬──────────┐
# │ top_k ┆ bottom_k │
# │ ---   ┆ ---      │
# │ i64   ┆ i64      │
# ╞═══════╪══════════╡
# │ 99    ┆ 1        │
# │ 98    ┆ 2        │
# │ 4     ┆ 3        │
# │ 3     ┆ 4        │
# │ 2     ┆ 98       │
# └───────┴──────────┘

Parameters:

  • k (Integer) (defaults to: 5) —

    Number of elements to return.

Returns:



1391
1392
1393
1394
 
# File 'lib/polars/expr.rb', line 1391

def bottom_k(k: 5)
  k = Utils.parse_as_expression(k)
  wrap_expr(_rbexpr.bottom_k(k))
end

#cast(dtype, strict: true) ⇒ Expr

Cast between data types.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [1, 2, 3],
    "b" => ["4", "5", "6"]
  }
)
df.with_columns(
  [
    Polars.col("a").cast(:f64),
    Polars.col("b").cast(:i32)
  ]
)
# =>
# shape: (3, 2)
# ┌─────┬─────┐
# │ a   ┆ b   │
# │ --- ┆ --- │
# │ f64 ┆ i32 │
# ╞═════╪═════╡
# │ 1.0 ┆ 4   │
# │ 2.0 ┆ 5   │
# │ 3.0 ┆ 6   │
# └─────┴─────┘

Parameters:

  • dtype (Symbol) —

    DataType to cast to.

  • strict (Boolean) (defaults to: true) —

    Throw an error if a cast could not be done. For instance, due to an overflow.

Returns:



1238
1239
1240
1241
 
# File 'lib/polars/expr.rb', line 1238

def cast(dtype, strict: true)
  dtype = Utils.rb_type_to_dtype(dtype)
  wrap_expr(_rbexpr.cast(dtype, strict))
end

#cat ⇒ CatExpr

Create an object namespace of all categorical related methods.

Returns:



5323
5324
5325
 
# File 'lib/polars/expr.rb', line 5323

def cat
  CatExpr.new(self)
end

#ceil ⇒ Expr

Rounds up to the nearest integer value.

Only works on floating point Series.

Examples:

df = Polars::DataFrame.new({"a" => [0.3, 0.5, 1.0, 1.1]})
df.select(Polars.col("a").ceil)
# =>
# shape: (4, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 1.0 │
# │ 1.0 │
# │ 1.0 │
# │ 2.0 │
# └─────┘

Returns:



1117
1118
1119
 
# File 'lib/polars/expr.rb', line 1117

def ceil
  wrap_expr(_rbexpr.ceil)
end

#clip(lower_bound, upper_bound) ⇒ Expr

Set values outside the given boundaries to the boundary value.

Only works for numeric and temporal columns. If you want to clip other data types, consider writing a when-then-otherwise expression.

Examples:

df = Polars::DataFrame.new({"foo" => [-50, 5, nil, 50]})
df.with_column(Polars.col("foo").clip(1, 10).alias("foo_clipped"))
# =>
# shape: (4, 2)
# ┌──────┬─────────────┐
# │ foo  ┆ foo_clipped │
# │ ---  ┆ ---         │
# │ i64  ┆ i64         │
# ╞══════╪═════════════╡
# │ -50  ┆ 1           │
# │ 5    ┆ 5           │
# │ null ┆ null        │
# │ 50   ┆ 10          │
# └──────┴─────────────┘

Parameters:

  • lower_bound (Numeric) —

    Minimum value.

  • upper_bound (Numeric) —

    Maximum value.

Returns:



4411
4412
4413
4414
4415
4416
4417
4418
4419
 
# File 'lib/polars/expr.rb', line 4411

def clip(lower_bound, upper_bound)
  if !lower_bound.nil?
    lower_bound = Utils.parse_as_expression(lower_bound, str_as_lit: true)
  end
  if !upper_bound.nil?
    upper_bound = Utils.parse_as_expression(upper_bound, str_as_lit: true)
  end
  wrap_expr(_rbexpr.clip(lower_bound, upper_bound))
end

#clip_max(upper_bound) ⇒ Expr

Clip (limit) the values in an array to a max boundary.

Only works for numerical types.

If you want to clip other dtypes, consider writing a "when, then, otherwise" expression. See when for more information.

Examples:

df = Polars::DataFrame.new({"foo" => [-50, 5, nil, 50]})
df.with_column(Polars.col("foo").clip_max(0).alias("foo_clipped"))
# =>
# shape: (4, 2)
# ┌──────┬─────────────┐
# │ foo  ┆ foo_clipped │
# │ ---  ┆ ---         │
# │ i64  ┆ i64         │
# ╞══════╪═════════════╡
# │ -50  ┆ -50         │
# │ 5    ┆ 0           │
# │ null ┆ null        │
# │ 50   ┆ 0           │
# └──────┴─────────────┘

Parameters:

  • upper_bound (Numeric) —

    Maximum value.

Returns:



4479
4480
4481
 
# File 'lib/polars/expr.rb', line 4479

def clip_max(upper_bound)
  clip(nil, upper_bound)
end

#clip_min(lower_bound) ⇒ Expr

Clip (limit) the values in an array to a min boundary.

Only works for numerical types.

If you want to clip other dtypes, consider writing a "when, then, otherwise" expression. See when for more information.

Examples:

df = Polars::DataFrame.new({"foo" => [-50, 5, nil, 50]})
df.with_column(Polars.col("foo").clip_min(0).alias("foo_clipped"))
# =>
# shape: (4, 2)
# ┌──────┬─────────────┐
# │ foo  ┆ foo_clipped │
# │ ---  ┆ ---         │
# │ i64  ┆ i64         │
# ╞══════╪═════════════╡
# │ -50  ┆ 0           │
# │ 5    ┆ 5           │
# │ null ┆ null        │
# │ 50   ┆ 50          │
# └──────┴─────────────┘

Parameters:

  • lower_bound (Numeric) —

    Minimum value.

Returns:



4448
4449
4450
 
# File 'lib/polars/expr.rb', line 4448

def clip_min(lower_bound)
  clip(lower_bound, nil)
end

#cos ⇒ Expr

Compute the element-wise value for the cosine.

Examples:

df = Polars::DataFrame.new({"a" => [0.0]})
df.select(Polars.col("a").cos)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 1.0 │
# └─────┘

Returns:



4589
4590
4591
 
# File 'lib/polars/expr.rb', line 4589

def cos
  wrap_expr(_rbexpr.cos)
end

#cosh ⇒ Expr

Compute the element-wise value for the hyperbolic cosine.

Examples:

df = Polars::DataFrame.new({"a" => [1.0]})
df.select(Polars.col("a").cosh)
# =>
# shape: (1, 1)
# ┌──────────┐
# │ a        │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ 1.543081 │
# └──────────┘

Returns:



4709
4710
4711
 
# File 'lib/polars/expr.rb', line 4709

def cosh
  wrap_expr(_rbexpr.cosh)
end

#count ⇒ Expr

Count the number of values in this expression.

Examples:

df = Polars::DataFrame.new({"a" => [8, 9, 10], "b" => [nil, 4, 4]})
df.select(Polars.all.count)
# =>
# shape: (1, 2)
# ┌─────┬─────┐
# │ a   ┆ b   │
# │ --- ┆ --- │
# │ u32 ┆ u32 │
# ╞═════╪═════╡
# │ 3   ┆ 3   │
# └─────┴─────┘

Returns:



723
724
725
726
727
 
# File 'lib/polars/expr.rb', line 723

def count
  warn "`Expr#count` will exclude null values in 0.9.0. Use `Expr#length` instead."
  # wrap_expr(_rbexpr.count)
  wrap_expr(_rbexpr.len)
end

#cum_count(reverse: false) ⇒ Expr Also known as: cumcount

Get an array with the cumulative count computed at every element.

Counting from 0 to len

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3, 4]})
df.select(
  [
    Polars.col("a").cum_count,
    Polars.col("a").cum_count(reverse: true).alias("a_reverse")
  ]
)
# =>
# shape: (4, 2)
# ┌─────┬───────────┐
# │ a   ┆ a_reverse │
# │ --- ┆ ---       │
# │ u32 ┆ u32       │
# ╞═════╪═══════════╡
# │ 0   ┆ 3         │
# │ 1   ┆ 2         │
# │ 2   ┆ 1         │
# │ 3   ┆ 0         │
# └─────┴───────────┘

Parameters:

  • reverse (Boolean) (defaults to: false) —

    Reverse the operation.

Returns:



1066
1067
1068
 
# File 'lib/polars/expr.rb', line 1066

def cum_count(reverse: false)
  wrap_expr(_rbexpr.cum_count(reverse))
end

#cum_max(reverse: false) ⇒ Expr Also known as: cummax

Get an array with the cumulative max computed at every element.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3, 4]})
df.select(
  [
    Polars.col("a").cum_max,
    Polars.col("a").cum_max(reverse: true).alias("a_reverse")
  ]
)
# =>
# shape: (4, 2)
# ┌─────┬───────────┐
# │ a   ┆ a_reverse │
# │ --- ┆ ---       │
# │ i64 ┆ i64       │
# ╞═════╪═══════════╡
# │ 1   ┆ 4         │
# │ 2   ┆ 4         │
# │ 3   ┆ 4         │
# │ 4   ┆ 4         │
# └─────┴───────────┘

Parameters:

  • reverse (Boolean) (defaults to: false) —

    Reverse the operation.

Returns:



1032
1033
1034
 
# File 'lib/polars/expr.rb', line 1032

def cum_max(reverse: false)
  wrap_expr(_rbexpr.cum_max(reverse))
end

#cum_min(reverse: false) ⇒ Expr Also known as: cummin

Get an array with the cumulative min computed at every element.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3, 4]})
df.select(
  [
    Polars.col("a").cum_min,
    Polars.col("a").cum_min(reverse: true).alias("a_reverse")
  ]
)
# =>
# shape: (4, 2)
# ┌─────┬───────────┐
# │ a   ┆ a_reverse │
# │ --- ┆ ---       │
# │ i64 ┆ i64       │
# ╞═════╪═══════════╡
# │ 1   ┆ 1         │
# │ 1   ┆ 2         │
# │ 1   ┆ 3         │
# │ 1   ┆ 4         │
# └─────┴───────────┘

Parameters:

  • reverse (Boolean) (defaults to: false) —

    Reverse the operation.

Returns:



1000
1001
1002
 
# File 'lib/polars/expr.rb', line 1000

def cum_min(reverse: false)
  wrap_expr(_rbexpr.cum_min(reverse))
end

#cum_prod(reverse: false) ⇒ Expr Also known as: cumprod

Note:

Dtypes in :i8, :u8, :i16, and :u16 are cast to :i64 before summing to prevent overflow issues.

Get an array with the cumulative product computed at every element.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3, 4]})
df.select(
  [
    Polars.col("a").cum_prod,
    Polars.col("a").cum_prod(reverse: true).alias("a_reverse")
  ]
)
# =>
# shape: (4, 2)
# ┌─────┬───────────┐
# │ a   ┆ a_reverse │
# │ --- ┆ ---       │
# │ i64 ┆ i64       │
# ╞═════╪═══════════╡
# │ 1   ┆ 24        │
# │ 2   ┆ 24        │
# │ 6   ┆ 12        │
# │ 24  ┆ 4         │
# └─────┴───────────┘

Parameters:

  • reverse (Boolean) (defaults to: false) —

    Reverse the operation.

Returns:



968
969
970
 
# File 'lib/polars/expr.rb', line 968

def cum_prod(reverse: false)
  wrap_expr(_rbexpr.cum_prod(reverse))
end

#cum_sum(reverse: false) ⇒ Expr Also known as: cumsum

Note:

Dtypes in :i8, :u8, :i16, and :u16 are cast to :i64 before summing to prevent overflow issues.

Get an array with the cumulative sum computed at every element.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3, 4]})
df.select(
  [
    Polars.col("a").cum_sum,
    Polars.col("a").cum_sum(reverse: true).alias("a_reverse")
  ]
)
# =>
# shape: (4, 2)
# ┌─────┬───────────┐
# │ a   ┆ a_reverse │
# │ --- ┆ ---       │
# │ i64 ┆ i64       │
# ╞═════╪═══════════╡
# │ 1   ┆ 10        │
# │ 3   ┆ 9         │
# │ 6   ┆ 7         │
# │ 10  ┆ 4         │
# └─────┴───────────┘

Parameters:

  • reverse (Boolean) (defaults to: false) —

    Reverse the operation.

Returns:



932
933
934
 
# File 'lib/polars/expr.rb', line 932

def cum_sum(reverse: false)
  wrap_expr(_rbexpr.cum_sum(reverse))
end

#cumulative_eval(expr, min_periods: 1, parallel: false) ⇒ Expr

Note:

This functionality is experimental and may change without it being considered a breaking change.

Note:

This can be really slow as it can have O(n^2) complexity. Don't use this for operations that visit all elements.

Run an expression over a sliding window that increases 1 slot every iteration.

Examples:

df = Polars::DataFrame.new({"values" => [1, 2, 3, 4, 5]})
df.select(
  [
    Polars.col("values").cumulative_eval(
      Polars.element.first - Polars.element.last ** 2
    )
  ]
)
# =>
# shape: (5, 1)
# ┌────────┐
# │ values │
# │ ---    │
# │ f64    │
# ╞════════╡
# │ 0.0    │
# │ -3.0   │
# │ -8.0   │
# │ -15.0  │
# │ -24.0  │
# └────────┘

Parameters:

  • expr (Expr) —

    Expression to evaluate

  • min_periods (Integer) (defaults to: 1) —

    Number of valid values there should be in the window before the expression is evaluated. valid values = length - null_count

  • parallel (Boolean) (defaults to: false) —

    Run in parallel. Don't do this in a group by or another operation that already has much parallelization.

Returns:



5204
5205
5206
5207
5208
 
# File 'lib/polars/expr.rb', line 5204

def cumulative_eval(expr, min_periods: 1, parallel: false)
  wrap_expr(
    _rbexpr.cumulative_eval(expr._rbexpr, min_periods, parallel)
  )
end

#cut(breaks, labels: nil, left_closed: false, include_breaks: false) ⇒ Expr

Bin continuous values into discrete categories.

Examples:

Divide a column into three categories.

df = Polars::DataFrame.new({"foo" => [-2, -1, 0, 1, 2]})
df.with_columns(
  Polars.col("foo").cut([-1, 1], labels: ["a", "b", "c"]).alias("cut")
)
# =>
# shape: (5, 2)
# ┌─────┬─────┐
# │ foo ┆ cut │
# │ --- ┆ --- │
# │ i64 ┆ cat │
# ╞═════╪═════╡
# │ -2  ┆ a   │
# │ -1  ┆ a   │
# │ 0   ┆ b   │
# │ 1   ┆ b   │
# │ 2   ┆ c   │
# └─────┴─────┘

Add both the category and the breakpoint.

df.with_columns(
  Polars.col("foo").cut([-1, 1], include_breaks: true).alias("cut")
).unnest("cut")
# =>
# shape: (5, 3)
# ┌─────┬──────┬────────────┐
# │ foo ┆ brk  ┆ foo_bin    │
# │ --- ┆ ---  ┆ ---        │
# │ i64 ┆ f64  ┆ cat        │
# ╞═════╪══════╪════════════╡
# │ -2  ┆ -1.0 ┆ (-inf, -1] │
# │ -1  ┆ -1.0 ┆ (-inf, -1] │
# │ 0   ┆ 1.0  ┆ (-1, 1]    │
# │ 1   ┆ 1.0  ┆ (-1, 1]    │
# │ 2   ┆ inf  ┆ (1, inf]   │
# └─────┴──────┴────────────┘

Parameters:

  • breaks (Array) —

    List of unique cut points.

  • labels (Array) (defaults to: nil) —

    Names of the categories. The number of labels must be equal to the number of cut points plus one.

  • left_closed (Boolean) (defaults to: false) —

    Set the intervals to be left-closed instead of right-closed.

  • include_breaks (Boolean) (defaults to: false) —

    Include a column with the right endpoint of the bin each observation falls in. This will change the data type of the output from a Categorical to a Struct.

Returns:



2534
2535
2536
 
# File 'lib/polars/expr.rb', line 2534

def cut(breaks, labels: nil, left_closed: false, include_breaks: false)
  wrap_expr(_rbexpr.cut(breaks, labels, left_closed, include_breaks))
end

#diff(n: 1, null_behavior: "ignore") ⇒ Expr

Calculate the n-th discrete difference.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [20, 10, 30]
  }
)
df.select(Polars.col("a").diff)
# =>
# shape: (3, 1)
# ┌──────┐
# │ a    │
# │ ---  │
# │ i64  │
# ╞══════╡
# │ null │
# │ -10  │
# │ 20   │
# └──────┘

Parameters:

  • n (Integer) (defaults to: 1) —

    Number of slots to shift.

  • null_behavior ("ignore", "drop") (defaults to: "ignore") —

    How to handle null values.

Returns:



4282
4283
4284
 
# File 'lib/polars/expr.rb', line 4282

def diff(n: 1, null_behavior: "ignore")
  wrap_expr(_rbexpr.diff(n, null_behavior))
end

#dot(other) ⇒ Expr

Compute the dot/inner product between two Expressions.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [1, 3, 5],
    "b" => [2, 4, 6]
  }
)
df.select(Polars.col("a").dot(Polars.col("b")))
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ i64 │
# ╞═════╡
# │ 44  │
# └─────┘

Parameters:

  • other (Expr) —

    Expression to compute dot product with.

Returns:



1171
1172
1173
1174
 
# File 'lib/polars/expr.rb', line 1171

def dot(other)
  other = Utils.expr_to_lit_or_expr(other, str_to_lit: false)
  wrap_expr(_rbexpr.dot(other._rbexpr))
end

#drop_nans ⇒ Expr

Drop floating point NaN values.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [8, 9, 10, 11],
    "b" => [nil, 4.0, 4.0, Float::NAN]
  }
)
df.select(Polars.col("b").drop_nans)
# =>
# shape: (3, 1)
# ┌──────┐
# │ b    │
# │ ---  │
# │ f64  │
# ╞══════╡
# │ null │
# │ 4.0  │
# │ 4.0  │
# └──────┘

Returns:



897
898
899
 
# File 'lib/polars/expr.rb', line 897

def drop_nans
  wrap_expr(_rbexpr.drop_nans)
end

#drop_nulls ⇒ Expr

Drop null values.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [8, 9, 10, 11],
    "b" => [nil, 4.0, 4.0, Float::NAN]
  }
)
df.select(Polars.col("b").drop_nulls)
# =>
# shape: (3, 1)
# ┌─────┐
# │ b   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 4.0 │
# │ 4.0 │
# │ NaN │
# └─────┘

Returns:



870
871
872
 
# File 'lib/polars/expr.rb', line 870

def drop_nulls
  wrap_expr(_rbexpr.drop_nulls)
end

#dt ⇒ DateTimeExpr

Create an object namespace of all datetime related methods.

Returns:



5330
5331
5332
 
# File 'lib/polars/expr.rb', line 5330

def dt
  DateTimeExpr.new(self)
end

#entropy(base: 2, normalize: true) ⇒ Expr

Computes the entropy.

Uses the formula -sum(pk * log(pk) where pk are discrete probabilities.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3]})
df.select(Polars.col("a").entropy(base: 2))
# =>
# shape: (1, 1)
# ┌──────────┐
# │ a        │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ 1.459148 │
# └──────────┘

df.select(Polars.col("a").entropy(base: 2, normalize: false))
# =>
# shape: (1, 1)
# ┌───────────┐
# │ a         │
# │ ---       │
# │ f64       │
# ╞═══════════╡
# │ -6.754888 │
# └───────────┘

Parameters:

  • base (Float) (defaults to: 2) —

    Given base, defaults to e.

  • normalize (Boolean) (defaults to: true) —

    Normalize pk if it doesn't sum to 1.

Returns:



5157
5158
5159
 
# File 'lib/polars/expr.rb', line 5157

def entropy(base: 2, normalize: true)
  wrap_expr(_rbexpr.entropy(base, normalize))
end

#ewm_mean(com: nil, span: nil, half_life: nil, alpha: nil, adjust: true, min_periods: 1, ignore_nulls: true) ⇒ Expr

Exponentially-weighted moving average.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3]})
df.select(Polars.col("a").ewm_mean(com: 1))
# =>
# shape: (3, 1)
# ┌──────────┐
# │ a        │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ 1.0      │
# │ 1.666667 │
# │ 2.428571 │
# └──────────┘

Returns:



4921
4922
4923
4924
4925
4926
4927
4928
4929
4930
4931
4932
 
# File 'lib/polars/expr.rb', line 4921

def ewm_mean(
  com: nil,
  span: nil,
  half_life: nil,
  alpha: nil,
  adjust: true,
  min_periods: 1,
  ignore_nulls: true
)
  alpha = _prepare_alpha(com, span, half_life, alpha)
  wrap_expr(_rbexpr.ewm_mean(alpha, adjust, min_periods, ignore_nulls))
end

#ewm_std(com: nil, span: nil, half_life: nil, alpha: nil, adjust: true, bias: false, min_periods: 1, ignore_nulls: true) ⇒ Expr

Exponentially-weighted moving standard deviation.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3]})
df.select(Polars.col("a").ewm_std(com: 1))
# =>
# shape: (3, 1)
# ┌──────────┐
# │ a        │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ 0.0      │
# │ 0.707107 │
# │ 0.963624 │
# └──────────┘

Returns:



4952
4953
4954
4955
4956
4957
4958
4959
4960
4961
4962
4963
4964
 
# File 'lib/polars/expr.rb', line 4952

def ewm_std(
  com: nil,
  span: nil,
  half_life: nil,
  alpha: nil,
  adjust: true,
  bias: false,
  min_periods: 1,
  ignore_nulls: true
)
  alpha = _prepare_alpha(com, span, half_life, alpha)
  wrap_expr(_rbexpr.ewm_std(alpha, adjust, bias, min_periods, ignore_nulls))
end

#ewm_var(com: nil, span: nil, half_life: nil, alpha: nil, adjust: true, bias: false, min_periods: 1, ignore_nulls: true) ⇒ Expr

Exponentially-weighted moving variance.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3]})
df.select(Polars.col("a").ewm_var(com: 1))
# =>
# shape: (3, 1)
# ┌──────────┐
# │ a        │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ 0.0      │
# │ 0.5      │
# │ 0.928571 │
# └──────────┘

Returns:



4984
4985
4986
4987
4988
4989
4990
4991
4992
4993
4994
4995
4996
 
# File 'lib/polars/expr.rb', line 4984

def ewm_var(
  com: nil,
  span: nil,
  half_life: nil,
  alpha: nil,
  adjust: true,
  bias: false,
  min_periods: 1,
  ignore_nulls: true
)
  alpha = _prepare_alpha(com, span, half_life, alpha)
  wrap_expr(_rbexpr.ewm_var(alpha, adjust, bias, min_periods, ignore_nulls))
end

#exclude(columns) ⇒ Expr

Exclude certain columns from a wildcard/regex selection.

You may also use regexes in the exclude list. They must start with ^ and end with $.

Examples:

df = Polars::DataFrame.new(
  {
    "aa" => [1, 2, 3],
    "ba" => ["a", "b", nil],
    "cc" => [nil, 2.5, 1.5]
  }
)
df.select(Polars.all.exclude("ba"))
# =>
# shape: (3, 2)
# ┌─────┬──────┐
# │ aa  ┆ cc   │
# │ --- ┆ ---  │
# │ i64 ┆ f64  │
# ╞═════╪══════╡
# │ 1   ┆ null │
# │ 2   ┆ 2.5  │
# │ 3   ┆ 1.5  │
# └─────┴──────┘

Parameters:

  • columns (Object) —

    Column(s) to exclude from selection. This can be:

    • a column name, or multiple column names
    • a regular expression starting with ^ and ending with $
    • a dtype or multiple dtypes

Returns:



368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
 
# File 'lib/polars/expr.rb', line 368

def exclude(columns)
  if columns.is_a?(::String)
    columns = [columns]
    return wrap_expr(_rbexpr.exclude(columns))
  elsif !columns.is_a?(::Array)
    columns = [columns]
    return wrap_expr(_rbexpr.exclude_dtype(columns))
  end

  if !columns.all? { |a| a.is_a?(::String) } || !columns.all? { |a| Utils.is_polars_dtype(a) }
    raise ArgumentError, "input should be all string or all DataType"
  end

  if columns[0].is_a?(::String)
    wrap_expr(_rbexpr.exclude(columns))
  else
    wrap_expr(_rbexpr.exclude_dtype(columns))
  end
end

#exp ⇒ Expr

Compute the exponential, element-wise.

Examples:

df = Polars::DataFrame.new({"values" => [1.0, 2.0, 4.0]})
df.select(Polars.col("values").exp)
# =>
# shape: (3, 1)
# ┌──────────┐
# │ values   │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ 2.718282 │
# │ 7.389056 │
# │ 54.59815 │
# └──────────┘

Returns:



292
293
294
 
# File 'lib/polars/expr.rb', line 292

def exp
  wrap_expr(_rbexpr.exp)
end

#explode ⇒ Expr

Explode a list or utf8 Series.

This means that every item is expanded to a new row.

Examples:

df = Polars::DataFrame.new({"b" => [[1, 2, 3], [4, 5, 6]]})
df.select(Polars.col("b").explode)
# =>
# shape: (6, 1)
# ┌─────┐
# │ b   │
# │ --- │
# │ i64 │
# ╞═════╡
# │ 1   │
# │ 2   │
# │ 3   │
# │ 4   │
# │ 5   │
# │ 6   │
# └─────┘

Returns:



2933
2934
2935
 
# File 'lib/polars/expr.rb', line 2933

def explode
  wrap_expr(_rbexpr.explode)
end

#extend_constant(value, n) ⇒ Expr

Extend the Series with given number of values.

Examples:

df = Polars::DataFrame.new({"values" => [1, 2, 3]})
df.select(Polars.col("values").extend_constant(99, 2))
# =>
# shape: (5, 1)
# ┌────────┐
# │ values │
# │ ---    │
# │ i64    │
# ╞════════╡
# │ 1      │
# │ 2      │
# │ 3      │
# │ 99     │
# │ 99     │
# └────────┘

Parameters:

  • value (Object) —

    The value to extend the Series with. This value may be nil to fill with nulls.

  • n (Integer) —

    The number of values to extend.

Returns:



5024
5025
5026
 
# File 'lib/polars/expr.rb', line 5024

def extend_constant(value, n)
  wrap_expr(_rbexpr.extend_constant(value, n))
end

#fill_nan(fill_value) ⇒ Expr

Fill floating point NaN value with a fill value.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [1.0, nil, Float::NAN],
    "b" => [4.0, Float::NAN, 6]
  }
)
df.fill_nan("zero")
# =>
# shape: (3, 2)
# ┌──────┬──────┐
# │ a    ┆ b    │
# │ ---  ┆ ---  │
# │ str  ┆ str  │
# ╞══════╪══════╡
# │ 1.0  ┆ 4.0  │
# │ null ┆ zero │
# │ zero ┆ 6.0  │
# └──────┴──────┘

Returns:



1766
1767
1768
1769
 
# File 'lib/polars/expr.rb', line 1766

def fill_nan(fill_value)
  fill_value = Utils.expr_to_lit_or_expr(fill_value, str_to_lit: true)
  wrap_expr(_rbexpr.fill_nan(fill_value._rbexpr))
end

#fill_null(value = nil, strategy: nil, limit: nil) ⇒ Expr

Fill null values using the specified value or strategy.

To interpolate over null values see interpolate.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [1, 2, nil],
    "b" => [4, nil, 6]
  }
)
df.fill_null(strategy: "zero")
# =>
# shape: (3, 2)
# ┌─────┬─────┐
# │ a   ┆ b   │
# │ --- ┆ --- │
# │ i64 ┆ i64 │
# ╞═════╪═════╡
# │ 1   ┆ 4   │
# │ 2   ┆ 0   │
# │ 0   ┆ 6   │
# └─────┴─────┘

df.fill_null(99)
# =>
# shape: (3, 2)
# ┌─────┬─────┐
# │ a   ┆ b   │
# │ --- ┆ --- │
# │ i64 ┆ i64 │
# ╞═════╪═════╡
# │ 1   ┆ 4   │
# │ 2   ┆ 99  │
# │ 99  ┆ 6   │
# └─────┴─────┘

df.fill_null(strategy: "forward")
# =>
# shape: (3, 2)
# ┌─────┬─────┐
# │ a   ┆ b   │
# │ --- ┆ --- │
# │ i64 ┆ i64 │
# ╞═════╪═════╡
# │ 1   ┆ 4   │
# │ 2   ┆ 4   │
# │ 2   ┆ 6   │
# └─────┴─────┘

Parameters:

  • value (Object) (defaults to: nil) —

    Value used to fill null values.

  • strategy (nil, "forward", "backward", "min", "max", "mean", "zero", "one") (defaults to: nil) —

    Strategy used to fill null values.

  • limit (Integer) (defaults to: nil) —

    Number of consecutive null values to fill when using the 'forward' or 'backward' strategy.

Returns:



1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
 
# File 'lib/polars/expr.rb', line 1726

def fill_null(value = nil, strategy: nil, limit: nil)
  if !value.nil? && !strategy.nil?
    raise ArgumentError, "cannot specify both 'value' and 'strategy'."
  elsif value.nil? && strategy.nil?
    raise ArgumentError, "must specify either a fill 'value' or 'strategy'"
  elsif ["forward", "backward"].include?(strategy) && !limit.nil?
    raise ArgumentError, "can only specify 'limit' when strategy is set to 'backward' or 'forward'"
  end

  if !value.nil?
    value = Utils.expr_to_lit_or_expr(value, str_to_lit: true)
    wrap_expr(_rbexpr.fill_null(value._rbexpr))
  else
    wrap_expr(_rbexpr.fill_null_with_strategy(strategy, limit))
  end
end

#filter(predicate) ⇒ Expr

Filter a single column.

Mostly useful in an aggregation context. If you want to filter on a DataFrame level, use LazyFrame#filter.

Examples:

df = Polars::DataFrame.new(
  {
    "group_col" => ["g1", "g1", "g2"],
    "b" => [1, 2, 3]
  }
)
(
  df.group_by("group_col").agg(
    [
      Polars.col("b").filter(Polars.col("b") < 2).sum.alias("lt"),
      Polars.col("b").filter(Polars.col("b") >= 2).sum.alias("gte")
    ]
  )
).sort("group_col")
# =>
# shape: (2, 3)
# ┌───────────┬─────┬─────┐
# │ group_col ┆ lt  ┆ gte │
# │ ---       ┆ --- ┆ --- │
# │ str       ┆ i64 ┆ i64 │
# ╞═══════════╪═════╪═════╡
# │ g1        ┆ 1   ┆ 2   │
# │ g2        ┆ 0   ┆ 3   │
# └───────────┴─────┴─────┘

Parameters:

  • predicate (Expr) —

    Boolean expression.

Returns:



2717
2718
2719
 
# File 'lib/polars/expr.rb', line 2717

def filter(predicate)
  wrap_expr(_rbexpr.filter(predicate._rbexpr))
end

#first ⇒ Expr

Get the first value.

Examples:

df = Polars::DataFrame.new({"a" => [1, 1, 2]})
df.select(Polars.col("a").first)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ i64 │
# ╞═════╡
# │ 1   │
# └─────┘

Returns:



2199
2200
2201
 
# File 'lib/polars/expr.rb', line 2199

def first
  wrap_expr(_rbexpr.first)
end

#flatten ⇒ Expr

Explode a list or utf8 Series. This means that every item is expanded to a new row.

Alias for #explode.

Examples:

df = Polars::DataFrame.new(
  {
    "group" => ["a", "b", "b"],
    "values" => [[1, 2], [2, 3], [4]]
  }
)
df.group_by("group").agg(Polars.col("values").flatten)
# =>
# shape: (2, 2)
# ┌───────┬───────────┐
# │ group ┆ values    │
# │ ---   ┆ ---       │
# │ str   ┆ list[i64] │
# ╞═══════╪═══════════╡
# │ a     ┆ [1, 2]    │
# │ b     ┆ [2, 3, 4] │
# └───────┴───────────┘

Returns:



2906
2907
2908
 
# File 'lib/polars/expr.rb', line 2906

def flatten
  wrap_expr(_rbexpr.explode)
end

#floor ⇒ Expr

Rounds down to the nearest integer value.

Only works on floating point Series.

Examples:

df = Polars::DataFrame.new({"a" => [0.3, 0.5, 1.0, 1.1]})
df.select(Polars.col("a").floor)
# =>
# shape: (4, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 0.0 │
# │ 0.0 │
# │ 1.0 │
# │ 1.0 │
# └─────┘

Returns:



1092
1093
1094
 
# File 'lib/polars/expr.rb', line 1092

def floor
  wrap_expr(_rbexpr.floor)
end

#floordiv(other) ⇒ Expr

Performs floor division.

Returns:



74
75
76
 
# File 'lib/polars/expr.rb', line 74

def floordiv(other)
  wrap_expr(_rbexpr.floordiv(_to_rbexpr(other)))
end

#forward_fill(limit: nil) ⇒ Expr

Fill missing values with the latest seen values.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [1, 2, nil],
    "b" => [4, nil, 6]
  }
)
df.select(Polars.all.forward_fill)
# =>
# shape: (3, 2)
# ┌─────┬─────┐
# │ a   ┆ b   │
# │ --- ┆ --- │
# │ i64 ┆ i64 │
# ╞═════╪═════╡
# │ 1   ┆ 4   │
# │ 2   ┆ 4   │
# │ 2   ┆ 6   │
# └─────┴─────┘

Parameters:

  • limit (Integer) (defaults to: nil) —

    The number of consecutive null values to forward fill.

Returns:



1797
1798
1799
 
# File 'lib/polars/expr.rb', line 1797

def forward_fill(limit: nil)
  wrap_expr(_rbexpr.forward_fill(limit))
end

#gather(indices) ⇒ Expr Also known as: take

Take values by index.

Examples:

df = Polars::DataFrame.new(
  {
    "group" => [
      "one",
      "one",
      "one",
      "two",
      "two",
      "two"
    ],
    "value" => [1, 98, 2, 3, 99, 4]
  }
)
df.group_by("group", maintain_order: true).agg(Polars.col("value").take([2, 1]))
# =>
# shape: (2, 2)
# ┌───────┬───────────┐
# │ group ┆ value     │
# │ ---   ┆ ---       │
# │ str   ┆ list[i64] │
# ╞═══════╪═══════════╡
# │ one   ┆ [2, 98]   │
# │ two   ┆ [4, 99]   │
# └───────┴───────────┘

Parameters:

  • indices (Expr) —

    An expression that leads to a :u32 dtyped Series.

Returns:



1595
1596
1597
1598
1599
1600
1601
1602
 
# File 'lib/polars/expr.rb', line 1595

def gather(indices)
  if indices.is_a?(::Array)
    indices_lit = Polars.lit(Series.new("", indices, dtype: :u32))
  else
    indices_lit = Utils.expr_to_lit_or_expr(indices, str_to_lit: false)
  end
  wrap_expr(_rbexpr.gather(indices_lit._rbexpr))
end

#gather_every(n, offset = 0) ⇒ Expr Also known as: take_every

Take every nth value in the Series and return as a new Series.

Examples:

df = Polars::DataFrame.new({"foo" => [1, 2, 3, 4, 5, 6, 7, 8, 9]})
df.select(Polars.col("foo").gather_every(3))
# =>
# shape: (3, 1)
# ┌─────┐
# │ foo │
# │ --- │
# │ i64 │
# ╞═════╡
# │ 1   │
# │ 4   │
# │ 7   │
# └─────┘

Returns:



2955
2956
2957
 
# File 'lib/polars/expr.rb', line 2955

def gather_every(n, offset = 0)
  wrap_expr(_rbexpr.gather_every(n, offset))
end

#head(n = 10) ⇒ Expr

Get the first n rows.

Examples:

df = Polars::DataFrame.new({"foo" => [1, 2, 3, 4, 5, 6, 7]})
df.head(3)
# =>
# shape: (3, 1)
# ┌─────┐
# │ foo │
# │ --- │
# │ i64 │
# ╞═════╡
# │ 1   │
# │ 2   │
# │ 3   │
# └─────┘

Parameters:

  • n (Integer) (defaults to: 10) —

    Number of rows to return.

Returns:



2981
2982
2983
 
# File 'lib/polars/expr.rb', line 2981

def head(n = 10)
  wrap_expr(_rbexpr.head(n))
end

#implode ⇒ Expr

Aggregate to list.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [1, 2, 3],
    "b" => [4, 5, 6]
  }
)
df.select(Polars.all.implode)
# =>
# shape: (1, 2)
# ┌───────────┬───────────┐
# │ a         ┆ b         │
# │ ---       ┆ ---       │
# │ list[i64] ┆ list[i64] │
# ╞═══════════╪═══════════╡
# │ [1, 2, 3] ┆ [4, 5, 6] │
# └───────────┴───────────┘

Returns:



5260
5261
5262
 
# File 'lib/polars/expr.rb', line 5260

def implode
  wrap_expr(_rbexpr.implode)
end

#interpolate(method: "linear") ⇒ Expr

Fill nulls with linear interpolation over missing values.

Can also be used to regrid data to a new grid - see examples below.

Examples:

Fill nulls with linear interpolation

df = Polars::DataFrame.new(
  {
    "a" => [1, nil, 3],
    "b" => [1.0, Float::NAN, 3.0]
  }
)
df.select(Polars.all.interpolate)
# =>
# shape: (3, 2)
# ┌─────┬─────┐
# │ a   ┆ b   │
# │ --- ┆ --- │
# │ f64 ┆ f64 │
# ╞═════╪═════╡
# │ 1.0 ┆ 1.0 │
# │ 2.0 ┆ NaN │
# │ 3.0 ┆ 3.0 │
# └─────┴─────┘

Returns:



3337
3338
3339
 
# File 'lib/polars/expr.rb', line 3337

def interpolate(method: "linear")
  wrap_expr(_rbexpr.interpolate(method))
end

#is_between(start, _end, closed: "both") ⇒ Expr

Check if this expression is between start and end.

Examples:

df = Polars::DataFrame.new({"num" => [1, 2, 3, 4, 5]})
df.with_columns(Polars.col("num").is_between(2, 4).alias("is_between"))
# =>
# shape: (5, 2)
# ┌─────┬────────────┐
# │ num ┆ is_between │
# │ --- ┆ ---        │
# │ i64 ┆ bool       │
# ╞═════╪════════════╡
# │ 1   ┆ false      │
# │ 2   ┆ true       │
# │ 3   ┆ true       │
# │ 4   ┆ true       │
# │ 5   ┆ false      │
# └─────┴────────────┘

Use the closed argument to include or exclude the values at the bounds:

df.with_columns(
  Polars.col("num").is_between(2, 4, closed: "left").alias("is_between")
)
# =>
# shape: (5, 2)
# ┌─────┬────────────┐
# │ num ┆ is_between │
# │ --- ┆ ---        │
# │ i64 ┆ bool       │
# ╞═════╪════════════╡
# │ 1   ┆ false      │
# │ 2   ┆ true       │
# │ 3   ┆ true       │
# │ 4   ┆ false      │
# │ 5   ┆ false      │
# └─────┴────────────┘

You can also use strings as well as numeric/temporal values:

df = Polars::DataFrame.new({"a" => ["a", "b", "c", "d", "e"]})
df.with_columns(
  Polars.col("a")
    .is_between(Polars.lit("a"), Polars.lit("c"), closed: "both")
    .alias("is_between")
)
# =>
# shape: (5, 2)
# ┌─────┬────────────┐
# │ a   ┆ is_between │
# │ --- ┆ ---        │
# │ str ┆ bool       │
# ╞═════╪════════════╡
# │ a   ┆ true       │
# │ b   ┆ true       │
# │ c   ┆ true       │
# │ d   ┆ false      │
# │ e   ┆ false      │
# └─────┴────────────┘

Parameters:

  • start (Object) —

    Lower bound as primitive type or datetime.

  • _end (Object) —

    Upper bound as primitive type or datetime.

  • closed ("both", "left", "right", "none") (defaults to: "both") —

    Define which sides of the interval are closed (inclusive).

Returns:



3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
 
# File 'lib/polars/expr.rb', line 3184

def is_between(start, _end, closed: "both")
  start = Utils.expr_to_lit_or_expr(start, str_to_lit: false)
  _end = Utils.expr_to_lit_or_expr(_end, str_to_lit: false)

  case closed
  when "none"
    (self > start) & (self < _end)
  when "both"
    (self >= start) & (self <= _end)
  when "right"
    (self > start) & (self <= _end)
  when "left"
    (self >= start) & (self < _end)
  else
    raise ArgumentError, "closed must be one of 'left', 'right', 'both', or 'none'"
  end
end

#is_duplicated ⇒ Expr

Get mask of duplicated values.

Examples:

df = Polars::DataFrame.new({"a" => [1, 1, 2]})
df.select(Polars.col("a").is_duplicated)
# =>
# shape: (3, 1)
# ┌───────┐
# │ a     │
# │ ---   │
# │ bool  │
# ╞═══════╡
# │ true  │
# │ true  │
# │ false │
# └───────┘

Returns:



2356
2357
2358
 
# File 'lib/polars/expr.rb', line 2356

def is_duplicated
  wrap_expr(_rbexpr.is_duplicated)
end

#is_finite ⇒ Expr

Returns a boolean Series indicating which values are finite.

Examples:

df = Polars::DataFrame.new(
  {
    "A" => [1.0, 2],
    "B" => [3.0, Float::INFINITY]
  }
)
df.select(Polars.all.is_finite)
# =>
# shape: (2, 2)
# ┌──────┬───────┐
# │ A    ┆ B     │
# │ ---  ┆ ---   │
# │ bool ┆ bool  │
# ╞══════╪═══════╡
# │ true ┆ true  │
# │ true ┆ false │
# └──────┴───────┘

Returns:



576
577
578
 
# File 'lib/polars/expr.rb', line 576

def is_finite
  wrap_expr(_rbexpr.is_finite)
end

#is_first_distinct ⇒ Expr Also known as: is_first

Get a mask of the first unique value.

Examples:

df = Polars::DataFrame.new(
  {
    "num" => [1, 2, 3, 1, 5]
  }
)
df.with_column(Polars.col("num").is_first.alias("is_first"))
# =>
# shape: (5, 2)
# ┌─────┬──────────┐
# │ num ┆ is_first │
# │ --- ┆ ---      │
# │ i64 ┆ bool     │
# ╞═════╪══════════╡
# │ 1   ┆ true     │
# │ 2   ┆ true     │
# │ 3   ┆ true     │
# │ 1   ┆ false    │
# │ 5   ┆ true     │
# └─────┴──────────┘

Returns:



2333
2334
2335
 
# File 'lib/polars/expr.rb', line 2333

def is_first_distinct
  wrap_expr(_rbexpr.is_first_distinct)
end

#is_in(other) ⇒ Expr Also known as: in?

Check if elements of this expression are present in the other Series.

Examples:

df = Polars::DataFrame.new(
  {"sets" => [[1, 2, 3], [1, 2], [9, 10]], "optional_members" => [1, 2, 3]}
)
df.select([Polars.col("optional_members").is_in("sets").alias("contains")])
# =>
# shape: (3, 1)
# ┌──────────┐
# │ contains │
# │ ---      │
# │ bool     │
# ╞══════════╡
# │ true     │
# │ true     │
# │ false    │
# └──────────┘

Parameters:

  • other (Object) —

    Series or sequence of primitive type.

Returns:



3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
 
# File 'lib/polars/expr.rb', line 3069

def is_in(other)
  if other.is_a?(::Array)
    if other.length == 0
      other = Polars.lit(nil)
    else
      other = Polars.lit(Series.new(other))
    end
  else
    other = Utils.expr_to_lit_or_expr(other, str_to_lit: false)
  end
  wrap_expr(_rbexpr.is_in(other._rbexpr))
end

#is_infinite ⇒ Expr

Returns a boolean Series indicating which values are infinite.

Examples:

df = Polars::DataFrame.new(
  {
    "A" => [1.0, 2],
    "B" => [3.0, Float::INFINITY]
  }
)
df.select(Polars.all.is_infinite)
# =>
# shape: (2, 2)
# ┌───────┬───────┐
# │ A     ┆ B     │
# │ ---   ┆ ---   │
# │ bool  ┆ bool  │
# ╞═══════╪═══════╡
# │ false ┆ false │
# │ false ┆ true  │
# └───────┴───────┘

Returns:



602
603
604
 
# File 'lib/polars/expr.rb', line 602

def is_infinite
  wrap_expr(_rbexpr.is_infinite)
end

#is_nan ⇒ Expr

Note:

Floating point NaN (Not A Number) should not be confused with missing data represented as nil.

Returns a boolean Series indicating which values are NaN.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [1, 2, nil, 1, 5],
    "b" => [1.0, 2.0, Float::NAN, 1.0, 5.0]
  }
)
df.with_column(Polars.col(Polars::Float64).is_nan.suffix("_isnan"))
# =>
# shape: (5, 3)
# ┌──────┬─────┬─────────┐
# │ a    ┆ b   ┆ b_isnan │
# │ ---  ┆ --- ┆ ---     │
# │ i64  ┆ f64 ┆ bool    │
# ╞══════╪═════╪═════════╡
# │ 1    ┆ 1.0 ┆ false   │
# │ 2    ┆ 2.0 ┆ false   │
# │ null ┆ NaN ┆ true    │
# │ 1    ┆ 1.0 ┆ false   │
# │ 5    ┆ 5.0 ┆ false   │
# └──────┴─────┴─────────┘

Returns:



635
636
637
 
# File 'lib/polars/expr.rb', line 635

def is_nan
  wrap_expr(_rbexpr.is_nan)
end

#is_not ⇒ Expr

Negate a boolean expression.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [true, false, false],
    "b" => ["a", "b", nil]
  }
)
# =>
# shape: (3, 2)
# ┌───────┬──────┐
# │ a     ┆ b    │
# │ ---   ┆ ---  │
# │ bool  ┆ str  │
# ╞═══════╪══════╡
# │ true  ┆ a    │
# │ false ┆ b    │
# │ false ┆ null │
# └───────┴──────┘

df.select(Polars.col("a").is_not)
# =>
# shape: (3, 1)
# ┌───────┐
# │ a     │
# │ ---   │
# │ bool  │
# ╞═══════╡
# │ false │
# │ true  │
# │ true  │
# └───────┘

Returns:



492
493
494
 
# File 'lib/polars/expr.rb', line 492

def is_not
  wrap_expr(_rbexpr.is_not)
end

#is_not_nan ⇒ Expr

Note:

Floating point NaN (Not A Number) should not be confused with missing data represented as nil.

Returns a boolean Series indicating which values are not NaN.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [1, 2, nil, 1, 5],
    "b" => [1.0, 2.0, Float::NAN, 1.0, 5.0]
  }
)
df.with_column(Polars.col(Polars::Float64).is_not_nan.suffix("_is_not_nan"))
# =>
# shape: (5, 3)
# ┌──────┬─────┬──────────────┐
# │ a    ┆ b   ┆ b_is_not_nan │
# │ ---  ┆ --- ┆ ---          │
# │ i64  ┆ f64 ┆ bool         │
# ╞══════╪═════╪══════════════╡
# │ 1    ┆ 1.0 ┆ true         │
# │ 2    ┆ 2.0 ┆ true         │
# │ null ┆ NaN ┆ false        │
# │ 1    ┆ 1.0 ┆ true         │
# │ 5    ┆ 5.0 ┆ true         │
# └──────┴─────┴──────────────┘

Returns:



668
669
670
 
# File 'lib/polars/expr.rb', line 668

def is_not_nan
  wrap_expr(_rbexpr.is_not_nan)
end

#is_not_null ⇒ Expr

Returns a boolean Series indicating which values are not null.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [1, 2, nil, 1, 5],
    "b" => [1.0, 2.0, Float::NAN, 1.0, 5.0]
  }
)
df.with_column(Polars.all.is_not_null.suffix("_not_null"))
# =>
# shape: (5, 4)
# ┌──────┬─────┬────────────┬────────────┐
# │ a    ┆ b   ┆ a_not_null ┆ b_not_null │
# │ ---  ┆ --- ┆ ---        ┆ ---        │
# │ i64  ┆ f64 ┆ bool       ┆ bool       │
# ╞══════╪═════╪════════════╪════════════╡
# │ 1    ┆ 1.0 ┆ true       ┆ true       │
# │ 2    ┆ 2.0 ┆ true       ┆ true       │
# │ null ┆ NaN ┆ false      ┆ true       │
# │ 1    ┆ 1.0 ┆ true       ┆ true       │
# │ 5    ┆ 5.0 ┆ true       ┆ true       │
# └──────┴─────┴────────────┴────────────┘

Returns:



550
551
552
 
# File 'lib/polars/expr.rb', line 550

def is_not_null
  wrap_expr(_rbexpr.is_not_null)
end

#is_null ⇒ Expr

Returns a boolean Series indicating which values are null.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [1, 2, nil, 1, 5],
    "b" => [1.0, 2.0, Float::NAN, 1.0, 5.0]
  }
)
df.with_column(Polars.all.is_null.suffix("_isnull"))
# =>
# shape: (5, 4)
# ┌──────┬─────┬──────────┬──────────┐
# │ a    ┆ b   ┆ a_isnull ┆ b_isnull │
# │ ---  ┆ --- ┆ ---      ┆ ---      │
# │ i64  ┆ f64 ┆ bool     ┆ bool     │
# ╞══════╪═════╪══════════╪══════════╡
# │ 1    ┆ 1.0 ┆ false    ┆ false    │
# │ 2    ┆ 2.0 ┆ false    ┆ false    │
# │ null ┆ NaN ┆ true     ┆ false    │
# │ 1    ┆ 1.0 ┆ false    ┆ false    │
# │ 5    ┆ 5.0 ┆ false    ┆ false    │
# └──────┴─────┴──────────┴──────────┘

Returns:



521
522
523
 
# File 'lib/polars/expr.rb', line 521

def is_null
  wrap_expr(_rbexpr.is_null)
end

#is_unique ⇒ Expr

Get mask of unique values.

Examples:

df = Polars::DataFrame.new({"a" => [1, 1, 2]})
df.select(Polars.col("a").is_unique)
# =>
# shape: (3, 1)
# ┌───────┐
# │ a     │
# │ ---   │
# │ bool  │
# ╞═══════╡
# │ false │
# │ false │
# │ true  │
# └───────┘

Returns:



2305
2306
2307
 
# File 'lib/polars/expr.rb', line 2305

def is_unique
  wrap_expr(_rbexpr.is_unique)
end

#keep_name ⇒ Expr

Keep the original root name of the expression.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [1, 2],
    "b" => [3, 4]
  }
)
df.with_columns([(Polars.col("a") * 9).alias("c").keep_name])
# =>
# shape: (2, 2)
# ┌─────┬─────┐
# │ a   ┆ b   │
# │ --- ┆ --- │
# │ i64 ┆ i64 │
# ╞═════╪═════╡
# │ 9   ┆ 3   │
# │ 18  ┆ 4   │
# └─────┴─────┘

Returns:



410
411
412
 
# File 'lib/polars/expr.rb', line 410

def keep_name
  name.keep
end

#kurtosis(fisher: true, bias: true) ⇒ Expr

Compute the kurtosis (Fisher or Pearson) of a dataset.

Kurtosis is the fourth central moment divided by the square of the variance. If Fisher's definition is used, then 3.0 is subtracted from the result to give 0.0 for a normal distribution. If bias is False then the kurtosis is calculated using k statistics to eliminate bias coming from biased moment estimators

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3, 2, 1]})
df.select(Polars.col("a").kurtosis)
# =>
# shape: (1, 1)
# ┌───────────┐
# │ a         │
# │ ---       │
# │ f64       │
# ╞═══════════╡
# │ -1.153061 │
# └───────────┘

Parameters:

  • fisher (Boolean) (defaults to: true) —

    If true, Fisher's definition is used (normal ==> 0.0). If false, Pearson's definition is used (normal ==> 3.0).

  • bias (Boolean) (defaults to: true) —

    If false, the calculations are corrected for statistical bias.

Returns:



4380
4381
4382
 
# File 'lib/polars/expr.rb', line 4380

def kurtosis(fisher: true, bias: true)
  wrap_expr(_rbexpr.kurtosis(fisher, bias))
end

#last ⇒ Expr

Get the last value.

Examples:

df = Polars::DataFrame.new({"a" => [1, 1, 2]})
df.select(Polars.col("a").last)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ i64 │
# ╞═════╡
# │ 2   │
# └─────┘

Returns:



2219
2220
2221
 
# File 'lib/polars/expr.rb', line 2219

def last
  wrap_expr(_rbexpr.last)
end

#len ⇒ Expr Also known as: length

Count the number of values in this expression.

Examples:

df = Polars::DataFrame.new({"a" => [8, 9, 10], "b" => [nil, 4, 4]})
df.select(Polars.all.len)
# =>
# shape: (1, 2)
# ┌─────┬─────┐
# │ a   ┆ b   │
# │ --- ┆ --- │
# │ u32 ┆ u32 │
# ╞═════╪═════╡
# │ 3   ┆ 3   │
# └─────┴─────┘

Returns:



745
746
747
 
# File 'lib/polars/expr.rb', line 745

def len
  wrap_expr(_rbexpr.len)
end

#limit(n = 10) ⇒ Expr

Get the first n rows.

Alias for #head.

Parameters:

  • n (Integer) (defaults to: 10) —

    Number of rows to return.

Returns:



3018
3019
3020
 
# File 'lib/polars/expr.rb', line 3018

def limit(n = 10)
  head(n)
end

#list ⇒ ListExpr

Create an object namespace of all list related methods.

Returns:



5302
5303
5304
 
# File 'lib/polars/expr.rb', line 5302

def list
  ListExpr.new(self)
end

#log(base = Math::E) ⇒ Expr

Compute the logarithm to a given base.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3]})
df.select(Polars.col("a").log(2))
# =>
# shape: (3, 1)
# ┌──────────┐
# │ a        │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ 0.0      │
# │ 1.0      │
# │ 1.584963 │
# └──────────┘

Parameters:

  • base (Float) (defaults to: Math::E) —

    Given base, defaults to e.

Returns:



5118
5119
5120
 
# File 'lib/polars/expr.rb', line 5118

def log(base = Math::E)
  wrap_expr(_rbexpr.log(base))
end

#log10 ⇒ Expr

Compute the base 10 logarithm of the input array, element-wise.

Examples:

df = Polars::DataFrame.new({"values" => [1.0, 2.0, 4.0]})
df.select(Polars.col("values").log10)
# =>
# shape: (3, 1)
# ┌─────────┐
# │ values  │
# │ ---     │
# │ f64     │
# ╞═════════╡
# │ 0.0     │
# │ 0.30103 │
# │ 0.60206 │
# └─────────┘

Returns:



270
271
272
 
# File 'lib/polars/expr.rb', line 270

def log10
  log(10)
end

#lower_bound ⇒ Expr

Calculate the lower bound.

Returns a unit Series with the lowest value possible for the dtype of this expression.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3, 2, 1]})
df.select(Polars.col("a").lower_bound)
# =>
# shape: (1, 1)
# ┌──────────────────────┐
# │ a                    │
# │ ---                  │
# │ i64                  │
# ╞══════════════════════╡
# │ -9223372036854775808 │
# └──────────────────────┘

Returns:



4502
4503
4504
 
# File 'lib/polars/expr.rb', line 4502

def lower_bound
  wrap_expr(_rbexpr.lower_bound)
end

#map_alias(&f) ⇒ Expr

Rename the output of an expression by mapping a function over the root name.

Examples:

df = Polars::DataFrame.new(
  {
    "A" => [1, 2],
    "B" => [3, 4]
  }
)
df.select(
  Polars.all.reverse.map_alias { |colName| colName + "_reverse" }
)
# =>
# shape: (2, 2)
# ┌───────────┬───────────┐
# │ A_reverse ┆ B_reverse │
# │ ---       ┆ ---       │
# │ i64       ┆ i64       │
# ╞═══════════╪═══════════╡
# │ 2         ┆ 4         │
# │ 1         ┆ 3         │
# └───────────┴───────────┘

Returns:



452
453
454
 
# File 'lib/polars/expr.rb', line 452

def map_alias(&f)
  name.map(&f)
end

#max ⇒ Expr

Get maximum value.

Examples:

df = Polars::DataFrame.new({"a" => [-1.0, Float::NAN, 1.0]})
df.select(Polars.col("a").max)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 1.0 │
# └─────┘

Returns:



1900
1901
1902
 
# File 'lib/polars/expr.rb', line 1900

def max
  wrap_expr(_rbexpr.max)
end

#mean ⇒ Expr

Get mean value.

Examples:

df = Polars::DataFrame.new({"a" => [-1, 0, 1]})
df.select(Polars.col("a").mean)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 0.0 │
# └─────┘

Returns:



2004
2005
2006
 
# File 'lib/polars/expr.rb', line 2004

def mean
  wrap_expr(_rbexpr.mean)
end

#median ⇒ Expr

Get median value using linear interpolation.

Examples:

df = Polars::DataFrame.new({"a" => [-1, 0, 1]})
df.select(Polars.col("a").median)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 0.0 │
# └─────┘

Returns:



2024
2025
2026
 
# File 'lib/polars/expr.rb', line 2024

def median
  wrap_expr(_rbexpr.median)
end

#meta ⇒ MetaExpr

Create an object namespace of all meta related expression methods.

Returns:



5337
5338
5339
 
# File 'lib/polars/expr.rb', line 5337

def meta
  MetaExpr.new(self)
end

#min ⇒ Expr

Get minimum value.

Examples:

df = Polars::DataFrame.new({"a" => [-1.0, Float::NAN, 1.0]})
df.select(Polars.col("a").min)
# =>
# shape: (1, 1)
# ┌──────┐
# │ a    │
# │ ---  │
# │ f64  │
# ╞══════╡
# │ -1.0 │
# └──────┘

Returns:



1920
1921
1922
 
# File 'lib/polars/expr.rb', line 1920

def min
  wrap_expr(_rbexpr.min)
end

#mode ⇒ Expr

Compute the most occurring value(s).

Can return multiple Values.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [1, 1, 2, 3],
    "b" => [1, 1, 2, 2]
  }
)
df.select(Polars.all.mode)
# =>
# shape: (2, 2)
# ┌─────┬─────┐
# │ a   ┆ b   │
# │ --- ┆ --- │
# │ i64 ┆ i64 │
# ╞═════╪═════╡
# │ 1   ┆ 1   │
# │ 1   ┆ 2   │
# └─────┴─────┘

Returns:



1200
1201
1202
 
# File 'lib/polars/expr.rb', line 1200

def mode
  wrap_expr(_rbexpr.mode)
end

#n_unique ⇒ Expr

Count unique values.

Examples:

df = Polars::DataFrame.new({"a" => [1, 1, 2]})
df.select(Polars.col("a").n_unique)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ u32 │
# ╞═════╡
# │ 2   │
# └─────┘

Returns:



2064
2065
2066
 
# File 'lib/polars/expr.rb', line 2064

def n_unique
  wrap_expr(_rbexpr.n_unique)
end

#name ⇒ NameExpr

Create an object namespace of all expressions that modify expression names.

Returns:



5344
5345
5346
 
# File 'lib/polars/expr.rb', line 5344

def name
  NameExpr.new(self)
end

#nan_max ⇒ Expr

Get maximum value, but propagate/poison encountered NaN values.

Examples:

df = Polars::DataFrame.new({"a" => [0.0, Float::NAN]})
df.select(Polars.col("a").nan_max)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ NaN │
# └─────┘

Returns:



1940
1941
1942
 
# File 'lib/polars/expr.rb', line 1940

def nan_max
  wrap_expr(_rbexpr.nan_max)
end

#nan_min ⇒ Expr

Get minimum value, but propagate/poison encountered NaN values.

Examples:

df = Polars::DataFrame.new({"a" => [0.0, Float::NAN]})
df.select(Polars.col("a").nan_min)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ NaN │
# └─────┘

Returns:



1960
1961
1962
 
# File 'lib/polars/expr.rb', line 1960

def nan_min
  wrap_expr(_rbexpr.nan_min)
end

#null_count ⇒ Expr

Count null values.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [nil, 1, nil],
    "b" => [1, 2, 3]
  }
)
df.select(Polars.all.null_count)
# =>
# shape: (1, 2)
# ┌─────┬─────┐
# │ a   ┆ b   │
# │ --- ┆ --- │
# │ u32 ┆ u32 │
# ╞═════╪═════╡
# │ 2   ┆ 0   │
# └─────┴─────┘

Returns:



2111
2112
2113
 
# File 'lib/polars/expr.rb', line 2111

def null_count
  wrap_expr(_rbexpr.null_count)
end

#over(expr) ⇒ Expr

Apply window function over a subgroup.

This is similar to a group by + aggregation + self join. Or similar to window functions in Postgres.

Examples:

df = Polars::DataFrame.new(
  {
    "groups" => ["g1", "g1", "g2"],
    "values" => [1, 2, 3]
  }
)
df.with_column(
  Polars.col("values").max.over("groups").alias("max_by_group")
)
# =>
# shape: (3, 3)
# ┌────────┬────────┬──────────────┐
# │ groups ┆ values ┆ max_by_group │
# │ ---    ┆ ---    ┆ ---          │
# │ str    ┆ i64    ┆ i64          │
# ╞════════╪════════╪══════════════╡
# │ g1     ┆ 1      ┆ 2            │
# │ g1     ┆ 2      ┆ 2            │
# │ g2     ┆ 3      ┆ 3            │
# └────────┴────────┴──────────────┘

df = Polars::DataFrame.new(
  {
    "groups" => [1, 1, 2, 2, 1, 2, 3, 3, 1],
    "values" => [1, 2, 3, 4, 5, 6, 7, 8, 8]
  }
)
df.lazy
  .select([Polars.col("groups").sum.over("groups")])
  .collect
# =>
# shape: (9, 1)
# ┌────────┐
# │ groups │
# │ ---    │
# │ i64    │
# ╞════════╡
# │ 4      │
# │ 4      │
# │ 6      │
# │ 6      │
# │ 4      │
# │ 6      │
# │ 6      │
# │ 6      │
# │ 4      │
# └────────┘

Parameters:

  • expr (Object) —

    Column(s) to group by.

Returns:



2282
2283
2284
2285
 
# File 'lib/polars/expr.rb', line 2282

def over(expr)
  rbexprs = Utils.selection_to_rbexpr_list(expr)
  wrap_expr(_rbexpr.over(rbexprs))
end

#pct_change(n: 1) ⇒ Expr

Computes percentage change between values.

Percentage change (as fraction) between current element and most-recent non-null element at least n period(s) before the current element.

Computes the change from the previous row by default.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [10, 11, 12, nil, 12]
  }
)
df.with_column(Polars.col("a").pct_change.alias("pct_change"))
# =>
# shape: (5, 2)
# ┌──────┬────────────┐
# │ a    ┆ pct_change │
# │ ---  ┆ ---        │
# │ i64  ┆ f64        │
# ╞══════╪════════════╡
# │ 10   ┆ null       │
# │ 11   ┆ 0.1        │
# │ 12   ┆ 0.090909   │
# │ null ┆ 0.0        │
# │ 12   ┆ 0.0        │
# └──────┴────────────┘

Parameters:

  • n (Integer) (defaults to: 1) —

    Periods to shift for forming percent change.

Returns:



4318
4319
4320
4321
 
# File 'lib/polars/expr.rb', line 4318

def pct_change(n: 1)
  n = Utils.parse_as_expression(n)
  wrap_expr(_rbexpr.pct_change(n))
end

#peak_max ⇒ Expr

Get a boolean mask of the local maximum peaks.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3, 4, 5]})
df.select(Polars.col("a").peak_max)
# =>
# shape: (5, 1)
# ┌───────┐
# │ a     │
# │ ---   │
# │ bool  │
# ╞═══════╡
# │ false │
# │ false │
# │ false │
# │ false │
# │ true  │
# └───────┘

Returns:



2380
2381
2382
 
# File 'lib/polars/expr.rb', line 2380

def peak_max
  wrap_expr(_rbexpr.peak_max)
end

#peak_min ⇒ Expr

Get a boolean mask of the local minimum peaks.

Examples:

df = Polars::DataFrame.new({"a" => [4, 1, 3, 2, 5]})
df.select(Polars.col("a").peak_min)
# =>
# shape: (5, 1)
# ┌───────┐
# │ a     │
# │ ---   │
# │ bool  │
# ╞═══════╡
# │ false │
# │ true  │
# │ false │
# │ true  │
# │ false │
# └───────┘

Returns:



2404
2405
2406
 
# File 'lib/polars/expr.rb', line 2404

def peak_min
  wrap_expr(_rbexpr.peak_min)
end

#pow(exponent) ⇒ Expr

Raise expression to the power of exponent.

Examples:

df = Polars::DataFrame.new({"foo" => [1, 2, 3, 4]})
df.select(Polars.col("foo").pow(3))
# =>
# shape: (4, 1)
# ┌──────┐
# │ foo  │
# │ ---  │
# │ f64  │
# ╞══════╡
# │ 1.0  │
# │ 8.0  │
# │ 27.0 │
# │ 64.0 │
# └──────┘

Returns:



3041
3042
3043
3044
 
# File 'lib/polars/expr.rb', line 3041

def pow(exponent)
  exponent = Utils.expr_to_lit_or_expr(exponent)
  wrap_expr(_rbexpr.pow(exponent._rbexpr))
end

#prefix(prefix) ⇒ Expr

Add a prefix to the root column name of the expression.

Returns:



417
418
419
 
# File 'lib/polars/expr.rb', line 417

def prefix(prefix)
  name.prefix(prefix)
end

#product ⇒ Expr

Compute the product of an expression.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3]})
df.select(Polars.col("a").product)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ i64 │
# ╞═════╡
# │ 6   │
# └─────┘

Returns:



2044
2045
2046
 
# File 'lib/polars/expr.rb', line 2044

def product
  wrap_expr(_rbexpr.product)
end

#qcut(quantiles, labels: nil, left_closed: false, allow_duplicates: false, include_breaks: false) ⇒ Expr

Bin continuous values into discrete categories based on their quantiles.

Examples:

Divide a column into three categories according to pre-defined quantile probabilities.

df = Polars::DataFrame.new({"foo" => [-2, -1, 0, 1, 2]})
df.with_columns(
  Polars.col("foo").qcut([0.25, 0.75], labels: ["a", "b", "c"]).alias("qcut")
)
# =>
# shape: (5, 2)
# ┌─────┬──────┐
# │ foo ┆ qcut │
# │ --- ┆ ---  │
# │ i64 ┆ cat  │
# ╞═════╪══════╡
# │ -2  ┆ a    │
# │ -1  ┆ a    │
# │ 0   ┆ b    │
# │ 1   ┆ b    │
# │ 2   ┆ c    │
# └─────┴──────┘

Divide a column into two categories using uniform quantile probabilities.

df.with_columns(
  Polars.col("foo")
    .qcut(2, labels: ["low", "high"], left_closed: true)
    .alias("qcut")
)
# =>
# shape: (5, 2)
# ┌─────┬──────┐
# │ foo ┆ qcut │
# │ --- ┆ ---  │
# │ i64 ┆ cat  │
# ╞═════╪══════╡
# │ -2  ┆ low  │
# │ -1  ┆ low  │
# │ 0   ┆ high │
# │ 1   ┆ high │
# │ 2   ┆ high │
# └─────┴──────┘

Add both the category and the breakpoint.

df.with_columns(
  Polars.col("foo").qcut([0.25, 0.75], include_breaks: true).alias("qcut")
).unnest("qcut")
# =>
# shape: (5, 3)
# ┌─────┬──────┬────────────┐
# │ foo ┆ brk  ┆ foo_bin    │
# │ --- ┆ ---  ┆ ---        │
# │ i64 ┆ f64  ┆ cat        │
# ╞═════╪══════╪════════════╡
# │ -2  ┆ -1.0 ┆ (-inf, -1] │
# │ -1  ┆ -1.0 ┆ (-inf, -1] │
# │ 0   ┆ 1.0  ┆ (-1, 1]    │
# │ 1   ┆ 1.0  ┆ (-1, 1]    │
# │ 2   ┆ inf  ┆ (1, inf]   │
# └─────┴──────┴────────────┘

Parameters:

  • quantiles (Array) —

    Either a list of quantile probabilities between 0 and 1 or a positive integer determining the number of bins with uniform probability.

  • labels (Array) (defaults to: nil) —

    Names of the categories. The number of labels must be equal to the number of categories.

  • left_closed (Boolean) (defaults to: false) —

    Set the intervals to be left-closed instead of right-closed.

  • allow_duplicates (Boolean) (defaults to: false) —

    If set to true, duplicates in the resulting quantiles are dropped, rather than raising a DuplicateError. This can happen even with unique probabilities, depending on the data.

  • include_breaks (Boolean) (defaults to: false) —

    Include a column with the right endpoint of the bin each observation falls in. This will change the data type of the output from a Categorical to a Struct.

Returns:



2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
 
# File 'lib/polars/expr.rb', line 2615

def qcut(quantiles, labels: nil, left_closed: false, allow_duplicates: false, include_breaks: false)
  if quantiles.is_a?(Integer)
    rbexpr = _rbexpr.qcut_uniform(
      quantiles, labels, left_closed, allow_duplicates, include_breaks
    )
  else
    rbexpr = _rbexpr.qcut(
      quantiles, labels, left_closed, allow_duplicates, include_breaks
    )
  end

  wrap_expr(rbexpr)
end

#quantile(quantile, interpolation: "nearest") ⇒ Expr

Get quantile value.

Examples:

df = Polars::DataFrame.new({"a" => [0, 1, 2, 3, 4, 5]})
df.select(Polars.col("a").quantile(0.3))
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 2.0 │
# └─────┘

df.select(Polars.col("a").quantile(0.3, interpolation: "higher"))
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 2.0 │
# └─────┘

df.select(Polars.col("a").quantile(0.3, interpolation: "lower"))
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 1.0 │
# └─────┘

df.select(Polars.col("a").quantile(0.3, interpolation: "midpoint"))
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 1.5 │
# └─────┘

df.select(Polars.col("a").quantile(0.3, interpolation: "linear"))
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 1.5 │
# └─────┘

Parameters:

  • quantile (Float) —

    Quantile between 0.0 and 1.0.

  • interpolation ("nearest", "higher", "lower", "midpoint", "linear") (defaults to: "nearest") —

    Interpolation method.

Returns:



2477
2478
2479
2480
 
# File 'lib/polars/expr.rb', line 2477

def quantile(quantile, interpolation: "nearest")
  quantile = Utils.expr_to_lit_or_expr(quantile, str_to_lit: false)
  wrap_expr(_rbexpr.quantile(quantile._rbexpr, interpolation))
end

#rank(method: "average", reverse: false, seed: nil) ⇒ Expr

Assign ranks to data, dealing with ties appropriately.

Examples:

The 'average' method:

df = Polars::DataFrame.new({"a" => [3, 6, 1, 1, 6]})
df.select(Polars.col("a").rank)
# =>
# shape: (5, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 3.0 │
# │ 4.5 │
# │ 1.5 │
# │ 1.5 │
# │ 4.5 │
# └─────┘

The 'ordinal' method:

df = Polars::DataFrame.new({"a" => [3, 6, 1, 1, 6]})
df.select(Polars.col("a").rank(method: "ordinal"))
# =>
# shape: (5, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ u32 │
# ╞═════╡
# │ 3   │
# │ 4   │
# │ 1   │
# │ 2   │
# │ 5   │
# └─────┘

Parameters:

  • method ("average", "min", "max", "dense", "ordinal", "random") (defaults to: "average") —

    The method used to assign ranks to tied elements. The following methods are available:

    • 'average' : The average of the ranks that would have been assigned to all the tied values is assigned to each value.
    • 'min' : The minimum of the ranks that would have been assigned to all the tied values is assigned to each value. (This is also referred to as "competition" ranking.)
    • 'max' : The maximum of the ranks that would have been assigned to all the tied values is assigned to each value.
    • 'dense' : Like 'min', but the rank of the next highest element is assigned the rank immediately after those assigned to the tied elements.
    • 'ordinal' : All values are given a distinct rank, corresponding to the order that the values occur in the Series.
    • 'random' : Like 'ordinal', but the rank for ties is not dependent on the order that the values occur in the Series.
  • reverse (Boolean) (defaults to: false) —

    Reverse the operation.

Returns:



4251
4252
4253
 
# File 'lib/polars/expr.rb', line 4251

def rank(method: "average", reverse: false, seed: nil)
  wrap_expr(_rbexpr.rank(method, reverse, seed))
end

#rechunk ⇒ Expr

Create a single chunk of memory for this Series.

Examples:

Create a Series with 3 nulls, append column a then rechunk

df = Polars::DataFrame.new({"a" => [1, 1, 2]})
df.select(Polars.repeat(nil, 3).append(Polars.col("a")).rechunk)
# =>
# shape: (6, 1)
# ┌────────┐
# │ repeat │
# │ ---    │
# │ i64    │
# ╞════════╡
# │ null   │
# │ null   │
# │ null   │
# │ 1      │
# │ 1      │
# │ 2      │
# └────────┘

Returns:



843
844
845
 
# File 'lib/polars/expr.rb', line 843

def rechunk
  wrap_expr(_rbexpr.rechunk)
end

#reinterpret(signed: false) ⇒ Expr

Reinterpret the underlying bits as a signed/unsigned integer.

This operation is only allowed for 64bit integers. For lower bits integers, you can safely use that cast operation.

Examples:

s = Polars::Series.new("a", [1, 1, 2], dtype: :u64)
df = Polars::DataFrame.new([s])
df.select(
  [
    Polars.col("a").reinterpret(signed: true).alias("reinterpreted"),
    Polars.col("a").alias("original")
  ]
)
# =>
# shape: (3, 2)
# ┌───────────────┬──────────┐
# │ reinterpreted ┆ original │
# │ ---           ┆ ---      │
# │ i64           ┆ u64      │
# ╞═══════════════╪══════════╡
# │ 1             ┆ 1        │
# │ 1             ┆ 1        │
# │ 2             ┆ 2        │
# └───────────────┴──────────┘

Parameters:

  • signed (Boolean) (defaults to: false) —

    If true, reinterpret as :i64. Otherwise, reinterpret as :u64.

Returns:



3274
3275
3276
 
# File 'lib/polars/expr.rb', line 3274

def reinterpret(signed: false)
  wrap_expr(_rbexpr.reinterpret(signed))
end

#repeat_by(by) ⇒ Expr

Repeat the elements in this Series as specified in the given expression.

The repeated elements are expanded into a List.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => ["x", "y", "z"],
    "n" => [1, 2, 3]
  }
)
df.select(Polars.col("a").repeat_by("n"))
# =>
# shape: (3, 1)
# ┌─────────────────┐
# │ a               │
# │ ---             │
# │ list[str]       │
# ╞═════════════════╡
# │ ["x"]           │
# │ ["y", "y"]      │
# │ ["z", "z", "z"] │
# └─────────────────┘

Parameters:

  • by (Object) —

    Numeric column that determines how often the values will be repeated. The column will be coerced to UInt32. Give this dtype to make the coercion a no-op.

Returns:



3113
3114
3115
3116
 
# File 'lib/polars/expr.rb', line 3113

def repeat_by(by)
  by = Utils.expr_to_lit_or_expr(by, str_to_lit: false)
  wrap_expr(_rbexpr.repeat_by(by._rbexpr))
end

#reshape(dims) ⇒ Expr

Reshape this Expr to a flat Series or a Series of Lists.

Examples:

df = Polars::DataFrame.new({"foo" => [1, 2, 3, 4, 5, 6, 7, 8, 9]})
df.select(Polars.col("foo").reshape([3, 3]))
# =>
# shape: (3, 1)
# ┌───────────┐
# │ foo       │
# │ ---       │
# │ list[i64] │
# ╞═══════════╡
# │ [1, 2, 3] │
# │ [4, 5, 6] │
# │ [7, 8, 9] │
# └───────────┘

Parameters:

  • dims (Array) —

    Tuple of the dimension sizes. If a -1 is used in any of the dimensions, that dimension is inferred.

Returns:



4815
4816
4817
 
# File 'lib/polars/expr.rb', line 4815

def reshape(dims)
  wrap_expr(_rbexpr.reshape(dims))
end

#reverse ⇒ Expr

Reverse the selection.

Returns:



1834
1835
1836
 
# File 'lib/polars/expr.rb', line 1834

def reverse
  wrap_expr(_rbexpr.reverse)
end

#rle ⇒ Expr

Get the lengths of runs of identical values.

Examples:

df = Polars::DataFrame.new(Polars::Series.new("s", [1, 1, 2, 1, nil, 1, 3, 3]))
df.select(Polars.col("s").rle).unnest("s")
# =>
# shape: (6, 2)
# ┌─────────┬────────┐
# │ lengths ┆ values │
# │ ---     ┆ ---    │
# │ i32     ┆ i64    │
# ╞═════════╪════════╡
# │ 2       ┆ 1      │
# │ 1       ┆ 2      │
# │ 1       ┆ 1      │
# │ 1       ┆ null   │
# │ 1       ┆ 1      │
# │ 2       ┆ 3      │
# └─────────┴────────┘

Returns:



2650
2651
2652
 
# File 'lib/polars/expr.rb', line 2650

def rle
  wrap_expr(_rbexpr.rle)
end

#rle_id ⇒ Expr

Map values to run IDs.

Similar to RLE, but it maps each value to an ID corresponding to the run into which it falls. This is especially useful when you want to define groups by runs of identical values rather than the values themselves.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 1, 1, 1], "b" => ["x", "x", nil, "y", "y"]})
df.with_columns([Polars.col("a").rle_id.alias("a_r"), Polars.struct(["a", "b"]).rle_id.alias("ab_r")])
# =>
# shape: (5, 4)
# ┌─────┬──────┬─────┬──────┐
# │ a   ┆ b    ┆ a_r ┆ ab_r │
# │ --- ┆ ---  ┆ --- ┆ ---  │
# │ i64 ┆ str  ┆ u32 ┆ u32  │
# ╞═════╪══════╪═════╪══════╡
# │ 1   ┆ x    ┆ 0   ┆ 0    │
# │ 2   ┆ x    ┆ 1   ┆ 1    │
# │ 1   ┆ null ┆ 2   ┆ 2    │
# │ 1   ┆ y    ┆ 2   ┆ 3    │
# │ 1   ┆ y    ┆ 2   ┆ 3    │
# └─────┴──────┴─────┴──────┘

Returns:



2678
2679
2680
 
# File 'lib/polars/expr.rb', line 2678

def rle_id
  wrap_expr(_rbexpr.rle_id)
end

#rolling_max(window_size, weights: nil, min_periods: nil, center: false, by: nil, closed: "left") ⇒ Expr

Note:

This functionality is experimental and may change without it being considered a breaking change.

Note:

If you want to compute multiple aggregation statistics over the same dynamic window, consider using group_by_rolling this method can cache the window size computation.

Apply a rolling max (moving max) over the values in this array.

A window of length window_size will traverse the array. The values that fill this window will (optionally) be multiplied with the weights given by the weight vector. The resulting values will be aggregated to their sum.

Examples:

df = Polars::DataFrame.new({"A" => [1.0, 2.0, 3.0, 4.0, 5.0, 6.0]})
df.select(
  [
    Polars.col("A").rolling_max(2)
  ]
)
# =>
# shape: (6, 1)
# ┌──────┐
# │ A    │
# │ ---  │
# │ f64  │
# ╞══════╡
# │ null │
# │ 2.0  │
# │ 3.0  │
# │ 4.0  │
# │ 5.0  │
# │ 6.0  │
# └──────┘

Parameters:

  • window_size (Integer) —

    The length of the window. Can be a fixed integer size, or a dynamic temporal size indicated by a timedelta or the following string language:

    • 1ns (1 nanosecond)
    • 1us (1 microsecond)
    • 1ms (1 millisecond)
    • 1s (1 second)
    • 1m (1 minute)
    • 1h (1 hour)
    • 1d (1 day)
    • 1w (1 week)
    • 1mo (1 calendar month)
    • 1y (1 calendar year)
    • 1i (1 index count)

    If a timedelta or the dynamic string language is used, the by and closed arguments must also be set.

  • weights (Array) (defaults to: nil) —

    An optional slice with the same length as the window that will be multiplied elementwise with the values in the window.

  • min_periods (Integer) (defaults to: nil) —

    The number of values in the window that should be non-null before computing a result. If None, it will be set equal to window size.

  • center (Boolean) (defaults to: false) —

    Set the labels at the center of the window

  • by (String) (defaults to: nil) —

    If the window_size is temporal for instance "5h" or "3s, you must set the column that will be used to determine the windows. This column must be of dtype {Date, Datetime}

  • closed ("left", "right", "both", "none") (defaults to: "left") —

    Define whether the temporal window interval is closed or not.

Returns:



3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
 
# File 'lib/polars/expr.rb', line 3501

def rolling_max(
  window_size,
  weights: nil,
  min_periods: nil,
  center: false,
  by: nil,
  closed: "left"
)
  window_size, min_periods = _prepare_rolling_window_args(
    window_size, min_periods
  )
  wrap_expr(
    _rbexpr.rolling_max(
      window_size, weights, min_periods, center, by, closed
    )
  )
end

#rolling_mean(window_size, weights: nil, min_periods: nil, center: false, by: nil, closed: "left") ⇒ Expr

Note:

This functionality is experimental and may change without it being considered a breaking change.

Note:

If you want to compute multiple aggregation statistics over the same dynamic window, consider using group_by_rolling this method can cache the window size computation.

Apply a rolling mean (moving mean) over the values in this array.

A window of length window_size will traverse the array. The values that fill this window will (optionally) be multiplied with the weights given by the weight vector. The resulting values will be aggregated to their sum.

Examples:

df = Polars::DataFrame.new({"A" => [1.0, 8.0, 6.0, 2.0, 16.0, 10.0]})
df.select(
  [
    Polars.col("A").rolling_mean(2)
  ]
)
# =>
# shape: (6, 1)
# ┌──────┐
# │ A    │
# │ ---  │
# │ f64  │
# ╞══════╡
# │ null │
# │ 4.5  │
# │ 7.0  │
# │ 4.0  │
# │ 9.0  │
# │ 13.0 │
# └──────┘

Parameters:

  • window_size (Integer) —

    The length of the window. Can be a fixed integer size, or a dynamic temporal size indicated by a timedelta or the following string language:

    • 1ns (1 nanosecond)
    • 1us (1 microsecond)
    • 1ms (1 millisecond)
    • 1s (1 second)
    • 1m (1 minute)
    • 1h (1 hour)
    • 1d (1 day)
    • 1w (1 week)
    • 1mo (1 calendar month)
    • 1y (1 calendar year)
    • 1i (1 index count)

    If a timedelta or the dynamic string language is used, the by and closed arguments must also be set.

  • weights (Array) (defaults to: nil) —

    An optional slice with the same length as the window that will be multiplied elementwise with the values in the window.

  • min_periods (Integer) (defaults to: nil) —

    The number of values in the window that should be non-null before computing a result. If None, it will be set equal to window size.

  • center (Boolean) (defaults to: false) —

    Set the labels at the center of the window

  • by (String) (defaults to: nil) —

    If the window_size is temporal for instance "5h" or "3s, you must set the column that will be used to determine the windows. This column must be of dtype {Date, Datetime}

  • closed ("left", "right", "both", "none") (defaults to: "left") —

    Define whether the temporal window interval is closed or not.

Returns:



3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
 
# File 'lib/polars/expr.rb', line 3590

def rolling_mean(
  window_size,
  weights: nil,
  min_periods: nil,
  center: false,
  by: nil,
  closed: "left"
)
  window_size, min_periods = _prepare_rolling_window_args(
    window_size, min_periods
  )
  wrap_expr(
    _rbexpr.rolling_mean(
      window_size, weights, min_periods, center, by, closed
    )
  )
end

#rolling_median(window_size, weights: nil, min_periods: nil, center: false, by: nil, closed: "left", warn_if_unsorted: true) ⇒ Expr

Note:

This functionality is experimental and may change without it being considered a breaking change.

Note:

If you want to compute multiple aggregation statistics over the same dynamic window, consider using group_by_rolling this method can cache the window size computation.

Compute a rolling median.

Examples:

df = Polars::DataFrame.new({"A" => [1.0, 2.0, 3.0, 4.0, 6.0, 8.0]})
df.select(
  [
    Polars.col("A").rolling_median(3)
  ]
)
# =>
# shape: (6, 1)
# ┌──────┐
# │ A    │
# │ ---  │
# │ f64  │
# ╞══════╡
# │ null │
# │ null │
# │ 2.0  │
# │ 3.0  │
# │ 4.0  │
# │ 6.0  │
# └──────┘

Parameters:

  • window_size (Integer) —

    The length of the window. Can be a fixed integer size, or a dynamic temporal size indicated by a timedelta or the following string language:

    • 1ns (1 nanosecond)
    • 1us (1 microsecond)
    • 1ms (1 millisecond)
    • 1s (1 second)
    • 1m (1 minute)
    • 1h (1 hour)
    • 1d (1 day)
    • 1w (1 week)
    • 1mo (1 calendar month)
    • 1y (1 calendar year)
    • 1i (1 index count)

    If a timedelta or the dynamic string language is used, the by and closed arguments must also be set.

  • weights (Array) (defaults to: nil) —

    An optional slice with the same length as the window that will be multiplied elementwise with the values in the window.

  • min_periods (Integer) (defaults to: nil) —

    The number of values in the window that should be non-null before computing a result. If None, it will be set equal to window size.

  • center (Boolean) (defaults to: false) —

    Set the labels at the center of the window

  • by (String) (defaults to: nil) —

    If the window_size is temporal for instance "5h" or "3s, you must set the column that will be used to determine the windows. This column must be of dtype {Date, Datetime}

  • closed ("left", "right", "both", "none") (defaults to: "left") —

    Define whether the temporal window interval is closed or not.

Returns:



3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
 
# File 'lib/polars/expr.rb', line 3946

def rolling_median(
  window_size,
  weights: nil,
  min_periods: nil,
  center: false,
  by: nil,
  closed: "left",
  warn_if_unsorted: true
)
  window_size, min_periods = _prepare_rolling_window_args(
    window_size, min_periods
  )
  wrap_expr(
    _rbexpr.rolling_median(
      window_size, weights, min_periods, center, by, closed, warn_if_unsorted
    )
  )
end

#rolling_min(window_size, weights: nil, min_periods: nil, center: false, by: nil, closed: "left") ⇒ Expr

Note:

This functionality is experimental and may change without it being considered a breaking change.

Note:

If you want to compute multiple aggregation statistics over the same dynamic window, consider using group_by_rolling this method can cache the window size computation.

Apply a rolling min (moving min) over the values in this array.

A window of length window_size will traverse the array. The values that fill this window will (optionally) be multiplied with the weights given by the weight vector. The resulting values will be aggregated to their sum.

Examples:

df = Polars::DataFrame.new({"A" => [1.0, 2.0, 3.0, 4.0, 5.0, 6.0]})
df.select(
  [
    Polars.col("A").rolling_min(2)
  ]
)
# =>
# shape: (6, 1)
# ┌──────┐
# │ A    │
# │ ---  │
# │ f64  │
# ╞══════╡
# │ null │
# │ 1.0  │
# │ 2.0  │
# │ 3.0  │
# │ 4.0  │
# │ 5.0  │
# └──────┘

Parameters:

  • window_size (Integer) —

    The length of the window. Can be a fixed integer size, or a dynamic temporal size indicated by a timedelta or the following string language:

    • 1ns (1 nanosecond)
    • 1us (1 microsecond)
    • 1ms (1 millisecond)
    • 1s (1 second)
    • 1m (1 minute)
    • 1h (1 hour)
    • 1d (1 day)
    • 1w (1 week)
    • 1mo (1 calendar month)
    • 1y (1 calendar year)
    • 1i (1 index count)

    If a timedelta or the dynamic string language is used, the by and closed arguments must also be set.

  • weights (Array) (defaults to: nil) —

    An optional slice with the same length as the window that will be multiplied elementwise with the values in the window.

  • min_periods (Integer) (defaults to: nil) —

    The number of values in the window that should be non-null before computing a result. If None, it will be set equal to window size.

  • center (Boolean) (defaults to: false) —

    Set the labels at the center of the window

  • by (String) (defaults to: nil) —

    If the window_size is temporal for instance "5h" or "3s, you must set the column that will be used to determine the windows. This column must be of dtype {Date, Datetime}

  • closed ("left", "right", "both", "none") (defaults to: "left") —

    Define whether the temporal window interval is closed or not.

Returns:



3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
 
# File 'lib/polars/expr.rb', line 3412

def rolling_min(
  window_size,
  weights: nil,
  min_periods: nil,
  center: false,
  by: nil,
  closed: "left"
)
  window_size, min_periods = _prepare_rolling_window_args(
    window_size, min_periods
  )
  wrap_expr(
    _rbexpr.rolling_min(
      window_size, weights, min_periods, center, by, closed
    )
  )
end

#rolling_quantile(quantile, interpolation: "nearest", window_size: 2, weights: nil, min_periods: nil, center: false, by: nil, closed: "left", warn_if_unsorted: true) ⇒ Expr

Note:

This functionality is experimental and may change without it being considered a breaking change.

Note:

If you want to compute multiple aggregation statistics over the same dynamic window, consider using group_by_rolling this method can cache the window size computation.

Compute a rolling quantile.

Examples:

df = Polars::DataFrame.new({"A" => [1.0, 2.0, 3.0, 4.0, 6.0, 8.0]})
df.select(
  [
    Polars.col("A").rolling_quantile(0.33, window_size: 3)
  ]
)
# =>
# shape: (6, 1)
# ┌──────┐
# │ A    │
# │ ---  │
# │ f64  │
# ╞══════╡
# │ null │
# │ null │
# │ 1.0  │
# │ 2.0  │
# │ 3.0  │
# │ 4.0  │
# └──────┘

Parameters:

  • quantile (Float) —

    Quantile between 0.0 and 1.0.

  • interpolation ("nearest", "higher", "lower", "midpoint", "linear") (defaults to: "nearest") —

    Interpolation method.

  • window_size (Integer) (defaults to: 2) —

    The length of the window. Can be a fixed integer size, or a dynamic temporal size indicated by a timedelta or the following string language:

    • 1ns (1 nanosecond)
    • 1us (1 microsecond)
    • 1ms (1 millisecond)
    • 1s (1 second)
    • 1m (1 minute)
    • 1h (1 hour)
    • 1d (1 day)
    • 1w (1 week)
    • 1mo (1 calendar month)
    • 1y (1 calendar year)
    • 1i (1 index count)

    If a timedelta or the dynamic string language is used, the by and closed arguments must also be set.

  • weights (Array) (defaults to: nil) —

    An optional slice with the same length as the window that will be multiplied elementwise with the values in the window.

  • min_periods (Integer) (defaults to: nil) —

    The number of values in the window that should be non-null before computing a result. If None, it will be set equal to window size.

  • center (Boolean) (defaults to: false) —

    Set the labels at the center of the window

  • by (String) (defaults to: nil) —

    If the window_size is temporal for instance "5h" or "3s, you must set the column that will be used to determine the windows. This column must be of dtype {Date, Datetime}

  • closed ("left", "right", "both", "none") (defaults to: "left") —

    Define whether the temporal window interval is closed or not.

Returns:



4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
 
# File 'lib/polars/expr.rb', line 4036

def rolling_quantile(
  quantile,
  interpolation: "nearest",
  window_size: 2,
  weights: nil,
  min_periods: nil,
  center: false,
  by: nil,
  closed: "left",
  warn_if_unsorted: true
)
  window_size, min_periods = _prepare_rolling_window_args(
    window_size, min_periods
  )
  wrap_expr(
    _rbexpr.rolling_quantile(
      quantile, interpolation, window_size, weights, min_periods, center, by, closed, warn_if_unsorted
    )
  )
end

#rolling_skew(window_size, bias: true) ⇒ Expr

Compute a rolling skew.

Parameters:

  • window_size (Integer) —

    Integer size of the rolling window.

  • bias (Boolean) (defaults to: true) —

    If false, the calculations are corrected for statistical bias.

Returns:



4129
4130
4131
 
# File 'lib/polars/expr.rb', line 4129

def rolling_skew(window_size, bias: true)
  wrap_expr(_rbexpr.rolling_skew(window_size, bias))
end

#rolling_std(window_size, weights: nil, min_periods: nil, center: false, by: nil, closed: "left", ddof: 1, warn_if_unsorted: true) ⇒ Expr

Note:

This functionality is experimental and may change without it being considered a breaking change.

Note:

If you want to compute multiple aggregation statistics over the same dynamic window, consider using group_by_rolling this method can cache the window size computation.

Compute a rolling standard deviation.

A window of length window_size will traverse the array. The values that fill this window will (optionally) be multiplied with the weights given by the weight vector. The resulting values will be aggregated to their sum.

Examples:

df = Polars::DataFrame.new({"A" => [1.0, 2.0, 3.0, 4.0, 6.0, 8.0]})
df.select(
  [
    Polars.col("A").rolling_std(3)
  ]
)
# =>
# shape: (6, 1)
# ┌──────────┐
# │ A        │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ null     │
# │ null     │
# │ 1.0      │
# │ 1.0      │
# │ 1.527525 │
# │ 2.0      │
# └──────────┘

Parameters:

  • window_size (Integer) —

    The length of the window. Can be a fixed integer size, or a dynamic temporal size indicated by a timedelta or the following string language:

    • 1ns (1 nanosecond)
    • 1us (1 microsecond)
    • 1ms (1 millisecond)
    • 1s (1 second)
    • 1m (1 minute)
    • 1h (1 hour)
    • 1d (1 day)
    • 1w (1 week)
    • 1mo (1 calendar month)
    • 1y (1 calendar year)
    • 1i (1 index count)

    If a timedelta or the dynamic string language is used, the by and closed arguments must also be set.

  • weights (Array) (defaults to: nil) —

    An optional slice with the same length as the window that will be multiplied elementwise with the values in the window.

  • min_periods (Integer) (defaults to: nil) —

    The number of values in the window that should be non-null before computing a result. If None, it will be set equal to window size.

  • center (Boolean) (defaults to: false) —

    Set the labels at the center of the window

  • by (String) (defaults to: nil) —

    If the window_size is temporal for instance "5h" or "3s, you must set the column that will be used to determine the windows. This column must be of dtype {Date, Datetime}

  • closed ("left", "right", "both", "none") (defaults to: "left") —

    Define whether the temporal window interval is closed or not.

Returns:



3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
 
# File 'lib/polars/expr.rb', line 3768

def rolling_std(
  window_size,
  weights: nil,
  min_periods: nil,
  center: false,
  by: nil,
  closed: "left",
  ddof: 1,
  warn_if_unsorted: true
)
  window_size, min_periods = _prepare_rolling_window_args(
    window_size, min_periods
  )
  wrap_expr(
    _rbexpr.rolling_std(
      window_size, weights, min_periods, center, by, closed, ddof, warn_if_unsorted
    )
  )
end

#rolling_sum(window_size, weights: nil, min_periods: nil, center: false, by: nil, closed: "left") ⇒ Expr

Note:

This functionality is experimental and may change without it being considered a breaking change.

Note:

If you want to compute multiple aggregation statistics over the same dynamic window, consider using group_by_rolling this method can cache the window size computation.

Apply a rolling sum (moving sum) over the values in this array.

A window of length window_size will traverse the array. The values that fill this window will (optionally) be multiplied with the weights given by the weight vector. The resulting values will be aggregated to their sum.

Examples:

df = Polars::DataFrame.new({"A" => [1.0, 2.0, 3.0, 4.0, 5.0, 6.0]})
df.select(
  [
    Polars.col("A").rolling_sum(2)
  ]
)
# =>
# shape: (6, 1)
# ┌──────┐
# │ A    │
# │ ---  │
# │ f64  │
# ╞══════╡
# │ null │
# │ 3.0  │
# │ 5.0  │
# │ 7.0  │
# │ 9.0  │
# │ 11.0 │
# └──────┘

Parameters:

  • window_size (Integer) —

    The length of the window. Can be a fixed integer size, or a dynamic temporal size indicated by a timedelta or the following string language:

    • 1ns (1 nanosecond)
    • 1us (1 microsecond)
    • 1ms (1 millisecond)
    • 1s (1 second)
    • 1m (1 minute)
    • 1h (1 hour)
    • 1d (1 day)
    • 1w (1 week)
    • 1mo (1 calendar month)
    • 1y (1 calendar year)
    • 1i (1 index count)

    If a timedelta or the dynamic string language is used, the by and closed arguments must also be set.

  • weights (Array) (defaults to: nil) —

    An optional slice with the same length as the window that will be multiplied elementwise with the values in the window.

  • min_periods (Integer) (defaults to: nil) —

    The number of values in the window that should be non-null before computing a result. If None, it will be set equal to window size.

  • center (Boolean) (defaults to: false) —

    Set the labels at the center of the window

  • by (String) (defaults to: nil) —

    If the window_size is temporal for instance "5h" or "3s, you must set the column that will be used to determine the windows. This column must be of dtype {Date, Datetime}

  • closed ("left", "right", "both", "none") (defaults to: "left") —

    Define whether the temporal window interval is closed or not.

Returns:



3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
 
# File 'lib/polars/expr.rb', line 3679

def rolling_sum(
  window_size,
  weights: nil,
  min_periods: nil,
  center: false,
  by: nil,
  closed: "left"
)
  window_size, min_periods = _prepare_rolling_window_args(
    window_size, min_periods
  )
  wrap_expr(
    _rbexpr.rolling_sum(
      window_size, weights, min_periods, center, by, closed
    )
  )
end

#rolling_var(window_size, weights: nil, min_periods: nil, center: false, by: nil, closed: "left", ddof: 1, warn_if_unsorted: true) ⇒ Expr

Note:

This functionality is experimental and may change without it being considered a breaking change.

Note:

If you want to compute multiple aggregation statistics over the same dynamic window, consider using group_by_rolling this method can cache the window size computation.

Compute a rolling variance.

A window of length window_size will traverse the array. The values that fill this window will (optionally) be multiplied with the weights given by the weight vector. The resulting values will be aggregated to their sum.

Examples:

df = Polars::DataFrame.new({"A" => [1.0, 2.0, 3.0, 4.0, 6.0, 8.0]})
df.select(
  [
    Polars.col("A").rolling_var(3)
  ]
)
# =>
# shape: (6, 1)
# ┌──────────┐
# │ A        │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ null     │
# │ null     │
# │ 1.0      │
# │ 1.0      │
# │ 2.333333 │
# │ 4.0      │
# └──────────┘

Parameters:

  • window_size (Integer) —

    The length of the window. Can be a fixed integer size, or a dynamic temporal size indicated by a timedelta or the following string language:

    • 1ns (1 nanosecond)
    • 1us (1 microsecond)
    • 1ms (1 millisecond)
    • 1s (1 second)
    • 1m (1 minute)
    • 1h (1 hour)
    • 1d (1 day)
    • 1w (1 week)
    • 1mo (1 calendar month)
    • 1y (1 calendar year)
    • 1i (1 index count)

    If a timedelta or the dynamic string language is used, the by and closed arguments must also be set.

  • weights (Array) (defaults to: nil) —

    An optional slice with the same length as the window that will be multiplied elementwise with the values in the window.

  • min_periods (Integer) (defaults to: nil) —

    The number of values in the window that should be non-null before computing a result. If None, it will be set equal to window size.

  • center (Boolean) (defaults to: false) —

    Set the labels at the center of the window

  • by (String) (defaults to: nil) —

    If the window_size is temporal for instance "5h" or "3s, you must set the column that will be used to determine the windows. This column must be of dtype {Date, Datetime}

  • closed ("left", "right", "both", "none") (defaults to: "left") —

    Define whether the temporal window interval is closed or not.

Returns:



3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
 
# File 'lib/polars/expr.rb', line 3859

def rolling_var(
  window_size,
  weights: nil,
  min_periods: nil,
  center: false,
  by: nil,
  closed: "left",
  ddof: 1,
  warn_if_unsorted: true
)
  window_size, min_periods = _prepare_rolling_window_args(
    window_size, min_periods
  )
  wrap_expr(
    _rbexpr.rolling_var(
      window_size, weights, min_periods, center, by, closed, ddof, warn_if_unsorted
    )
  )
end

#round(decimals = 0) ⇒ Expr

Round underlying floating point data by decimals digits.

Examples:

df = Polars::DataFrame.new({"a" => [0.33, 0.52, 1.02, 1.17]})
df.select(Polars.col("a").round(1))
# =>
# shape: (4, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 0.3 │
# │ 0.5 │
# │ 1.0 │
# │ 1.2 │
# └─────┘

Parameters:

  • decimals (Integer) (defaults to: 0) —

    Number of decimals to round by.

Returns:



1143
1144
1145
 
# File 'lib/polars/expr.rb', line 1143

def round(decimals = 0)
  wrap_expr(_rbexpr.round(decimals))
end

#sample(frac: nil, with_replacement: true, shuffle: false, seed: nil, n: nil) ⇒ Expr

Sample from this expression.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3]})
df.select(Polars.col("a").sample(frac: 1.0, with_replacement: true, seed: 1))
# =>
# shape: (3, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ i64 │
# ╞═════╡
# │ 3   │
# │ 1   │
# │ 1   │
# └─────┘

Parameters:

  • frac (Float) (defaults to: nil) —

    Fraction of items to return. Cannot be used with n.

  • with_replacement (Boolean) (defaults to: true) —

    Allow values to be sampled more than once.

  • shuffle (Boolean) (defaults to: false) —

    Shuffle the order of sampled data points.

  • seed (Integer) (defaults to: nil) —

    Seed for the random number generator. If set to None (default), a random seed is used.

  • n (Integer) (defaults to: nil) —

    Number of items to return. Cannot be used with frac.

Returns:



4878
4879
4880
4881
4882
4883
4884
4885
4886
4887
4888
4889
4890
4891
4892
4893
4894
4895
4896
4897
4898
4899
4900
4901
 
# File 'lib/polars/expr.rb', line 4878

def sample(
  frac: nil,
  with_replacement: true,
  shuffle: false,
  seed: nil,
  n: nil
)
  if !n.nil? && !frac.nil?
    raise ArgumentError, "cannot specify both `n` and `frac`"
  end

  if !n.nil? && frac.nil?
    n = Utils.parse_as_expression(n)
    return wrap_expr(_rbexpr.sample_n(n, with_replacement, shuffle, seed))
  end

  if frac.nil?
    frac = 1.0
  end
  frac = Utils.parse_as_expression(frac)
  wrap_expr(
    _rbexpr.sample_frac(frac, with_replacement, shuffle, seed)
  )
end

#search_sorted(element, side: "any") ⇒ Expr

Find indices where elements should be inserted to maintain order.

Examples:

df = Polars::DataFrame.new(
  {
    "values" => [1, 2, 3, 5]
  }
)
df.select(
  [
    Polars.col("values").search_sorted(0).alias("zero"),
    Polars.col("values").search_sorted(3).alias("three"),
    Polars.col("values").search_sorted(6).alias("six")
  ]
)
# =>
# shape: (1, 3)
# ┌──────┬───────┬─────┐
# │ zero ┆ three ┆ six │
# │ ---  ┆ ---   ┆ --- │
# │ u32  ┆ u32   ┆ u32 │
# ╞══════╪═══════╪═════╡
# │ 0    ┆ 2     ┆ 4   │
# └──────┴───────┴─────┘

Parameters:

  • element (Object) —

    Expression or scalar value.

Returns:



1504
1505
1506
1507
 
# File 'lib/polars/expr.rb', line 1504

def search_sorted(element, side: "any")
  element = Utils.expr_to_lit_or_expr(element, str_to_lit: false)
  wrap_expr(_rbexpr.search_sorted(element._rbexpr, side))
end

#set_sorted(descending: false) ⇒ Expr

Note:

This can lead to incorrect results if this Series is not sorted!! Use with care!

Flags the expression as 'sorted'.

Enables downstream code to user fast paths for sorted arrays.

Examples:

df = Polars::DataFrame.new({"values" => [1, 2, 3]})
df.select(Polars.col("values").set_sorted.max)
# =>
# shape: (1, 1)
# ┌────────┐
# │ values │
# │ ---    │
# │ i64    │
# ╞════════╡
# │ 3      │
# └────────┘

Parameters:

  • descending (Boolean) (defaults to: false) —

    Whether the Series order is descending.

Returns:



5235
5236
5237
 
# File 'lib/polars/expr.rb', line 5235

def set_sorted(descending: false)
  wrap_expr(_rbexpr.set_sorted_flag(descending))
end

#shift(n = 1, fill_value: nil) ⇒ Expr

Shift the values by a given period.

Examples:

df = Polars::DataFrame.new({"foo" => [1, 2, 3, 4]})
df.select(Polars.col("foo").shift(1))
# =>
# shape: (4, 1)
# ┌──────┐
# │ foo  │
# │ ---  │
# │ i64  │
# ╞══════╡
# │ null │
# │ 1    │
# │ 2    │
# │ 3    │
# └──────┘

Parameters:

  • n (Integer) (defaults to: 1) —

    Number of places to shift (may be negative).

  • fill_value (Object) (defaults to: nil) —

    Fill the resulting null values with this value.

Returns:



1629
1630
1631
1632
1633
1634
1635
 
# File 'lib/polars/expr.rb', line 1629

def shift(n = 1, fill_value: nil)
  if !fill_value.nil?
    fill_value = Utils.parse_as_expression(fill_value, str_as_lit: true)
  end
  n = Utils.parse_as_expression(n)
  wrap_expr(_rbexpr.shift(n, fill_value))
end

#shift_and_fill(periods, fill_value) ⇒ Expr

Shift the values by a given period and fill the resulting null values.

Examples:

df = Polars::DataFrame.new({"foo" => [1, 2, 3, 4]})
df.select(Polars.col("foo").shift_and_fill(1, "a"))
# =>
# shape: (4, 1)
# ┌─────┐
# │ foo │
# │ --- │
# │ str │
# ╞═════╡
# │ a   │
# │ 1   │
# │ 2   │
# │ 3   │
# └─────┘

Parameters:

  • periods (Integer) —

    Number of places to shift (may be negative).

  • fill_value (Object) —

    Fill nil values with the result of this expression.

Returns:



1661
1662
1663
 
# File 'lib/polars/expr.rb', line 1661

def shift_and_fill(periods, fill_value)
  shift(periods, fill_value: fill_value)
end

#shrink_dtype ⇒ Expr

Shrink numeric columns to the minimal required datatype.

Shrink to the dtype needed to fit the extrema of this Series. This can be used to reduce memory pressure.

Examples:

Polars::DataFrame.new(
  {
    "a" => [1, 2, 3],
    "b" => [1, 2, 2 << 32],
    "c" => [-1, 2, 1 << 30],
    "d" => [-112, 2, 112],
    "e" => [-112, 2, 129],
    "f" => ["a", "b", "c"],
    "g" => [0.1, 1.32, 0.12],
    "h" => [true, nil, false]
  }
).select(Polars.all.shrink_dtype)
# =>
# shape: (3, 8)
# ┌─────┬────────────┬────────────┬──────┬──────┬─────┬──────┬───────┐
# │ a   ┆ b          ┆ c          ┆ d    ┆ e    ┆ f   ┆ g    ┆ h     │
# │ --- ┆ ---        ┆ ---        ┆ ---  ┆ ---  ┆ --- ┆ ---  ┆ ---   │
# │ i8  ┆ i64        ┆ i32        ┆ i8   ┆ i16  ┆ str ┆ f32  ┆ bool  │
# ╞═════╪════════════╪════════════╪══════╪══════╪═════╪══════╪═══════╡
# │ 1   ┆ 1          ┆ -1         ┆ -112 ┆ -112 ┆ a   ┆ 0.1  ┆ true  │
# │ 2   ┆ 2          ┆ 2          ┆ 2    ┆ 2    ┆ b   ┆ 1.32 ┆ null  │
# │ 3   ┆ 8589934592 ┆ 1073741824 ┆ 112  ┆ 129  ┆ c   ┆ 0.12 ┆ false │
# └─────┴────────────┴────────────┴──────┴──────┴─────┴──────┴───────┘

Returns:



5295
5296
5297
 
# File 'lib/polars/expr.rb', line 5295

def shrink_dtype
  wrap_expr(_rbexpr.shrink_dtype)
end

#shuffle(seed: nil) ⇒ Expr

Shuffle the contents of this expr.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3]})
df.select(Polars.col("a").shuffle(seed: 1))
# =>
# shape: (3, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ i64 │
# ╞═════╡
# │ 2   │
# │ 1   │
# │ 3   │
# └─────┘

Parameters:

  • seed (Integer) (defaults to: nil) —

    Seed for the random number generator. If set to None (default), a random seed is generated using the random module.

Returns:



4841
4842
4843
4844
4845
4846
 
# File 'lib/polars/expr.rb', line 4841

def shuffle(seed: nil)
  if seed.nil?
    seed = rand(10000)
  end
  wrap_expr(_rbexpr.shuffle(seed))
end

#sign ⇒ Expr

Compute the element-wise indication of the sign.

Examples:

df = Polars::DataFrame.new({"a" => [-9.0, -0.0, 0.0, 4.0, nil]})
df.select(Polars.col("a").sign)
# =>
# shape: (5, 1)
# ┌──────┐
# │ a    │
# │ ---  │
# │ i64  │
# ╞══════╡
# │ -1   │
# │ 0    │
# │ 0    │
# │ 1    │
# │ null │
# └──────┘

Returns:



4549
4550
4551
 
# File 'lib/polars/expr.rb', line 4549

def sign
  wrap_expr(_rbexpr.sign)
end

#sin ⇒ Expr

Compute the element-wise value for the sine.

Examples:

df = Polars::DataFrame.new({"a" => [0.0]})
df.select(Polars.col("a").sin)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 0.0 │
# └─────┘

Returns:



4569
4570
4571
 
# File 'lib/polars/expr.rb', line 4569

def sin
  wrap_expr(_rbexpr.sin)
end

#sinh ⇒ Expr

Compute the element-wise value for the hyperbolic sine.

Examples:

df = Polars::DataFrame.new({"a" => [1.0]})
df.select(Polars.col("a").sinh)
# =>
# shape: (1, 1)
# ┌──────────┐
# │ a        │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ 1.175201 │
# └──────────┘

Returns:



4689
4690
4691
 
# File 'lib/polars/expr.rb', line 4689

def sinh
  wrap_expr(_rbexpr.sinh)
end

#skew(bias: true) ⇒ Expr

Compute the sample skewness of a data set.

For normally distributed data, the skewness should be about zero. For unimodal continuous distributions, a skewness value greater than zero means that there is more weight in the right tail of the distribution. The function skewtest can be used to determine if the skewness value is close enough to zero, statistically speaking.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3, 2, 1]})
df.select(Polars.col("a").skew)
# =>
# shape: (1, 1)
# ┌──────────┐
# │ a        │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ 0.343622 │
# └──────────┘

Parameters:

  • bias (Boolean) (defaults to: true) —

    If false, the calculations are corrected for statistical bias.

Returns:



4348
4349
4350
 
# File 'lib/polars/expr.rb', line 4348

def skew(bias: true)
  wrap_expr(_rbexpr.skew(bias))
end

#slice(offset, length = nil) ⇒ Expr

Get a slice of this expression.

Examples:

df = Polars::DataFrame.new(
  {
    "a" => [8, 9, 10, 11],
    "b" => [nil, 4, 4, 4]
  }
)
df.select(Polars.all.slice(1, 2))
# =>
# shape: (2, 2)
# ┌─────┬─────┐
# │ a   ┆ b   │
# │ --- ┆ --- │
# │ i64 ┆ i64 │
# ╞═════╪═════╡
# │ 9   ┆ 4   │
# │ 10  ┆ 4   │
# └─────┴─────┘

Parameters:

  • offset (Integer) —

    Start index. Negative indexing is supported.

  • length (Integer) (defaults to: nil) —

    Length of the slice. If set to nil, all rows starting at the offset will be selected.

Returns:



778
779
780
781
782
783
784
785
786
 
# File 'lib/polars/expr.rb', line 778

def slice(offset, length = nil)
  if !offset.is_a?(Expr)
    offset = Polars.lit(offset)
  end
  if !length.is_a?(Expr)
    length = Polars.lit(length)
  end
  wrap_expr(_rbexpr.slice(offset._rbexpr, length._rbexpr))
end

#sort(reverse: false, nulls_last: false) ⇒ Expr

Sort this column. In projection/ selection context the whole column is sorted.

If used in a group by context, the groups are sorted.

Examples:

df = Polars::DataFrame.new(
  {
    "group" => [
        "one",
        "one",
        "one",
        "two",
        "two",
        "two"
    ],
    "value" => [1, 98, 2, 3, 99, 4]
  }
)
df.select(Polars.col("value").sort)
# =>
# shape: (6, 1)
# ┌───────┐
# │ value │
# │ ---   │
# │ i64   │
# ╞═══════╡
# │ 1     │
# │ 2     │
# │ 3     │
# │ 4     │
# │ 98    │
# │ 99    │
# └───────┘

df.select(Polars.col("value").sort)
# =>
# shape: (6, 1)
# ┌───────┐
# │ value │
# │ ---   │
# │ i64   │
# ╞═══════╡
# │ 1     │
# │ 2     │
# │ 3     │
# │ 4     │
# │ 98    │
# │ 99    │
# └───────┘

df.group_by("group").agg(Polars.col("value").sort)
# =>
# shape: (2, 2)
# ┌───────┬────────────┐
# │ group ┆ value      │
# │ ---   ┆ ---        │
# │ str   ┆ list[i64]  │
# ╞═══════╪════════════╡
# │ two   ┆ [3, 4, 99] │
# │ one   ┆ [1, 2, 98] │
# └───────┴────────────┘

Parameters:

  • reverse (Boolean) (defaults to: false) —

    false -> order from small to large. true -> order from large to small.

  • nulls_last (Boolean) (defaults to: false) —

    If true nulls are considered to be larger than any valid value.

Returns:



1314
1315
1316
 
# File 'lib/polars/expr.rb', line 1314

def sort(reverse: false, nulls_last: false)
  wrap_expr(_rbexpr.sort_with(reverse, nulls_last))
end

#sort_by(by, reverse: false) ⇒ Expr

Sort this column by the ordering of another column, or multiple other columns.

In projection/ selection context the whole column is sorted. If used in a group by context, the groups are sorted.

Examples:

df = Polars::DataFrame.new(
  {
    "group" => [
      "one",
      "one",
      "one",
      "two",
      "two",
      "two"
    ],
    "value" => [1, 98, 2, 3, 99, 4]
  }
)
df.select(Polars.col("group").sort_by("value"))
# =>
# shape: (6, 1)
# ┌───────┐
# │ group │
# │ ---   │
# │ str   │
# ╞═══════╡
# │ one   │
# │ one   │
# │ two   │
# │ two   │
# │ one   │
# │ two   │
# └───────┘

Parameters:

  • by (Object) —

    The column(s) used for sorting.

  • reverse (Boolean) (defaults to: false) —

    false -> order from small to large. true -> order from large to small.

Returns:



1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
 
# File 'lib/polars/expr.rb', line 1551

def sort_by(by, reverse: false)
  if !by.is_a?(::Array)
    by = [by]
  end
  if !reverse.is_a?(::Array)
    reverse = [reverse]
  end
  by = Utils.selection_to_rbexpr_list(by)

  wrap_expr(_rbexpr.sort_by(by, reverse))
end

#sqrt ⇒ Expr

Compute the square root of the elements.

Examples:

df = Polars::DataFrame.new({"values" => [1.0, 2.0, 4.0]})
df.select(Polars.col("values").sqrt)
# =>
# shape: (3, 1)
# ┌──────────┐
# │ values   │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ 1.0      │
# │ 1.414214 │
# │ 2.0      │
# └──────────┘

Returns:



248
249
250
 
# File 'lib/polars/expr.rb', line 248

def sqrt
  self**0.5
end

#std(ddof: 1) ⇒ Expr

Get standard deviation.

Examples:

df = Polars::DataFrame.new({"a" => [-1, 0, 1]})
df.select(Polars.col("a").std)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 1.0 │
# └─────┘

Parameters:

  • ddof (Integer) (defaults to: 1) —

    Degrees of freedom.

Returns:



1857
1858
1859
 
# File 'lib/polars/expr.rb', line 1857

def std(ddof: 1)
  wrap_expr(_rbexpr.std(ddof))
end

#str ⇒ StringExpr

Create an object namespace of all string related methods.

Returns:



5351
5352
5353
 
# File 'lib/polars/expr.rb', line 5351

def str
  StringExpr.new(self)
end

#struct ⇒ StructExpr

Create an object namespace of all struct related methods.

Returns:



5358
5359
5360
 
# File 'lib/polars/expr.rb', line 5358

def struct
  StructExpr.new(self)
end

#suffix(suffix) ⇒ Expr

Add a suffix to the root column name of the expression.

Returns:



424
425
426
 
# File 'lib/polars/expr.rb', line 424

def suffix(suffix)
  name.suffix(suffix)
end

#sum ⇒ Expr

Note:

Dtypes in :i8, :u8, :i16, and :u16 are cast to :i64 before summing to prevent overflow issues.

Get sum value.

Examples:

df = Polars::DataFrame.new({"a" => [-1, 0, 1]})
df.select(Polars.col("a").sum)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ i64 │
# ╞═════╡
# │ 0   │
# └─────┘

Returns:



1984
1985
1986
 
# File 'lib/polars/expr.rb', line 1984

def sum
  wrap_expr(_rbexpr.sum)
end

#tail(n = 10) ⇒ Expr

Get the last n rows.

Examples:

df = Polars::DataFrame.new({"foo" => [1, 2, 3, 4, 5, 6, 7]})
df.tail(3)
# =>
# shape: (3, 1)
# ┌─────┐
# │ foo │
# │ --- │
# │ i64 │
# ╞═════╡
# │ 5   │
# │ 6   │
# │ 7   │
# └─────┘

Parameters:

  • n (Integer) (defaults to: 10) —

    Number of rows to return.

Returns:



3006
3007
3008
 
# File 'lib/polars/expr.rb', line 3006

def tail(n = 10)
  wrap_expr(_rbexpr.tail(n))
end

#tan ⇒ Expr

Compute the element-wise value for the tangent.

Examples:

df = Polars::DataFrame.new({"a" => [1.0]})
df.select(Polars.col("a").tan)
# =>
# shape: (1, 1)
# ┌──────────┐
# │ a        │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ 1.557408 │
# └──────────┘

Returns:



4609
4610
4611
 
# File 'lib/polars/expr.rb', line 4609

def tan
  wrap_expr(_rbexpr.tan)
end

#tanh ⇒ Expr

Compute the element-wise value for the hyperbolic tangent.

Examples:

df = Polars::DataFrame.new({"a" => [1.0]})
df.select(Polars.col("a").tanh)
# =>
# shape: (1, 1)
# ┌──────────┐
# │ a        │
# │ ---      │
# │ f64      │
# ╞══════════╡
# │ 0.761594 │
# └──────────┘

Returns:



4729
4730
4731
 
# File 'lib/polars/expr.rb', line 4729

def tanh
  wrap_expr(_rbexpr.tanh)
end

#to_physical ⇒ Expr

Cast to physical representation of the logical dtype.

  • :date -> :i32
  • :datetime -> :i64
  • :time -> :i64
  • :duration -> :i64
  • :cat -> :u32
  • Other data types will be left unchanged.

Examples:

Polars::DataFrame.new({"vals" => ["a", "x", nil, "a"]}).with_columns(
  [
    Polars.col("vals").cast(:cat),
    Polars.col("vals")
      .cast(:cat)
      .to_physical
      .alias("vals_physical")
  ]
)
# =>
# shape: (4, 2)
# ┌──────┬───────────────┐
# │ vals ┆ vals_physical │
# │ ---  ┆ ---           │
# │ cat  ┆ u32           │
# ╞══════╪═══════════════╡
# │ a    ┆ 0             │
# │ x    ┆ 1             │
# │ null ┆ null          │
# │ a    ┆ 0             │
# └──────┴───────────────┘

Returns:



181
182
183
 
# File 'lib/polars/expr.rb', line 181

def to_physical
  wrap_expr(_rbexpr.to_physical)
end

#to_s ⇒ String Also known as: inspect

Returns a string representing the Expr.

Returns:



17
18
19
 
# File 'lib/polars/expr.rb', line 17

def to_s
  _rbexpr.to_str
end

#top_k(k: 5) ⇒ Expr

Return the k largest elements.

If 'reverse: true` the smallest elements will be given.

Examples:

df = Polars::DataFrame.new(
  {
    "value" => [1, 98, 2, 3, 99, 4]
  }
)
df.select(
  [
    Polars.col("value").top_k.alias("top_k"),
    Polars.col("value").bottom_k.alias("bottom_k")
  ]
)
# =>
# shape: (5, 2)
# ┌───────┬──────────┐
# │ top_k ┆ bottom_k │
# │ ---   ┆ ---      │
# │ i64   ┆ i64      │
# ╞═══════╪══════════╡
# │ 99    ┆ 1        │
# │ 98    ┆ 2        │
# │ 4     ┆ 3        │
# │ 3     ┆ 4        │
# │ 2     ┆ 98       │
# └───────┴──────────┘

Parameters:

  • k (Integer) (defaults to: 5) —

    Number of elements to return.

Returns:



1352
1353
1354
1355
 
# File 'lib/polars/expr.rb', line 1352

def top_k(k: 5)
  k = Utils.parse_as_expression(k)
  wrap_expr(_rbexpr.top_k(k))
end

#unique(maintain_order: false) ⇒ Expr

Get unique values of this expression.

Examples:

df = Polars::DataFrame.new({"a" => [1, 1, 2]})
df.select(Polars.col("a").unique(maintain_order: true))
# =>
# shape: (2, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ i64 │
# ╞═════╡
# │ 1   │
# │ 2   │
# └─────┘

Parameters:

  • maintain_order (Boolean) (defaults to: false) —

    Maintain order of data. This requires more work.

Returns:



2175
2176
2177
2178
2179
2180
2181
 
# File 'lib/polars/expr.rb', line 2175

def unique(maintain_order: false)
  if maintain_order
    wrap_expr(_rbexpr.unique_stable)
  else
    wrap_expr(_rbexpr.unique)
  end
end

#unique_counts ⇒ Expr

Return a count of the unique values in the order of appearance.

This method differs from value_counts in that it does not return the values, only the counts and might be faster

Examples:

df = Polars::DataFrame.new(
  {
    "id" => ["a", "b", "b", "c", "c", "c"]
  }
)
df.select(
  [
    Polars.col("id").unique_counts
  ]
)
# =>
# shape: (3, 1)
# ┌─────┐
# │ id  │
# │ --- │
# │ u32 │
# ╞═════╡
# │ 1   │
# │ 2   │
# │ 3   │
# └─────┘

Returns:



5093
5094
5095
 
# File 'lib/polars/expr.rb', line 5093

def unique_counts
  wrap_expr(_rbexpr.unique_counts)
end

#upper_bound ⇒ Expr

Calculate the upper bound.

Returns a unit Series with the highest value possible for the dtype of this expression.

Examples:

df = Polars::DataFrame.new({"a" => [1, 2, 3, 2, 1]})
df.select(Polars.col("a").upper_bound)
# =>
# shape: (1, 1)
# ┌─────────────────────┐
# │ a                   │
# │ ---                 │
# │ i64                 │
# ╞═════════════════════╡
# │ 9223372036854775807 │
# └─────────────────────┘

Returns:



4525
4526
4527
 
# File 'lib/polars/expr.rb', line 4525

def upper_bound
  wrap_expr(_rbexpr.upper_bound)
end

#value_counts(multithreaded: false, sort: false) ⇒ Expr

Count all unique values and create a struct mapping value to count.

Examples:

df = Polars::DataFrame.new(
  {
    "id" => ["a", "b", "b", "c", "c", "c"]
  }
)
df.select(
  [
    Polars.col("id").value_counts(sort: true),
  ]
)
# =>
# shape: (3, 1)
# ┌───────────┐
# │ id        │
# │ ---       │
# │ struct[2] │
# ╞═══════════╡
# │ {"c",3}   │
# │ {"b",2}   │
# │ {"a",1}   │
# └───────────┘

Parameters:

  • multithreaded (Boolean) (defaults to: false) —

    Better to turn this off in the aggregation context, as it can lead to contention.

  • sort (Boolean) (defaults to: false) —

    Ensure the output is sorted from most values to least.

Returns:



5060
5061
5062
 
# File 'lib/polars/expr.rb', line 5060

def value_counts(multithreaded: false, sort: false)
  wrap_expr(_rbexpr.value_counts(multithreaded, sort))
end

#var(ddof: 1) ⇒ Expr

Get variance.

Examples:

df = Polars::DataFrame.new({"a" => [-1, 0, 1]})
df.select(Polars.col("a").var)
# =>
# shape: (1, 1)
# ┌─────┐
# │ a   │
# │ --- │
# │ f64 │
# ╞═════╡
# │ 1.0 │
# └─────┘

Parameters:

  • ddof (Integer) (defaults to: 1) —

    Degrees of freedom.

Returns:



1880
1881
1882
 
# File 'lib/polars/expr.rb', line 1880

def var(ddof: 1)
  wrap_expr(_rbexpr.var(ddof))
end

#where(predicate) ⇒ Expr

Filter a single column.

Alias for #filter.

Examples:

df = Polars::DataFrame.new(
  {
    "group_col" => ["g1", "g1", "g2"],
    "b" => [1, 2, 3]
  }
)
(
  df.group_by("group_col").agg(
    [
      Polars.col("b").where(Polars.col("b") < 2).sum.alias("lt"),
      Polars.col("b").where(Polars.col("b") >= 2).sum.alias("gte")
    ]
  )
).sort("group_col")
# =>
# shape: (2, 3)
# ┌───────────┬─────┬─────┐
# │ group_col ┆ lt  ┆ gte │
# │ ---       ┆ --- ┆ --- │
# │ str       ┆ i64 ┆ i64 │
# ╞═══════════╪═════╪═════╡
# │ g1        ┆ 1   ┆ 2   │
# │ g2        ┆ 0   ┆ 3   │
# └───────────┴─────┴─────┘

Parameters:

  • predicate (Expr) —

    Boolean expression.

Returns:



2755
2756
2757
 
# File 'lib/polars/expr.rb', line 2755

def where(predicate)
  filter(predicate)
end

#|(other) ⇒ Expr

Bitwise OR.

Returns:



39
40
41
 
# File 'lib/polars/expr.rb', line 39

def |(other)
  wrap_expr(_rbexpr._or(_to_rbexpr(other)))
end