Aidge backend CPU#

When imported automatically update availables implementations for aidge core objects.

Supported operators (✔️ Yes, ✅ Generic impl., ☑️ MetaOp impl., ❌ No)#

Operator

Support
Impl specs
(— = any ; ✱ = additional requirements)
Abs
(aidge_core.Abs)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int32 — → int32 —
Add
(aidge_core.Add)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int8 — → int8 —
4. int8 — → int32 —
5. int8 — → int64 —
6. int32 — → int32 —
7. int64 — → int64 —
8. uint8 — → uint8 —
And
(aidge_core.And)

✔️
1. boolean — → boolean —
ArgMax
(aidge_core.ArgMax)

✔️
1. float64 — → int64 —
2. float32 — → int64 —
3. int32 — → int64 —
4. int64 — → int64 —
Atan
(aidge_core.Atan)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
AvgPooling1D
(aidge_core.AvgPooling1D)

✅
AvgPooling2D
(aidge_core.AvgPooling2D)

✔️
1. float64 nchw → float64 nchw
2. float32 nchw → float32 nchw
3. int8 nchw → int8 nchw
4. int32 nchw → int32 nchw
AvgPooling3D
(aidge_core.AvgPooling3D)

✅
BatchNorm2D
(aidge_core.BatchNorm2D)

✔️
1. float32 nchw → float32 nchw
BitErrorRate
(aidge_core.BitErrorRate)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int8 — → int8 —
4. int32 — → int32 —
BitShift
(aidge_core.BitShift)

✔️
1. int32 — → int32 —
2. int64 — → int64 —
3. uint32 — → uint32 —
4. uint64 — → uint64 —
CastLike
(aidge_core.CastLike)

✅
Cast
(aidge_core.Cast)

✅
Ceil
(aidge_core.Ceil)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
Clip
(aidge_core.Clip)

✔️
1. float64 — | — — | — — → float64 —
2. float32 — → float32 —
3. int8 — | — — | — — → int8 —
4. int16 — | — — | — — → int16 —
5. int32 — | — — | — — → int32 —
6. int64 — | — — | — — → int64 —
ComplexToInnerPair
(aidge_core.ComplexToInnerPair)

✅
Concat
(aidge_core.Concat)

✅
ConstantOfShape
(aidge_core.ConstantOfShape)

✔️
1. int64 — → float64 —
2. int64 — → float32 —
3. int64 — → float16 —
4. int64 — → int16 —
5. int64 — → int32 —
6. int64 — → int64 —
Conv1D
(aidge_core.Conv1D)

✔️
1. — nchw → float64 nchw
2. — nchw → float32 nchw
3. — nchw → float16 nchw
4. — nchw → int32 nchw
Conv2D
(aidge_core.Conv2D)

✔️
1. — nchw → float64 nchw
2. — nchw → float32 nchw
3. — nchw → float16 nchw
4. — nchw → int32 nchw
Conv3D
(aidge_core.Conv3D)

✔️
1. — nchw → float64 nchw
2. — nchw → float32 nchw
3. — nchw → float16 nchw
4. — nchw → int32 nchw
ConvDepthWise1D
(aidge_core.ConvDepthWise1D)

✔️
1. — nchw → float64 nchw
2. — nchw → float32 nchw
3. — nchw → int32 nchw
ConvDepthWise2D
(aidge_core.ConvDepthWise2D)

✔️
1. — nchw → float64 nchw
2. — nchw → float32 nchw
3. — nchw → int32 nchw
ConvTranspose1D
(aidge_core.ConvTranspose1D)

✔️
1. — nchw → float64 nchw
2. — nchw → float32 nchw
3. — nchw → float16 nchw
4. — nchw → int32 nchw
ConvTranspose2D
(aidge_core.ConvTranspose2D)

✔️
1. — nchw → float64 nchw
2. — nchw → float32 nchw
3. — nchw → float16 nchw
4. — nchw → int32 nchw
ConvTranspose3D
(aidge_core.ConvTranspose3D)

✔️
1. — nchw → float64 nchw
2. — nchw → float32 nchw
3. — nchw → float16 nchw
4. — nchw → int32 nchw
Cosh
(aidge_core.Cosh)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
Cos
(aidge_core.Cos)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
CryptoHash
(aidge_core.CryptoHash)

✅
DepthToSpace
(aidge_core.DepthToSpace)

✅
DFT
(aidge_core.DFT)

✔️
1. float64 — | — — → complex64 —
2. float32 — | — — → complex32 —
3. complex64 — | — — → complex64 —
4. complex32 — | — — → complex32 —
Div
(aidge_core.Div)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int32 — → int32 —
4. int64 — → int64 —
5. uint8 — → uint8 —
Dropout
(aidge_core.Dropout)

✔️
1. float64 — | — — | boolean — → float64 — | boolean — ✱ 1
2. float64 — | — — | boolean — → float64 — | boolean — ✱ 1
3. float32 — | — — | boolean — → float32 — | boolean — ✱ 1
4. float32 — | — — | boolean — → float32 — | boolean — ✱ 1
Equal
(aidge_core.Equal)

✔️
1. float64 — | float64 — → boolean —
2. float32 — | float32 — → boolean —
3. boolean — | boolean — → boolean —
4. int32 — | int32 — → boolean —
5. int64 — | int64 — → boolean —
Erf
(aidge_core.Erf)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int32 — → int32 —
Expand
(aidge_core.Expand)

✔️
1. float64 — | int64 — → float64 —
2. float32 — | int64 — → float32 —
3. float16 — | int64 — → float16 —
4. int16 — | int64 — → int16 —
5. int32 — | int64 — → int32 —
6. int64 — | int64 — → int64 —
Exp
(aidge_core.Exp)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. float16 — → float16 —
FC
(aidge_core.FC)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int8 — | int8 — | int32 — → int32 —
4. int32 — → int32 —
FixedNBitFlip
(aidge_core.FixedNBitFlip)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int32 — → int32 —
Flatten
(aidge_core.Flatten)

✅
Floor
(aidge_core.Floor)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
Fold2D
(aidge_core.Fold2D)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int32 — → int32 —
Gather
(aidge_core.Gather)

✅
Gemm
(aidge_core.Gemm)

☑️
GlobalAveragePooling
(aidge_core.GlobalAveragePooling)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int8 — → int8 —
4. int32 — → int32 —
GridSample
(aidge_core.GridSample)

✅
GRU
(aidge_core.GRU)

☑️
HannWindow
(aidge_core.HannWindow)

☑️
Hardmax
(aidge_core.Hardmax)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int32 — → int32 —
HardSigmoid
(aidge_core.HardSigmoid)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
HardSwish
(aidge_core.HardSwish)

☑️
Heaviside
(aidge_core.Heaviside)

✔️
1. float32 — → float32 —
Identity
(aidge_core.Identity)

✅
InnerPairToComplex
(aidge_core.InnerPairToComplex)

✅
InstanceNorm
(aidge_core.InstanceNorm)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
LayerNorm
(aidge_core.LayerNorm)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
LeakyReLU
(aidge_core.LeakyReLU)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int32 — → int32 —
Ln
(aidge_core.Ln)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
LRN
(aidge_core.LRN)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
LSTM
(aidge_core.LSTM)

☑️
MatMul
(aidge_core.MatMul)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int8 — → int32 —
4. int32 — → int32 —
Max
(aidge_core.Max)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int8 — → int8 —
4. int16 — → int16 —
5. int32 — → int32 —
6. int64 — → int64 —
7. uint8 — → uint8 —
8. uint16 — → uint16 —
MaxPooling1D
(aidge_core.MaxPooling1D)

✅
MaxPooling2D
(aidge_core.MaxPooling2D)

✔️
1. float64 — → float64 — | int64 —
2. float32 — → float32 — | int64 —
3. int8 — → int8 — | int64 —
4. int32 — → int32 — | int64 —
MaxPooling3D
(aidge_core.MaxPooling3D)

✅
Memorize
(aidge_core.Memorize)

✅
Min
(aidge_core.Min)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int8 — → int8 —
4. int16 — → int16 —
5. int32 — → int32 —
6. int64 — → int64 —
7. uint8 — → uint8 —
8. uint16 — → uint16 —
Mod
(aidge_core.Mod)

✔️
1. float64 — | float64 — → float64 —
2. float32 — | float32 — → float32 —
3. float16 — | float16 — → float16 —
4. int8 — | int8 — → int8 —
5. int16 — | int16 — → int16 —
6. int32 — | int32 — → int32 —
7. int64 — | int64 — → int64 —
8. uint8 — | uint8 — → uint8 —
9. uint16 — | uint16 — → uint16 —
10. uint32 — | uint32 — → uint32 —
11. uint64 — | uint64 — → uint64 —
Mul
(aidge_core.Mul)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. float32 — | float64 — → float32 —
4. int8 — | int8 — → int32 —
5. int32 — → int32 —
6. int32 — | int8 — → int32 —
7. int64 — → int64 —
8. uint8 — → uint8 —
NBitFlip
(aidge_core.NBitFlip)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int32 — → int32 —
Neg
(aidge_core.Neg)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. float16 — → float16 —
4. int32 — → int32 —
5. int64 — → int64 —
OneHot
(aidge_core.OneHot)

✔️
1. float64 — | — — → float64 —
2. float32 — | — — → float32 —
3. int32 — | — — → int32 —
4. int64 — | — — → int64 —
PaddedAvgPooling2D
(aidge_core.PaddedAvgPooling2D)

☑️
PaddedConv1D
(aidge_core.PaddedConv1D)

☑️
PaddedConv2D
(aidge_core.PaddedConv2D)

☑️
PaddedConv3D
(aidge_core.PaddedConv3D)

☑️
PaddedConvDepthWise1D
(aidge_core.PaddedConvDepthWise1D)

☑️
PaddedConvDepthWise2D
(aidge_core.PaddedConvDepthWise2D)

☑️
PaddedConvTranspose1D
(aidge_core.PaddedConvTranspose1D)

☑️
PaddedConvTranspose2D
(aidge_core.PaddedConvTranspose2D)

☑️
PaddedConvTranspose3D
(aidge_core.PaddedConvTranspose3D)

☑️
PaddedMaxPooling2D
(aidge_core.PaddedMaxPooling2D)

☑️
Pad
(aidge_core.Pad)

✔️
1. float64 —1 | — — | — — | — — → float64 —1
2. float32 —1 | — — | — — | — — → float32 —1
3. float16 —1 | — — | — — | — — → float16 —1
4. int8 —1 | — — | — — | — — → int8 —1
5. int32 —1 | — — | — — | — — → int32 —1
Pop
(aidge_core.Pop)

✅
Pow
(aidge_core.Pow)

✔️
1. — — → float64 —
2. — — → float32 —
3. — — → int8 —
4. — — → int32 —
5. — — → int64 —
6. — — → uint8 —
Producer
(aidge_core.Producer)

✅
RandomNormalLike
(aidge_core.RandomNormalLike)

✔️
1. — — → float64 —
2. — — → float32 —
Range
(aidge_core.Range)

✅
Reciprocal
(aidge_core.Reciprocal)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
ReduceMax
(aidge_core.ReduceMax)

✔️
1. float64 — | — — → float64 —
2. float32 — | — — → float32 —
3. boolean — | — — → boolean —
4. int32 — | — — → int32 —
ReduceMean
(aidge_core.ReduceMean)

✔️
1. float64 — | — — → float64 —
2. float32 — | — — → float32 —
3. int32 — | — — → int32 —
ReduceMin
(aidge_core.ReduceMin)

✔️
1. float64 — | — — → float64 —
2. float32 — | — — → float32 —
3. boolean — | — — → boolean —
4. int32 — | — — → int32 —
ReduceSum
(aidge_core.ReduceSum)

✔️
1. float64 — | — — → float64 —
2. float32 — | — — → float32 —
3. int32 — | — — → int32 —
ReLU
(aidge_core.ReLU)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int8 — → int8 —
4. int32 — → int32 —
Reshape
(aidge_core.Reshape)

✅
Resize
(aidge_core.Resize)

✔️
1. float64 — | — — | — — | — — → float64 —
2. float32 — | — — | — — | — — → float32 —
3. float16 — | — — | — — | — — → float16 —
4. int8 — | — — | — — | — — → int8 —
5. int16 — | — — | — — | — — → int16 —
6. int32 — | — — | — — | — — → int32 —
7. int64 — | — — | — — | — — → uint64 —
Round
(aidge_core.Round)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
Scatter
(aidge_core.Scatter)

✔️
1. float64 — | — — → float64 —
2. float32 — | — — → float32 —
3. int8 — | — — → int8 —
4. int32 — | — — → int32 —
Select
(aidge_core.Select)

✅
Shape
(aidge_core.Shape)

✅
Sigmoid
(aidge_core.Sigmoid)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int32 — → int32 —
Sinh
(aidge_core.Sinh)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
Sin
(aidge_core.Sin)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
Slice
(aidge_core.Slice)

✔️
1. float64 — | — — → float64 —
2. float32 — | — — → float32 —
3. int32 — | — — → int32 —
4. int64 — | — — → int64 —
SLSTM
(aidge_core.SLSTM)

☑️
Softmax
(aidge_core.Softmax)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int32 — → int32 —
Split
(aidge_core.Split)

✅
Sqrt
(aidge_core.Sqrt)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int32 — → int32 —
Squeeze
(aidge_core.Squeeze)

✅
Stack
(aidge_core.Stack)

✅
STFT
(aidge_core.STFT)

✔️
1. float64 — | — — → complex64 —
2. float32 — | — — → complex32 —
3. complex64 — | — — → complex64 —
4. complex32 — | — — → complex32 —
Sub
(aidge_core.Sub)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
3. int8 — → int8 —
4. int32 — → int32 —
5. int64 — → int64 —
6. uint8 — → uint8 —
Sum
(aidge_core.Sum)

✔️
1. — — → float64 —
2. — — → float32 —
3. — — → int32 —
4. — — → int64 —
Tanh
(aidge_core.Tanh)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
Tan
(aidge_core.Tan)

✔️
1. float64 — → float64 —
2. float32 — → float32 —
Tile
(aidge_core.Tile)

✅
TopK
(aidge_core.TopK)

✔️
1. float64 — | — — → float64 — | int64 —
2. float32 — | — — → float32 — | int64 —
3. int8 — | — — → int8 — | int64 —
4. int32 — | — — → int32 — | int64 —
Transpose
(aidge_core.Transpose)

✅
Unfold2D
(aidge_core.Unfold2D)

✅
Unsqueeze
(aidge_core.Unsqueeze)

✅
WeightInterleaving
(aidge_core.WeightInterleaving)

✔️
1. binary — → octo_binary —
2. int2 — → quad_int2 —
3. uint2 — → quad_uint2 —
4. int3 — → dual_int3 —
5. uint3 — → dual_uint3 —
6. int4 — → dual_int4 —
7. uint4 — → dual_uint4 —
Where
(aidge_core.Where)

✔️
1. boolean — | — — | — — → float64 —
2. boolean — | — — | — — → float32 —
3. boolean — | — — | — — → int32 —
4. boolean — | — — | — — → int64 —

This Page