Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add Distance Functions #172

Merged
merged 2 commits into from
Jan 21, 2025
Merged

Add Distance Functions #172

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a6f63ec
Add Distance Functions
Jan 20, 2025
f6536ac
Add numerical type constraints
Jan 21, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 8 additions & 7 deletions dsl/src/main/scala/com/crobox/clickhouse/dsl/column/BitFunctions.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,14 +8,15 @@ trait BitFunctions { self: Magnets =>
case class BitAnd(a: NumericCol[_], b: NumericCol[_]) extends BitFunction(a)
case class BitOr(a: NumericCol[_], b: NumericCol[_]) extends BitFunction(a)
case class BitXor(a: NumericCol[_], b: NumericCol[_]) extends BitFunction(a)
case class BitNot(a: NumericCol[_]) extends BitFunction(a)
case class BitNot(a: NumericCol[_]) extends BitFunction(a)
case class BitShiftLeft(a: NumericCol[_], b: NumericCol[_]) extends BitFunction(a)
case class BitShiftRight(a: NumericCol[_], b: NumericCol[_]) extends BitFunction(a)

def bitAnd(a: NumericCol[_], b: NumericCol[_]) = BitAnd(a: NumericCol[_], b: NumericCol[_])
def bitOr(a: NumericCol[_], b: NumericCol[_]) = BitOr(a: NumericCol[_], b: NumericCol[_])
def bitXor(a: NumericCol[_], b: NumericCol[_]) = BitXor(a: NumericCol[_], b: NumericCol[_])
def bitNot(a: NumericCol[_]) = BitNot(a: NumericCol[_])
def bitShiftLeft(a: NumericCol[_], b: NumericCol[_]) = BitShiftLeft(a: NumericCol[_], b: NumericCol[_])
def bitShiftRight(a: NumericCol[_], b: NumericCol[_]) = BitShiftRight(a: NumericCol[_], b: NumericCol[_])
def bitAnd(a: NumericCol[_], b: NumericCol[_]): BitAnd = BitAnd(a: NumericCol[_], b: NumericCol[_])
def bitOr(a: NumericCol[_], b: NumericCol[_]): BitOr = BitOr(a: NumericCol[_], b: NumericCol[_])
def bitXor(a: NumericCol[_], b: NumericCol[_]): BitXor = BitXor(a: NumericCol[_], b: NumericCol[_])
def bitNot(a: NumericCol[_]): BitNot = BitNot(a: NumericCol[_])
def bitShiftLeft(a: NumericCol[_], b: NumericCol[_]): BitShiftLeft = BitShiftLeft(a: NumericCol[_], b: NumericCol[_])
def bitShiftRight(a: NumericCol[_], b: NumericCol[_]): BitShiftRight =
BitShiftRight(a: NumericCol[_], b: NumericCol[_])
}
1 change: 1 addition & 0 deletions dsl/src/main/scala/com/crobox/clickhouse/dsl/column/ClickhouseColumnFunctions.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,7 @@ trait ClickhouseColumnFunctions
with ComparisonFunctions
with DateTimeFunctions
with DictionaryFunctions
with DistanceFunctions
with EmptyFunctions
with EncodingFunctions
with HashFunctions
Expand Down
133 changes: 133 additions & 0 deletions dsl/src/main/scala/com/crobox/clickhouse/dsl/column/DistanceFunctions.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,133 @@
package com.crobox.clickhouse.dsl.column

import com.crobox.clickhouse.dsl.{EmptyColumn, ExpressionColumn}

trait DistanceFunctions { self: Magnets =>

sealed trait DistanceFunction
abstract class DistanceFunctionOp[V] extends ExpressionColumn[V](EmptyColumn) with DistanceFunction

// L1
case class L1Norm[V](vector: ArrayColMagnet[_ <: Iterable[V]])(implicit evidence: V => NumericCol[V])
extends DistanceFunctionOp[V]
case class L1Normalize[V](vector1: ArrayColMagnet[_ <: Iterable[V]], vector2: ArrayColMagnet[_ <: Iterable[V]])
extends DistanceFunctionOp[V]
case class L1Distance[V](vector1: ArrayColMagnet[_ <: Iterable[V]], vector2: ArrayColMagnet[_ <: Iterable[V]])(
implicit evidence: V => NumericCol[V]
) extends DistanceFunctionOp[V]

// L2
case class L2Norm[V](vector: ArrayColMagnet[_ <: Iterable[V]])(implicit evidence: V => NumericCol[V])
extends DistanceFunctionOp[V]
case class L2Normalize[V](vector1: ArrayColMagnet[_ <: Iterable[V]], vector2: ArrayColMagnet[_ <: Iterable[V]])
extends DistanceFunctionOp[V]
case class L2Distance[V](vector1: ArrayColMagnet[_ <: Iterable[V]], vector2: ArrayColMagnet[_ <: Iterable[V]])(
implicit evidence: V => NumericCol[V]
) extends DistanceFunctionOp[V]

// L2 Squared
case class L2SquaredNorm[V](vector: ArrayColMagnet[_ <: Iterable[V]])(implicit evidence: V => NumericCol[V])
extends DistanceFunctionOp[V]
case class L2SquaredDistance[V](vector1: ArrayColMagnet[_ <: Iterable[V]], vector2: ArrayColMagnet[_ <: Iterable[V]])(
implicit evidence: V => NumericCol[V]
) extends DistanceFunctionOp[V]

// LInf
case class LInfNorm[V](vector: ArrayColMagnet[_ <: Iterable[V]])(implicit evidence: V => NumericCol[V])
extends DistanceFunctionOp[V]
case class LInfNormalize[V](vector1: ArrayColMagnet[_ <: Iterable[V]], vector2: ArrayColMagnet[_ <: Iterable[V]])
extends DistanceFunctionOp[V]
case class LInfDistance[V](vector1: ArrayColMagnet[_ <: Iterable[V]], vector2: ArrayColMagnet[_ <: Iterable[V]])(
implicit evidence: V => NumericCol[V]
) extends DistanceFunctionOp[V]

// LP
case class LPNorm[V](vector: ArrayColMagnet[_ <: Iterable[V]], p: Float)(implicit evidence: V => NumericCol[V])
extends DistanceFunctionOp[V]
case class LPNormalize[V](
vector1: ArrayColMagnet[_ <: Iterable[V]],
vector2: ArrayColMagnet[_ <: Iterable[V]],
p: Float
) extends DistanceFunctionOp[V]
case class LPDistance[V](
vector1: ArrayColMagnet[_ <: Iterable[V]],
vector2: ArrayColMagnet[_ <: Iterable[V]],
p: Float
)(implicit
evidence: V => NumericCol[V]
) extends DistanceFunctionOp[V]

// cosine
case class CosineDistance[V](
vector1: ArrayColMagnet[_ <: Iterable[V]],
vector2: ArrayColMagnet[_ <: Iterable[V]]
)(implicit
evidence: V => NumericCol[V]
) extends DistanceFunctionOp[V]

// utilities
def l1Norm[V](vector: ArrayColMagnet[_ <: Iterable[V]])(implicit evidence: V => NumericCol[V]): L1Norm[V] =
L1Norm(vector)

def l1Normalize[V](vector1: ArrayColMagnet[_ <: Iterable[V]], vector2: ArrayColMagnet[_ <: Iterable[V]])(implicit
evidence: V => NumericCol[V]
): L1Normalize[V] = L1Normalize(vector1, vector2)

def l1Distance[V](vector1: ArrayColMagnet[_ <: Iterable[V]], vector2: ArrayColMagnet[_ <: Iterable[V]])(implicit
evidence: V => NumericCol[V]
): L1Distance[V] = L1Distance(vector1, vector2)

def l2Norm[V](vector: ArrayColMagnet[_ <: Iterable[V]])(implicit evidence: V => NumericCol[V]): L2Norm[V] =
L2Norm(vector)

def l2Normalize[V](
vector1: ArrayColMagnet[_ <: Iterable[V]],
vector2: ArrayColMagnet[_ <: Iterable[V]]
): L2Normalize[V] =
L2Normalize(vector1, vector2)

def l2Distance[V](vector1: ArrayColMagnet[_ <: Iterable[V]], vector2: ArrayColMagnet[_ <: Iterable[V]])(implicit
evidence: V => NumericCol[V]
): L2Distance[V] = L2Distance(vector1, vector2)

def l2SquaredNorm[V](vector: ArrayColMagnet[_ <: Iterable[V]])(implicit
evidence: V => NumericCol[V]
): L2SquaredNorm[V] = L2SquaredNorm(vector)

def l2SquaredDistance[V](vector1: ArrayColMagnet[_ <: Iterable[V]], vector2: ArrayColMagnet[_ <: Iterable[V]])(
implicit evidence: V => NumericCol[V]
): L2SquaredDistance[V] = L2SquaredDistance(vector1, vector2)

def lInfNorm[V](vector: ArrayColMagnet[_ <: Iterable[V]])(implicit evidence: V => NumericCol[V]): LInfNorm[V] =
LInfNorm(vector)

def lInfNormalize[V](
vector1: ArrayColMagnet[_ <: Iterable[V]],
vector2: ArrayColMagnet[_ <: Iterable[V]]
): LInfNormalize[V] = LInfNormalize(vector1, vector2)

def lInfDistance[V](vector1: ArrayColMagnet[_ <: Iterable[V]], vector2: ArrayColMagnet[_ <: Iterable[V]])(implicit
evidence: V => NumericCol[V]
): LInfDistance[V] = LInfDistance(vector1, vector2)

def lPNorm[V](vector: ArrayColMagnet[_ <: Iterable[V]], p: Float)(implicit evidence: V => NumericCol[V]): LPNorm[V] =
LPNorm(vector, p)

def lPNormalize[V](
vector1: ArrayColMagnet[_ <: Iterable[V]],
vector2: ArrayColMagnet[_ <: Iterable[V]],
p: Float
): LPNormalize[V] = LPNormalize(vector1, vector2, p)

def lPDistance[V](
vector1: ArrayColMagnet[_ <: Iterable[V]],
vector2: ArrayColMagnet[_ <: Iterable[V]],
p: Float
)(implicit evidence: V => NumericCol[V]): LPDistance[V] = LPDistance(vector1, vector2, p)

def cosineDistance[V](
vector1: ArrayColMagnet[_ <: Iterable[V]],
vector2: ArrayColMagnet[_ <: Iterable[V]]
)(implicit evidence: V => NumericCol[V]): CosineDistance[V] = CosineDistance(vector1, vector2)

}
2 changes: 2 additions & 0 deletions dsl/src/main/scala/com/crobox/clickhouse/dsl/language/ClickhouseTokenizerModule.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,6 +53,7 @@ trait ClickhouseTokenizerModule
with ComparisonFunctionTokenizer
with DateTimeFunctionTokenizer
with DictionaryFunctionTokenizer
with DistanceFunctionTokenizer
with EncodingFunctionTokenizer
with HashFunctionTokenizer
with HigherOrderFunctionTokenizer
Expand Down Expand Up @@ -168,6 +169,7 @@ trait ClickhouseTokenizerModule
case col: DateTimeFunctionCol[_] => tokenizeDateTimeColumn(col)
case col: DateTimeConst[_] => tokenizeDateTimeConst(col)
case col: DictionaryFuncColumn[_] => tokenizeDictionaryFunction(col)
case col: DistanceFunction => tokenizeDistanceFunction(col)
case col: EmptyFunction[_] => tokenizeEmptyCol(col)
case col: EncodingFunction[_] => tokenizeEncodingFunction(col)
case col: HashFunction => tokenizeHashFunction(col)
Expand Down
47 changes: 47 additions & 0 deletions dsl/src/main/scala/com/crobox/clickhouse/dsl/language/DistanceFunctionTokenizer.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,47 @@
package com.crobox.clickhouse.dsl.language

import com.crobox.clickhouse.dsl._

trait DistanceFunctionTokenizer {
self: ClickhouseTokenizerModule =>

def tokenizeDistanceFunction(col: DistanceFunction)(implicit ctx: TokenizeContext): String = col match {

// cosine
case CosineDistance(vector1, vector2) =>
s"cosineDistance(${tokenizeColumn(vector1.column)}, ${tokenizeColumn(vector2.column)})"

// L1
case L1Norm(vector) => s"L1Norm(${tokenizeColumn(vector.column)})"
case L1Normalize(vector1, vector2) =>
s"L1Normalize(${tokenizeColumn(vector1.column)}, ${tokenizeColumn(vector2.column)})"
case L1Distance(vector1, vector2) =>
s"L1Distance(${tokenizeColumn(vector1.column)}, ${tokenizeColumn(vector2.column)})"

// L2
case L2Norm(vector) => s"L2Norm(${tokenizeColumn(vector.column)})"
case L2Normalize(vector1, vector2) =>
s"L2Normalize(${tokenizeColumn(vector1.column)}, ${tokenizeColumn(vector2.column)})"
case L2Distance(vector1, vector2) =>
s"L2Distance(${tokenizeColumn(vector1.column)}, ${tokenizeColumn(vector2.column)})"

// L2Squared
case L2SquaredNorm(vector) => s"L2SquaredNorm(${tokenizeColumn(vector.column)})"
case L2SquaredDistance(vector1, vector2) =>
s"L2SquaredDistance(${tokenizeColumn(vector1.column)}, ${tokenizeColumn(vector2.column)})"

// LInf
case LInfNorm(vector) => s"LinfNorm(${tokenizeColumn(vector.column)})"
case LInfNormalize(vector1, vector2) =>
s"LinfNormalize(${tokenizeColumn(vector1.column)}, ${tokenizeColumn(vector2.column)})"
case LInfDistance(vector1, vector2) =>
s"LinfDistance(${tokenizeColumn(vector1.column)}, ${tokenizeColumn(vector2.column)})"

// LP
case LPNorm(vector, p) => s"LpNorm(${tokenizeColumn(vector.column)}, $p)"
case LPNormalize(vector1, vector2, p: Float) =>
s"LpNormalize(${tokenizeColumn(vector1.column)}, ${tokenizeColumn(vector2.column)}, $p)"
case LPDistance(vector1, vector2, p: Float) =>
s"LpDistance(${tokenizeColumn(vector1.column)}, ${tokenizeColumn(vector2.column)}, $p)"
}
}
Loading
Loading