'tfl' Dialek

Dialek TensorFlow Lite.

Dialek ini dipetakan ke operasi TensorFlow Lite.

Invarian:

• Semua nilai bertipe Tensor (khususnya, skalar direpresentasikan menggunakan tensor dimensi nol);

Operasi

tfl.abs (TFL::AbsOp)

Operator nilai absolut

Jika diberikan tensor x , operasi ini mengembalikan tensor yang berisi nilai absolut setiap elemen dalam x . Misalnya, jika x adalah elemen masukan dan y adalah elemen keluaran, operasi ini akan menghitung \(y = |x|\).

Sifat: AlwaysSpeculatableImplTrait , QuantizableResult , SameOperandsAndResultShape

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.add (TFL::TambahkanOp)

Operator tambahan

Operasi penjumlahan berdasarkan elemen.

Sifat: ::mlir::OpTrait::TFLRuntimeOpTrait , AlwaysSpeculatableImplTrait , Commutative , QuantizableResult , ResultsBroadcastableShape

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflArithmeticCountOpInterface , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.add_n (TFL::TambahkanNOp)

_Tambahkan n operator

Menambahkan semua tensor masukan berdasarkan elemen.

Ciri-ciri: AlwaysSpeculatableImplTrait , Commutative

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflArithmeticCountOpInterface , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.arg_max (TFL::ArgMaxOp)

Operator ArgMax

Mengembalikan indeks dengan nilai terbesar di seluruh dimensi tensor.

Sifat: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.arg_min (TFL::ArgMinOp)

Operator ArgMin

Mengembalikan indeks dengan nilai terkecil di seluruh dimensi tensor. a = [1, 10, 26.9, 2.8, 166.32, 62.3] b = tf.math.argmin(input = a) c = tf.keras.backend.eval(b)

Sifat: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.assign_variable (TFL::AssignVariableOp)

Menetapkan nilai baru ke variabel.

ReadVariableOp apa pun dengan ketergantungan kontrol pada operasi ini dijamin akan mengembalikan nilai ini atau nilai variabel berikutnya yang lebih baru.

Antarmuka: TflRuntimeVerifyOpInterface

Operan:

tfl.atan2 (TFL::Atan2Op)

Operasi Atan2

Operasi "atan2" menghitung tangen busur dari elemen y/x, dengan memperhatikan tanda-tanda argumen.

Sifat: AlwaysSpeculatableImplTrait , SameOperandsAndResultElementType , SameOperandsAndResultShape

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.average_pool_2d (TFL::AveragePool2DOp)

_Rata-rata_kumpulan operator 2 hari

Melakukan operasi pengumpulan rata-rata pada input.

Sifat: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflArithmeticCountOpInterface , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.basic_lstm (TFL::BasicLSTMOp)

Operator lstm dasar

Operator Seluler LSTM dasar.

Sifat: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.batch_matmul (TFL::BatchMatMulOp)

Operator Penggandaan Matriks Batch

Melakukan perkalian matriks batch pada input. Mengikuti konvensi TensorFlow BatchMatMulV2, dengan dukungan untuk dimensi yang tidak diketahui dalam dimensi batch dan penyiaran.

Inputs:
  `inputs[0]`: required: input LHS
  `inputs[1]`: required: input RHS
  `adjoint_lhs`: optional: Transpose LHS (default false)
  `adjoint_rhs`: optional: Transpose RHS (default false)

Sifat: ::mlir::OpTrait::TFLRuntimeOpTrait , AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , DynamicRangeQuantizedOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.batch_to_space_nd (TFL::BatchToSpaceNdOp)

Operator BatchToSpaceNd

Operasi ini membentuk ulang dimensi "batch" 0 menjadi dimensi ruang.

Sifat: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.bidirectional_sequence_lstm (TFL::BidirectSequenceLSTMOp)

Operator lstm urutan dua arah

Lstm dua arah pada dasarnya adalah dua lstm, satu berjalan maju & yang lainnya berjalan mundur. Dan outputnya adalah gabungan dari dua lstm.

Ciri-ciri: QuantizableResult

Antarmuka: DynamicRangeQuantizedOpInterface , TFL_StatefulOp , TflRuntimeVerifyOpInterface

Atribut:

Operan:

Hasil:

tfl.bitcast (TFL::BitcastOp)

Operator Bitcast

Mem-bitcast tensor dari satu jenis ke jenis lainnya.

Ciri-ciri: AlwaysSpeculatableImplTrait

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.bitwise_xor (TFL::BitwiseXorOp)

Operator Xor bitwise

Elementwise menghitung XOR bitwise dari lhs dan rhs .

Sifat: AlwaysSpeculatableImplTrait , Commutative , ResultsBroadcastableShape , SameOperandsAndResultElementType

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.broadcast_args (TFL::BroadcastArgsOp)

Kembalikan bentuk s0 op s1 dengan siaran.

Mengingat s0 dan s1 , tensor yang merepresentasikan bentuk, hitung r0 , bentuk yang disiarkan. s0 , s1 dan r0 semuanya adalah vektor bilangan bulat.

Ciri-ciri: AlwaysSpeculatableImplTrait

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.broadcast_to (TFL::BroadcastToOp)

Siarkan array untuk bentuk yang kompatibel.

Penyiaran adalah proses membuat array memiliki bentuk yang kompatibel untuk operasi aritmatika. Dua bentuk dikatakan kompatibel jika untuk setiap pasangan dimensi keduanya sama atau salah satunya adalah satu. Saat mencoba menyiarkan Tensor ke suatu bentuk, Tensor dimulai dengan dimensi tambahan, dan terus berlanjut.

Misalnya,

x = tf.constant([1, 2, 3]) y = tf.broadcast_to(x, [3, 3]) print(y) tf.Tensor( [[1 2 3] [1 2 3] [1 2 3]], bentuk=(3, 3), tiped=int32)

Pada contoh di atas, Tensor masukan berbentuk [1, 3] disiarkan ke Tensor keluaran berbentuk [3, 3] .

Saat melakukan operasi penyiaran seperti mengalikan tensor dengan skalar, penyiaran (biasanya) memberikan manfaat waktu atau ruang, karena tensor yang disiarkan tidak pernah terwujud.

Namun, broadcast_to tidak memberikan manfaat apa pun. Tensor yang baru dibuat mengambil memori penuh dari bentuk yang disiarkan. (Namun, dalam konteks grafik, broadcast_to mungkin digabungkan ke operasi berikutnya dan kemudian dioptimalkan.)

Sifat: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.bucketize (TFL::BucketizeOp)

Mengelompokkan 'masukan' berdasarkan 'batas'.

Contoh:

Jika masukannya adalah boundaries = [0, 10, 100] dan input = [[-5, 10000][150, 10][5, 100]] , maka keluarannya adalah output = [[0, 3][3, 2][1, 3]] .

Ciri-ciri: AlwaysSpeculatableImplTrait , SameOperandsAndResultShape

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.call_once (TFL::CallOnceOp)

Memanggil fungsi inisialisasi

Operasi ini memanggil fungsi inisialisasi yang diberikan untuk penginisialisasi sesi dalam dialek model yang disimpan tf.

Antarmuka: TflRuntimeVerifyOpInterface

Atribut:

tfl.cast (TFL::CastOp)

Operator pemeran

Melemparkan input dari tipe input ke tipe output.

Ciri-ciri: AlwaysSpeculatableImplTrait , SameOperandsAndResultShape

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.ceil (TFL::CeilOp)

Operator langit-langit

Mengembalikan nilai ceil berdasarkan elemen dari input.

Sifat: AlwaysSpeculatableImplTrait , InferTensorType , TF::SameOperandsAndResultTypeResolveRef

Antarmuka: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.complex_abs (TFL::ComplexAbsOp)

Menghitung nilai absolut kompleks dari sebuah tensor.

Diberikan tensor x bilangan kompleks, operasi ini mengembalikan tensor bertipe float atau double yang merupakan nilai absolut setiap elemen dalam x . Semua elemen dalam x harus berbentuk bilangan kompleks \(a + bj\). Nilai absolut dihitung sebagai \( \sqrt{a^2 + b^2}\).

Ciri-ciri: AlwaysSpeculatableImplTrait , SameOperandsAndResultShape

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.concatenation (TFL::PenggabunganOp)

Operator penggabungan

Menggabungkan tensor sepanjang satu dimensi

Sifat: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.control_node (TFL::ControlNodeOp)

Operasi TFL.control_node membungkus operasi blok tunggal untuk melampirkan tepi kontrol.

Ini digunakan untuk menggabungkan wilayah dan melampirkan dependensi kontrol ke wilayah tersebut. Biasanya, hal ini akan terjadi pada salah satu langkah terakhir sebelum memancarkan model flatbuffer untuk mengaktifkan pengoptimalan yang mengandalkan urutan operasi tetap (seperti rematerialisasi.) Eksportir flatbuffer akan membuka bungkus wilayah yang dibungkus dan memberi anotasi pada model yang dihasilkan dengan metadata sedemikian rupa sehingga setiap pemesanan ulang runtime akan mengikuti urutan yang diberikan oleh dependensi kontrol.

Sifat: HasParent<mlir::func::FuncOp> , RecursiveMemoryEffects , SingleBlockImplicitTerminator<YieldOp> , SingleBlock

Operan:

Hasil:

tfl.conv_2d (TFL::Konv2DOp)

Operator konvolusi

Melakukan operasi konvolusi pada input.

Input: inputs[0] : diperlukan: tensor aktivasi input inputs[1] : diperlukan: inputs[2] : opsional: tensor bias

Sifat: AlwaysSpeculatableImplTrait , QuantizableResult , quant::AccumulatorUniformScale<2, 0, 1> , quant::AffineOpCoefficient<0, 1>

Antarmuka: AffineQuantizedOpInterface , ConditionallySpeculatable , DynamicRangeQuantizedOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , TFL_SparseOp , TflArithmeticCountOpInterface , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.conv_3d (TFL::Konv3DOp)

Operator 3D konvolusi

Melakukan operasi konvolusi pada input 3D. Input: inputs[0] : diperlukan: tensor aktivasi input inputs[1] : diperlukan: inputs[2] : opsional: tensor bias

Sifat: AlwaysSpeculatableImplTrait , quant::AccumulatorUniformScale<2, 0, 1>

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.conv_3d_transpose (TFL::Conv3DTransposeOp)

Operator 3D Konvolusi yang Dialihkan

Melakukan operasi konvolusi yang dialihkan pada input 3D. Inputs: inputs[0] : diperlukan: bentuk tensor keluaran inputs[1] : diperlukan: filter tensor bobot inputs[2] : diperlukan: aktivasi masukan tensor inputs[3] : opsional: tensor bias

Sifat: AlwaysSpeculatableImplTrait , quant::AccumulatorUniformScale<2, 0, 1>

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.cos (TFL::CosOp)

Operator kosinus

Menghitung kosinus input berdasarkan elemen

Sifat: AlwaysSpeculatableImplTrait , InferTensorType , TF::SameOperandsAndResultTypeResolveRef

Antarmuka: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.cumsum (TFL::CumsumOp)

Operator cumsum

Hitung jumlah kumulatif tensor x sepanjang sumbu.

Ciri-ciri: AlwaysSpeculatableImplTrait

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.custom (TFL::CustomOp)

Operasi khusus

Operasi umum untuk operasi kustom TFLite apa pun.

input: Daftar input dalam operasi asli. custom_code: String yang digunakan untuk mengidentifikasi operasi mana yang sebenarnya, yang sesuai dengan operator_codes.custom_code di flatbuffer. custom_option: pemegang untuk menyimpan atribut op dalam mode byte. keluaran: Daftar keluaran dalam operasi asli.

Antarmuka: TflRuntimeVerifyOpInterface

Atribut:

Operan:

Hasil:

tfl.custom_tf (TFL::CustomTfOp)

Operasi Pembungkus untuk operasi khusus TF.

Operasi pembungkus di sekitar operasi TF Kustom apa pun. Ini termasuk operasi yang ditentukan menggunakan custom_opdefs atau tertaut yang tidak ditentukan dalam dialek TF. Operasi ini hanya menggabungkan operasi khusus di dalam suatu wilayah. Catatan #1, Operasi ini tidak akan menyertakan operasi kustom TF Lite yang ditentukan menggunakan CustomOp. Catatan #2, operasi ini hanyalah representasi internal di dalam konverter dan tidak diekspos/diekspor saat model diekspor ke Flatbuffer.

Sifat: IsolatedFromAbove , RecursiveMemoryEffects , SingleBlockImplicitTerminator<YieldOp> , SingleBlock

Antarmuka: InferTypeOpInterface , TflRuntimeVerifyOpInterface

Operan:

Hasil:

tfl.densify (TFL::DensifyOp)

Padatkan operator

Mengonversi tensor renggang menjadi format padat.

Ciri-ciri: AlwaysSpeculatableImplTrait

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.depth_to_space (TFL::DepthToSpaceOp)

Operator DepthToSpace

Menyusun ulang data dari kedalaman menjadi blok data spasial. Ini adalah transformasi kebalikan dari SpaceToDepth. Lebih khusus lagi, operasi ini menghasilkan salinan tensor masukan di mana nilai dari dimensi depth dipindahkan dalam blok spasial ke dimensi height dan width . attr block_size menunjukkan ukuran blok masukan dan cara data dipindahkan.

Sifat: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.depthwise_conv_2d (TFL::DepthwiseConv2DOp)

Operator konvolusi yang dapat dipisahkan secara mendalam

Melakukan operasi konvolusi pada input.

Input: inputs[0] : diperlukan: tensor aktivasi input inputs[1] : diperlukan: inputs[2] : opsional: tensor bias

Sifat: AlwaysSpeculatableImplTrait , QuantizableResult , quant::AccumulatorUniformScale<2, 0, 1> , quant::AffineOpCoefficient<3, 1>

Antarmuka: AffineQuantizedOpInterface , ConditionallySpeculatable , DynamicRangeQuantizedOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , TFL_SparseOp , TflArithmeticCountOpInterface , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.dequantize (TFL::DequantizeOp)

Dekuantisasi operator

Mengonversi array bilangan bulat terkuantisasi menjadi floating-point sesuai dengan parameter kuantisasi.

Antarmuka: NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.dilate (TFL::DilatOp)

Operator pelebaran

Memperluas tensor dengan menambahkan elemen baru di antara elemen yang sudah ada.

Ciri-ciri: AlwaysSpeculatableImplTrait

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.div (TFL::DivOp)

Operator divisi

Operasi pembagian berdasarkan elemen.

Sifat: ::mlir::OpTrait::TFLRuntimeOpTrait , AlwaysSpeculatableImplTrait , ResultsBroadcastableShape

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflArithmeticCountOpInterface , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.dynamic_update_slice (TFL::DynamicUpdateSliceOp)

Irisan Pembaruan Dinamis.

Operasi DynamicUpdateSlice yang memiliki semantik yang sama dengan XLA DynamicUpdateSlice. Menghasilkan hasil yang merupakan nilai operan larik masukan, dengan pembaruan irisan ditimpa di start_indices.

Lihat https://www.tensorflow.org/xla/operation_semantics#dynamicupdateslice

Ciri-ciri: AlwaysSpeculatableImplTrait

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.elu (tfl :: eluop)

Operator Unit Linear Eksponensial

Menghitung linear eksponensial F (x) -> exp (x) -1 untuk x <0, x untuk x> = 0. Elemen -bijaksana.

Ciri -ciri: AlwaysSpeculatableImplTrait , SameOperandsAndResultShape

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.embedding_lookup (tfl :: embeddinglookupop)

Operator pencarian yang menanamkan

Mencari ID dalam daftar tensor yang menanamkan.

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: AffineQuantizedOpInterface , ConditionallySpeculatable , DynamicRangeQuantizedOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.equal (tfl :: equalop)

Operator yang sama

Mengembalikan elemen kebenaran x == y elemen bijaksana

Ciri -ciri: ::mlir::OpTrait::TFLRuntimeOpTrait ResultsBroadcastableShape AlwaysSpeculatableImplTrait , Commutative , QuantizableResult ,

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.exp (tfl :: expop)

Operator eksponensial alami

Melakukan operasi eksponensial alami yang bijaksana pada input.

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.expand_dims (tfl :: expanddimsop)

Memasukkan dimensi 1 ke dalam bentuk tensor.

Diberi input tensor, operasi ini memasukkan dimensi 1 pada axis indeks dimensi bentuk input . axis indeks dimensi dimulai pada nol; Jika Anda menentukan angka negatif untuk axis itu dihitung mundur dari ujungnya.

Operasi ini berguna jika Anda ingin menambahkan dimensi batch ke satu elemen. Misalnya, jika Anda memiliki satu gambar bentuk [height, width, channels] , Anda dapat menjadikannya batch 1 gambar dengan expand_dims(image, 0) , yang akan membuat bentuk [1, height, width, channels] .

Contoh lain:

# 't' is a tensor of shape [2]
shape(expand_dims(t, 0)) ==> [1, 2]
shape(expand_dims(t, 1)) ==> [2, 1]
shape(expand_dims(t, -1)) ==> [2, 1]

# 't2' is a tensor of shape [2, 3, 5]
shape(expand_dims(t2, 0)) ==> [1, 2, 3, 5]
shape(expand_dims(t2, 2)) ==> [2, 3, 1, 5]
shape(expand_dims(t2, 3)) ==> [2, 3, 5, 1]

Operasi ini mensyaratkan itu:

-1-input.dims() <= dim <= input.dims()

Operasi ini terkait dengan squeeze() , yang menghilangkan dimensi ukuran 1.

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.external_const (tfl :: externalconstop)

Op Konstitusi Eksternal.

Eksternal const op memegang buffer_index yang menunjuk ke konstan di flatbuffer.

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Hasil:

tfl.fake_quant (tfl :: fakequantop)

Operator palsu

Palsu-quatize tensor 'input' tipe float melalui float skalar min dan max ke 'output' tensor dengan bentuk yang sama dengan input.

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.fill (tfl :: fillop)

Isi tensor dengan nilai yang diberikan.

Isi tensor dengan nilai yang diberikan.

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.floor (tfl :: floatop)

Operator lantai

Mengembalikan nilai lantai input yang bijaksana.

Ciri -ciri: AlwaysSpeculatableImplTrait , InferTensorType , TF::SameOperandsAndResultTypeResolveRef

Antarmuka: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.floor_div (tfl :: floordivop)

Operator Div Lantai

Operasi Div Lantai-Wise-Wise.

Ciri -ciri: ::mlir::OpTrait::TFLRuntimeOpTrait ResultsBroadcastableShape AlwaysSpeculatableImplTrait

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.floor_mod (tfl :: floorModop)

Pengingat Divisi

Operasi Pengingat Divisi Elemen.

Ciri -ciri: ::mlir::OpTrait::TFLRuntimeOpTrait ResultsBroadcastableShape AlwaysSpeculatableImplTrait

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.fully_connected (tfl :: fullConnectedop)

OP yang sepenuhnya terhubung

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult , quant::AccumulatorUniformScale<2, 0, 1> , quant::AffineOpCoefficient<0, 1>

Antarmuka: AffineQuantizedOpInterface , ConditionallySpeculatable , DynamicRangeQuantizedOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , TFL_SparseOp , TflArithmeticCountOpInterface , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.gather (tfl :: gatherop)

Kumpulkan operator

Kumpulkan irisan dari axis poros params sesuai dengan indices .

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , DynamicRangeQuantizedOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.gather_nd (tfl :: callingndop)

_Gather ND Operator

Kumpulkan irisan dari params menjadi tensor dengan bentuk yang ditentukan oleh indices .

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.gelu (tfl :: geluop)

Fungsi Aktivasi Gelu.

Menghitung elemen fungsi aktivasi gelu-bijaksana.

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult , SameOperandsAndResultShape

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.greater (tfl :: greatop)

Operator yang lebih besar

Operasi yang lebih besar dari elemen.

Ciri -ciri: ::mlir::OpTrait::TFLRuntimeOpTrait AlwaysSpeculatableImplTrait ResultsBroadcastableShape QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.greater_equal (tfl :: greaterequalop)

_ Operator yang sama

Operasi elemen-bijaksana Greater_Equal.

Ciri -ciri: ::mlir::OpTrait::TFLRuntimeOpTrait AlwaysSpeculatableImplTrait ResultsBroadcastableShape QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.hard_swish (tfl :: hardswishop)

Fungsi aktivasi hardswish.

Menghitung fungsi aktivasi keras-Swish f (x)-> (x * relu6 (x+3))/6 elemen-bijaksana.

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult , SameOperandsAndResultShape

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.hashtable (tfl :: hashtableop)

Membuat tabel hash yang tidak diinisialisasi.

OP ini membuat tabel hash, menentukan jenis kunci dan nilainya. Sebelum menggunakan tabel, Anda harus menginisialisasi. Setelah inisialisasi tabel akan tidak berubah.

Antarmuka: TflRuntimeVerifyOpInterface

Atribut:

Hasil:

tfl.hashtable_find (tfl :: hashtablefindop)

Mencari kunci dalam tabel, mengeluarkan nilai yang sesuai.

keys tensor harus dari jenis yang sama dengan kunci tabel. values output adalah jenis nilai tabel.

Scalar default_value adalah output nilai untuk kunci yang tidak ada dalam tabel. Itu juga harus memiliki tipe yang sama dengan nilai tabel.

Antarmuka: TflRuntimeVerifyOpInterface

Operan:

Hasil:

tfl.hashtable_import (tfl :: hashtableimportop)

Menggantikan isi tabel dengan kunci dan nilai yang ditentukan.

keys tensor harus memiliki jenis yang sama dengan kunci tabel. values tensor harus dari jenis nilai tabel.

Antarmuka: TflRuntimeVerifyOpInterface

Operan:

tfl.hashtable_size (tfl :: hashtablessizeop)

Menghitung jumlah elemen dalam tabel yang diberikan.

Antarmuka: TflRuntimeVerifyOpInterface

Operan:

Hasil:

tfl.if (tfl :: ifop)

Operasi If-Then-Else

Operasi tfl.if mewakili konstruk jika-kemudian-else untuk mengeksekusi secara kondisional dua wilayah kode. Operan ke operasi IF adalah nilai boolean. Misalnya:

tfl.if %b  {
  ...
} else {
  ...
}

tfl.if juga dapat mengembalikan hasil yang didefinisikan di daerahnya. Nilai -nilai yang ditentukan ditentukan oleh jalur eksekusi yang diambil.

Contoh:

%x, %y = tfl.if %b -> (tensor<f32>, tensor<f32>) {
  %x_true = ...
  %y_true = ...
  tfl.yield %x_true, %y_true : tensor<f32>, tensor<f32>
} else {
  %x_false = ...
  %y_false = ...
  tfl.yield %x_false, %y_false : tensor<f32>, tensor<f32>
}

tfl.if daerah selalu diakhiri dengan "tfl.yield". Jika "tfl.if" tidak mendefinisikan nilai, "tfl.yield" dapat ditinggalkan, dan akan dimasukkan secara implisit. Kalau tidak, itu harus eksplisit. Juga, jika "tfl.jika" mendefinisikan satu atau lebih nilai, blok 'lain' tidak dapat dihilangkan.

Contoh:

tfl.if %b  {
  ...
}

Ciri -ciri: NoRegionArguments , RecursiveMemoryEffects , SingleBlockImplicitTerminator<YieldOp> , SingleBlock

Antarmuka: RegionBranchOpInterface , TflRuntimeVerifyOpInterface

Operan:

Hasil:

tfl.imag (tfl :: imagop)

Mengembalikan bagian imajiner dari bilangan kompleks.

Dengan input tensor bilangan kompleks, operasi ini mengembalikan tensor tipe float yang merupakan bagian imajiner dari setiap elemen dalam input . Semua elemen dalam input harus berupa bilangan formulir yang kompleks \(a + bj\), di mana A adalah bagian nyata dan B adalah bagian imajiner yang dikembalikan oleh operasi ini.

Ciri -ciri: AlwaysSpeculatableImplTrait , SameOperandsAndResultShape

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.l2_normalization (tfl :: l2normalizationop)

L2 menormalkan operator

L2normalisasi op

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , FixedOutputRangeInterface , NoMemoryEffect (MemoryEffectOpInterface) , TflArithmeticCountOpInterface , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.leaky_relu (tfl :: leakyreluop)

Operator Relu Bocor

Operator Relu Bocor Elemen -bijaksana X -> X> = 0? x: (alpha * x)

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult , SameOperandsAndResultShape

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.less (tfl :: lessop)

Lebih sedikit operator

Elemen-bijaksana kurang operasi.

Ciri -ciri: ::mlir::OpTrait::TFLRuntimeOpTrait AlwaysSpeculatableImplTrait ResultsBroadcastableShape QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.less_equal (tfl :: lessequalop)

_Less operator yang sama

Operasi Less_Equal elemen-bijaksana.

Ciri -ciri: ::mlir::OpTrait::TFLRuntimeOpTrait AlwaysSpeculatableImplTrait ResultsBroadcastableShape QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.local_response_normalization (tfl :: localresponsenormalizationop)

Normalisasi respons lokal.

Tensor input 4-D diperlakukan sebagai array 3-D vektor 1-D (sepanjang dimensi terakhir), dan setiap vektor dinormalisasi secara independen. Dalam vektor yang diberikan, setiap komponen dibagi dengan jumlah input yang tertimbang, kuadrat dalam depth_radius . Secara terperinci,

sqr_sum[a, b, c, d] =
    sum(input[a, b, c, d - depth_radius : d + depth_radius + 1] ** 2)
output = input / (bias + alpha * sqr_sum) ** beta

Untuk detailnya, lihat Krizhevsky et al., Klasifikasi Imagenet dengan jaringan saraf konvolusional yang mendalam (NIPS 2012) .

Ciri -ciri: AlwaysSpeculatableImplTrait , InferTensorType , TF::SameOperandsAndResultTypeResolveRef

Antarmuka: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.log (tfl :: logop)

Operator logaritma alami

Melakukan operasi logaritma alami yang bijaksana pada input.

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.log_softmax (tfl :: logsoftmaxop)

Operator Log Softmax

Menghitung aktivasi log softmax log-wise dengan rumus berikut

Input - Log (redike_sum (exp (input), DIM))

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult , SameOperandsAndResultShape

Antarmuka: ConditionallySpeculatable , FixedOutputRangeInterface , NoMemoryEffect (MemoryEffectOpInterface) , TflArithmeticCountOpInterface , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.logical_and (tfl :: LogicalAndop)

Logis dan operator

Logis dan Operasi Elemen-bijaksana.

Ciri -ciri: AlwaysSpeculatableImplTrait , ResultsBroadcastableShape

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.logical_not (tfl :: LogicalNotop)

Logis bukan operator

Elemen-bijaksana logis bukan operasi.

Ciri -ciri: AlwaysSpeculatableImplTrait , SameOperandsAndResultShape

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.logical_or (TFL :: Logicalorop)

Logis atau operator

Logis atau operasi elemen.

Ciri -ciri: AlwaysSpeculatableImplTrait , ResultsBroadcastableShape

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.logistic (tfl :: logisticop)

Operator logistik

Menghitung input elemen bijaksana

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult , SameOperandsAndResultShape

Antarmuka: ConditionallySpeculatable , FixedOutputRangeInterface , NoMemoryEffect (MemoryEffectOpInterface) , TflArithmeticCountOpInterface , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.lstm (tfl :: lstmop)

Operator LSTM lengkap

Lapisan Jaringan Unit Berulang Lama (LSTM) Berulang. Implementasi non-peephole default didasarkan pada: http://deeplearning.cs.cmu.edu/pdfs/hochreiter97_lstm.pdf S. Hochreiter dan J. Schmidhuber. 'Memori jangka pendek yang panjang'. Komputasi Saraf, 9 (8): 1735-1780, 1997. Implementasi Peephole didasarkan pada: https://research.google.com/pubs/archive/43905.pdf Hasim Sak, Andrew Senior, dan Francoise Beaufays. 'Arsitektur jaringan saraf berulang memori jangka pendek untuk pemodelan akustik skala besar.' Interspeech, 2014. Kopling input dan lupa gerbang (CIFG) didasarkan pada: http://arxiv.org/pdf/1503.04069.pdf Greff et al. 'LSTM: ruang pencarian Odyssey' Normalisasi lapisan didasarkan pada: https://arxiv.org/pdf/1607.06450.pdf Ba et al. 'Normalisasi lapisan'

Ciri -ciri: QuantizableResult

Antarmuka: DynamicRangeQuantizedOpInterface , TFL_StatefulOp , TflRuntimeVerifyOpInterface

Atribut:

Operan:

Hasil:

tfl.matrix_diag (tfl :: matrixdiagop)

Mengembalikan tensor dengan diagonal yang disediakan dan yang lainnya empuk dengan nol.

Diberi diagonal, mengembalikan tensor dengan diagonal dan yang lainnya empuk dengan nol. Asumsikan diagonal memiliki dimensi k [I, J, K, ..., N] , maka outputnya adalah tensor peringkat k+1 dengan dimensi [I, J, K, ..., N, N] di mana: output[i, j, k, ..., m, n] = 1{m=n} * diagonal[i, j, k, ..., n].

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.matrix_set_diag (tfl :: matrixsetDiagop)

Mengembalikan tensor matriks batch dengan nilai diagonal batched baru.

Diberi input dan diagonal , operasi ini mengembalikan tensor dengan bentuk dan nilai yang sama seperti input , kecuali untuk diagonal utama dari matriks terdalam. Ini akan ditimpa oleh nilai -nilai di diagonal .

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.max_pool_2d (tfl :: maxpool2dop)

Max Pool 2D OP

Melakukan Max Pool 2D pada Input.

Input: inputs[0] : Diperlukan: Input Tensor

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflArithmeticCountOpInterface , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.maximum (tfl :: maksimum)

Operator Max

Operasi maksimum elemen-bijaksana.

Ciri -ciri: ::mlir::OpTrait::TFLRuntimeOpTrait ResultsBroadcastableShape AlwaysSpeculatableImplTrait , Commutative , QuantizableResult ,

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.mean (tfl :: meanop)

Operator berarti

Menghitung rata -rata elemen lintas dimensi tensor. Mengurangi input_tensor di sepanjang dimensi yang diberikan dalam sumbu. Kecuali KeepDims benar, peringkat tensor dikurangi dengan 1 untuk setiap entri dalam sumbu. Jika KeepDims benar, dimensi yang dikurangi dipertahankan dengan panjang 1.

Ciri -ciri: AlwaysSpeculatableImplTrait , QuantizableResult

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Atribut:

Operan:

Hasil:

tfl.minimum (tfl :: minimumop)

Operator min

Operasi Min-Wise Min.

Ciri -ciri: ::mlir::OpTrait::TFLRuntimeOpTrait ResultsBroadcastableShape AlwaysSpeculatableImplTrait , Commutative , QuantizableResult ,

Antarmuka: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Efek: MemoryEffects::Effect{}

Operan:

Hasil:

tfl.mirror_pad (tfl :: mirrorpadop)

Operator Mirrorpad. Bantalan tensor dengan nilai cermin.

Operasi ini menjadi input dengan nilai cermin sesuai dengan padding yang Anda tentukan. Paddings adalah tensor integer dengan bentuk [n, 2], di mana n adalah peringkat input. Untuk setiap dimensi D input, padding [d, 0] menunjukkan berapa banyak nilai yang akan ditambahkan sebelum konten input dalam dimensi itu, dan padding [d, 1] menunjukkan berapa banyak nilai yang ditambahkan setelah konten input dalam dimensi tersebut.

Kedua padding [d, 0] dan padding [d, 1] tidak boleh lebih besar dari input.dim_size (d) (atau input.dim_size (d) - 1) jika copy_border adalah benar (masing -masing salah).

Ukuran empuk setiap dimensi D output adalah:

paddings(D, 0) + input.dim_size(D) + paddings(D, 1)

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Atribut:

Operands:

Hasil:

tfl.mul (TFL::MulOp)

Multiplication operator

Element-wise multiplication operation.

Traits: ::mlir::OpTrait::TFLRuntimeOpTrait , AlwaysSpeculatableImplTrait , Commutative , QuantizableResult , ResultsBroadcastableShape

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflArithmeticCountOpInterface , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Atribut:

Operands:

Hasil:

tfl.multinomial (TFL::MultinomialOp)

Draws samples from a categorical distribution.

The generated values will have a categorical distribution based on the logits or unnormalized log-probabilities provided for all classes.

Interfaces: TflRuntimeVerifyOpInterface

Atribut:

Operands:

Hasil:

tfl.neg (TFL::NegOp)

Negation operator

Computes element-wise negation of input

Traits: AlwaysSpeculatableImplTrait , InferTensorType , TF::SameOperandsAndResultTypeResolveRef

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.no_value (TFL::NoValueOp)

Constant representing no value.

No value constant op.

Traits: AlwaysSpeculatableImplTrait , ConstantLike

Interfaces: ConditionallySpeculatable , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Atribut:

Hasil:

tfl.non_max_suppression_v4 (TFL::NonMaxSuppressionV4Op)

Greedily selects a subset of bounding boxes in descending order of score,

pruning away boxes that have high intersection-over-union (IOU) overlap with previously selected boxes. Bounding boxes with score less than score_threshold are removed. Bounding boxes are supplied as [y1, x1, y2, x2], where (y1, x1) and (y2, x2) are the coordinates of any diagonal pair of box corners and the coordinates can be provided as normalized (ie, lying in the interval [0, 1]) or absolute. Note that this algorithm is agnostic to where the origin is in the coordinate system and more generally is invariant to orthogonal transformations and translations of the coordinate system; thus translating or reflections of the coordinate system result in the same boxes being selected by the algorithm. The output of this operation is a set of integers indexing into the input collection of bounding boxes representing the selected boxes. The bounding box coordinates corresponding to the selected indices can then be obtained using the tf.gather operation . For example: selected_indices = tf.image.non_max_suppression_v2( boxes, scores, max_output_size, iou_threshold, score_threshold) selected_boxes = tf.gather(boxes, selected_indices)

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.non_max_suppression_v5 (TFL::NonMaxSuppressionV5Op)

Greedily selects a subset of bounding boxes in descending order of score,

pruning away boxes that have high intersection-over-union (IOU) overlap with previously selected boxes. Bounding boxes with score less than score_threshold are removed. Bounding boxes are supplied as [y1, x1, y2, x2], where (y1, x1) and (y2, x2) are the coordinates of any diagonal pair of box corners and the coordinates can be provided as normalized (ie, lying in the interval [0, 1]) or absolute. Note that this algorithm is agnostic to where the origin is in the coordinate system and more generally is invariant to orthogonal transformations and translations of the coordinate system; thus translating or reflections of the coordinate system result in the same boxes being selected by the algorithm. The output of this operation is a set of integers indexing into the input collection of bounding boxes representing the selected boxes. The bounding box coordinates corresponding to the selected indices can then be obtained using the tf.gather operation . For example: selected_indices = tf.image.non_max_suppression_v2( boxes, scores, max_output_size, iou_threshold, score_threshold) selected_boxes = tf.gather(boxes, selected_indices) This op also supports a Soft-NMS (with Gaussian weighting) mode (cf Bodla et al, https://arxiv.org/abs/1704.04503 ) where boxes reduce the score of other overlapping boxes instead of directly causing them to be pruned. To enable this Soft-NMS mode, set the soft_nms_sigma parameter to be larger than 0.

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.not_equal (TFL::NotEqualOp)

_Not equal operator

Element-wise not_equal operation.

Traits: ::mlir::OpTrait::TFLRuntimeOpTrait , AlwaysSpeculatableImplTrait , Commutative , ResultsBroadcastableShape

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.NumericVerify (TFL::NumericVerifyOp)

Verifies the numericals of the two operands

The NumericVerify op is a debugging op to verify the numericals of the two activations. It is a custom op in TFLite. If log_if_failed is true, the NumericVerify op calculates statistics on differences between float and quantized activations, output logs, set differences to the output tensors, and throws an error if errors above tolerance exist. If log_if_failed = false, then it doesn't care about errors.

Traits: QuantizableResult , SameOperandsShape

Interfaces: TflRuntimeVerifyOpInterface

Atribut:

Operands:

Hasil:

tfl.one_hot (TFL::OneHotOp)

OneHot operator

Returns a one-hot tensor.The locations represented by indices in indices take value on_value , while all other locations take value off_value .

If the input indices is rank N , the output will have rank N+1 , The new axis is created at dimension axis (default: the new axis is appended at the end).

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Atribut:

Operands:

Hasil:

tfl.pack (TFL::PackOp)

Packs a list of tensors along a dimension into one tensor

Packs a list of values_count rank- R tensors into one rank- (R+1) tensor.

Packs the values_count tensors in values into a tensor with rank one higher than each tensor in values , by packing them along the axis dimension.

Given a list of tensors of shape (A, B, C) ;

if axis == 0 then the output tensor will have the shape (N, A, B, C) . if axis == 1 then the output tensor will have the shape (A, N, B, C) . Dll.

Misalnya:

# 'x' is [1, 4]
# 'y' is [2, 5]
# 'z' is [3, 6]
pack([x, y, z]) => [[1, 4], [2, 5], [3, 6]]  # Pack along first dim.
pack([x, y, z], axis=1) => [[1, 2, 3], [4, 5, 6]]

This is the opposite of unpack .

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Atribut:

Operands:

Hasil:

tfl.pad (TFL::PadOp)

Padding operator

This operation pads a input with zeros according to the paddings you specify. paddings is an integer tensor with shape [Dn, 2] , where n is the rank of input . For each dimension D of input , paddings[D, 0] indicates how many zeros to add before the contents of input in that dimension, and paddings[D, 1] indicates how many zeros to add after the contents of input in that dimension.

The padded size of each dimension D of the output is:

paddings(D, 0) + input.dim_size(D) + paddings(D, 1)

Misalnya:

# 't' is [[1, 1], [2, 2]]
# 'paddings' is [[1, 1], [2, 2]]
# rank of 't' is 2
pad(t, paddings) ==> [[0, 0, 0, 0, 0, 0]
                      [0, 0, 1, 1, 0, 0]
                      [0, 0, 2, 2, 0, 0]
                      [0, 0, 0, 0, 0, 0]]

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.padv2 (TFL::PadV2Op)

Padding operator v2

This operation pads a input according to the paddings and constant_values you specify. paddings is an integer tensor with shape [Dn, 2] , where n is the rank of input . For each dimension D of input , paddings[D, 0] indicates how many zeros to add before the contents of input in that dimension, and paddings[D, 1] indicates how many zeros to add after the contents of input in that dimension. constant_values is a scalar tensor of the same type as input that indicates the value to use for padding input .

The padded size of each dimension D of the output is:

paddings(D, 0) + input.dim_size(D) + paddings(D, 1)

Misalnya:

# 't' is [[1, 1], [2, 2]]
# 'paddings' is [[1, 1], [2, 2]]
# rank of 't' is 2
pad(t, paddings) ==> [[0, 0, 0, 0, 0, 0]
                      [0, 0, 1, 1, 0, 0]
                      [0, 0, 2, 2, 0, 0]
                      [0, 0, 0, 0, 0, 0]]

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.poly_call (TFL::PolyCallOp)

Poly call

Have multiple function bodies for the same computation. This allows a program compiler/interpreter to choose one of the available options to execute the program based on which one is most suitable for the target backend.

input: A list of input tensors whose types are T. output: A list of output tensors whose types are T.

call: Multiple regions, each of which encapsulates the same semantic computation but in different forms.

Traits: SingleBlockImplicitTerminator<YieldOp> , SingleBlock

Interfaces: RegionBranchOpInterface

Operands:

Hasil:

tfl.pow (TFL::PowOp)

Power operator

Element-wise power operation.

Traits: ::mlir::OpTrait::TFLRuntimeOpTrait , AlwaysSpeculatableImplTrait , ResultsBroadcastableShape

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.prelu (TFL::PReluOp)

Parameterized Relu operator

Parameterized Relu operator x -> x >= 0 ? x : (alpha * x) where alpha is a trainable tensor. input and alpha should be the same size as input or be broadcastable.

Traits: ::mlir::OpTrait::TFLRuntimeOpTrait , AlwaysSpeculatableImplTrait , QuantizableResult , ResultsBroadcastableShape , quant::AffineOpCoefficient<-1, 1>

Interfaces: AffineQuantizedOpInterface , ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.pseudo_const (TFL::ConstOp)

Constant pseudo op.

Represents a constant value in TensorFlow Lite dialect. This is not an actual operation and it will be lowered to buffer instead.

The op is allowed to have all the same type of attributes as tf.Const does (eg, opaque TF attributes are allowed).

Traits: AlwaysSpeculatableImplTrait , ConstantLike , FirstAttrDerivedResultType , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Atribut:

Hasil:

tfl.pseudo_qconst (TFL::QConstOp)

Quantized constant pseudo op

Represents a quantized constant value in TensorFlow Lite dialect. This is not an actual operation and it will be lowered to buffer instead. The quantization parameters are stored as a type attribute in this constant.

Traits: AlwaysSpeculatableImplTrait , FirstAttrDerivedResultType

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface)

Effects: MemoryEffects::Effect{}

Atribut:

Hasil:

tfl.pseudo_sparse_const (TFL::SparseConstOp)

Sparse constant pseudo op.

Represents a sparse constant value in TensorFlow Lite dialect. This is not an actual operation and it will be lowered to buffer instead.

Traits: AlwaysSpeculatableImplTrait , FirstAttrDerivedResultType , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Atribut:

Hasil:

tfl.pseudo_sparse_qconst (TFL::SparseQConstOp)

Sparse quantized constant pseudo op

Represents a sparse quantized constant value in TensorFlow Lite dialect. This is not an actual operation and it will be lowered to buffer instead. The quantization parameters are stored as a type attribute in this constant.

Traits: AlwaysSpeculatableImplTrait , FirstAttrDerivedResultType

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Atribut:

Hasil:

tfl.quantize (TFL::QuantizeOp)

Quantize operator

Converts floating point tensors to quantized integer tensors according to the quantization parameters defined in the type attribute.

Traits: FirstAttrDerivedResultType , SameOperandsAndResultShape

Interfaces: NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Atribut:

Operands:

Hasil:

tfl.random_standard_normal (TFL::RandomStandardNormalOp)

Outputs random values from a normal distribution.

The generated values will have mean 0 and standard deviation 1.

Interfaces: TflRuntimeVerifyOpInterface

Atribut:

Operands:

Hasil:

tfl.random_uniform (TFL::RandomUniformOp)

Outputs random values from a uniform distribution.

The generated values follow a uniform distribution in the range [0, 1) . The lower bound 0 is included in the range, while the upper bound 1 is excluded.

Interfaces: TflRuntimeVerifyOpInterface

Atribut:

Operands:

Hasil:

tfl.range (TFL::RangeOp)

Range operator

Returns a 1D tensor defined by a sequence from start to limit with a given delta .

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.rank (TFL::RankOp)

Rank operator.

Returns the rank of a tensor.

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.read_variable (TFL::ReadVariableOp)

Reads variable value.

Read variable data identified by 'resource_id'.

Interfaces: TflRuntimeVerifyOpInterface

Operands:

Hasil:

tfl.real (TFL::RealOp)

Returns the real part of a complex number.

Given a tensor input of complex numbers, this operation returns a tensor of type float that is the real part of each element in input . All elements in input must be complex numbers of the form \(a + bj\), where a is the real part returned by this operation and b is the imaginary part.

Traits: AlwaysSpeculatableImplTrait , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.reduce_all (TFL::ReduceAllOp)

Computes the "logical and" of elements across dimensions of a tensor.

Reduces input along the dimensions given in axis . Unless keep_dims is true, the rank of the tensor is reduced by 1 for each entry in axis . If keep_dims is true, the reduced dimensions are retained with length 1.

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.reduce_any (TFL::ReduceAnyOp)

Computes the "logical or" of elements across dimensions of a tensor.

Reduces input along the dimensions given in axis . Unless keep_dims is true, the rank of the tensor is reduced by 1 for each entry in axis . If keep_dims is true, the reduced dimensions are retained with length 1.

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.reduce_max (TFL::ReduceMaxOp)

Max-reduction operator

Computes the max reduction along the specified axes

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.reduce_min (TFL::ReduceMinOp)

Min-reduction operator

Computes the min reduction along the specified axes

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.reduce_prod (TFL::ReduceProdOp)

Prod-reduction operator

Computes the product along the specified axes

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.relu (TFL::ReluOp)

Relu operator

Element-wise Relu operator x -> max(0, x)

Traits: AlwaysSpeculatableImplTrait , QuantizableResult , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.relu6 (TFL::Relu6Op)

Relu6 operator

Element-wise Relu6 operator x -> max(0, min(6, x))

Traits: AlwaysSpeculatableImplTrait , QuantizableResult , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.relu_0_to_1 (TFL::Relu0To1Op)

Relu0To1 operator

Element-wise Relu0To1 operator x -> max(0, min(1, x))

Traits: AlwaysSpeculatableImplTrait , QuantizableResult , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.relu_n1_to_1 (TFL::Relu1Op)

Relu1 operator

Element-wise Relu1 operator x -> max(-1, min(1, x))

Traits: AlwaysSpeculatableImplTrait , QuantizableResult , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.reshape (TFL::ReshapeOp)

Reshape operator

Produces a tensor with the same values but different static shape defined by the output type.

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.resize_bilinear (TFL::ResizeBilinearOp)

ResizeBilinear Op

Resize images to size using bilinear interpolation.

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.resize_nearest_neighbor (TFL::ResizeNearestNeighborOp)

ResizeNearestNeighbor Op

Resize images to size using nearest neighbor interpolation.

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.reverse_sequence (TFL::ReverseSequenceOp)

Reverses variable length slices.

This op first slices input along the dimension batch_dim , and for each slice i , reverses the first seq_lengths[i] elements along the dimension seq_dim .

The elements of seq_lengths must obey seq_lengths[i] <= input.dims[seq_dim] , and seq_lengths must be a vector of length input.dims[batch_dim] .

The output slice i along dimension batch_dim is then given by input slice i , with the first seq_lengths[i] slices along dimension seq_dim reversed.

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.reverse_v2 (TFL::ReverseV2Op)

ReverseV2 Operator

Reverses specific dimensions of a tensor.

Given a tensor, and a int32/int64 tensor axis representing the set of dimensions of tensor to reverse. This operation reverses each dimension i for which there exists j st axis[j] == i.

Args: tensor: A Tensor. Must be one of the following types: uint8, int8, int16, int32, int64, float32, bool Up to 8-D.

axis: A Tensor. Must be one of the following types: int32, int64. with only 1 element which is the axis index. TODO: Add support for multiple elements.

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.rfft2d (TFL::RFFT2dOp)

2D real-valued fast Fourier transform.

Computes the 2-dimensional discrete Fourier transform of a real-valued signal over the inner-most 2 dimensions of input .

Since the DFT of a real signal is Hermitian-symmetric, RFFT2D only returns the fft_length / 2 + 1 unique components of the FFT for the inner-most dimension of output : the zero-frequency term, followed by the fft_length / 2 positive-frequency ketentuan.

Along each axis RFFT2D is computed on, if fft_length is smaller than the corresponding dimension of input , the dimension is cropped. If it is larger, the dimension is padded with zeros.

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.right_shift (TFL::RightShiftOp)

Right Shift operator

Elementwise computes the bitwise right-shift of lhs by rhs .

Traits: AlwaysSpeculatableImplTrait , SameOperandsAndResultElementType

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.round (TFL::RoundOp)

Round operator

Rounds the values of a tensor to the nearest integer, element-wise.

Traits: AlwaysSpeculatableImplTrait , InferTensorType , TF::SameOperandsAndResultTypeResolveRef

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.rsqrt (TFL::RsqrtOp)

Reciprocal of square root operator

Computes element-wise reverse square root of input

Traits: AlwaysSpeculatableImplTrait , QuantizableResult , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.scatter_nd (TFL::ScatterNdOp)

_Scatter nd operator

Scatter updates into a new tensor according to indices

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.segment_sum (TFL::SegmentSumOp)

SegmentSum operator

Computes the sum along segments of a tensor.

Traits: AlwaysSpeculatableImplTrait

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.select (TFL::SelectOp)

Pilih operator

Select values of 'x' if the corresponding value of 'condition' is true or the value of 'y' if false. There are valid condition input sizes:

1. Either the same shape (in which case the select is elementwise), or
2. condition must be Rank 1 and match over the first dimension.

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.select_v2 (TFL::SelectV2Op)

SelectV2 operator

Select values of 'x' if the corresponding value of 'condition' is true or the value of 'y' if false. There are valid condition input sizes:

1. Either the same shape (in which case the select is elementwise), or
2. Broadcastable shapes between 'condition', 'x' and 'y'.

Traits: ::mlir::OpTrait::TFLRuntimeOpTrait , AlwaysSpeculatableImplTrait , QuantizableResult , ResultsBroadcastableShape

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.shape (TFL::ShapeOp)

Shape operator

Returns the shape of a tensor.

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.sign (TFL::SignOp)

Sign operation

Returns NaN if x is NaN, 0 if x is 0, -1 if x < 0 and 1 if x > 0.

Traits: AlwaysSpeculatableImplTrait , SameOperandsAndResultElementType , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.sin (TFL::SinOp)

Sine operator

Computes element-wise Sine of input

Traits: AlwaysSpeculatableImplTrait , InferTensorType , TF::SameOperandsAndResultTypeResolveRef

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.slice (TFL::SliceOp)

Return a slice from 'input'.

The output tensor is a tensor with dimensions described by 'size' whose values are extracted from 'input' starting at the offsets in 'begin'.

begin is zero-based; size is one-based. If size[i] is -1, all remaining elements in dimension i are included in the slice. In other words, this is equivalent to setting: size[i] = input.dim_size(i) - begin[i]

Requirements : 0 <= begin[i] <= begin[i] + size[i] <= Di for i in [0, n)

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.softmax (TFL::SoftmaxOp)

Softmax operator

Computes element-wise softmax activations with the following formula

exp(input) / tf.reduce_sum(exp(input * beta), dim)

Traits: AlwaysSpeculatableImplTrait , QuantizableResult , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , FixedOutputRangeInterface , NoMemoryEffect (MemoryEffectOpInterface) , TflArithmeticCountOpInterface , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.space_to_batch_nd (TFL::SpaceToBatchNdOp)

SpaceToBatchNd operator

This operation reshapes space dimensions into the "batch" dimension 0

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.space_to_depth (TFL::SpaceToDepthOp)

SpaceToDepth operator

Rearranges blocks of spatial data, into depth. More specifically, this op outputs a copy of the input tensor where values from the height and width dimensions are moved to the depth dimension. block_size indicates the input block size.

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.sparse_to_dense (TFL::SparseToDenseOp)

Converts a sparse representation into a dense tensor.

Builds an array dense with shape output_shape such that

# If sparse_indices is scalar
dense[i] = (i == sparse_indices ? sparse_values : default_value)

# If sparse_indices is a vector, then for each i
dense[sparse_indices[i]] = sparse_values[i]

# If sparse_indices is an n by d matrix, then for each i in [0, n)
dense[sparse_indices[i][0], ..., sparse_indices[i][d-1]] = sparse_values[i]

All other values in dense are set to default_value . If sparse_values is a scalar, all sparse indices are set to this single value.

Indices should be sorted in lexicographic order, and indices must not contain any repeats. If validate_indices is true, these properties are checked during execution.

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.split (TFL::SplitOp)

Splits a tensor into num_split tensors along one dimension.

Splits the value tensor along split_dim into a number of sub-tensors with same shape as the original one, except for split_dim . Same as tf.Split.

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.split_v (TFL::SplitVOp)

Splits a tensor into num_split tensors along one dimension.

Splits the value tensor along split_dim into a number of sub-tensors with same shape as the original one, except for split_dim . The grouping of the resultant sub-tensors is decided by size-splits . Same as tf.SplitV.

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.sqrt (TFL::SqrtOp)

Square root operator

Computes element-wise Square root of input

Traits: AlwaysSpeculatableImplTrait , InferTensorType , TF::SameOperandsAndResultTypeResolveRef

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.square (TFL::SquareOp)

Square operator

Computes element-wise Square of input

Traits: AlwaysSpeculatableImplTrait , InferTensorType , TF::SameOperandsAndResultTypeResolveRef

Interfaces: ConditionallySpeculatable , InferShapedTypeOpInterface , InferTypeOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.squared_difference (TFL::SquaredDifferenceOp)

Squared difference operator

Element-wise squared difference operation.

Traits: ::mlir::OpTrait::TFLRuntimeOpTrait , AlwaysSpeculatableImplTrait , QuantizableResult , ResultsBroadcastableShape

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.squeeze (TFL::SqueezeOp)

Removes dimensions of size 1 from the shape of a tensor.

Given a tensor input , this operation returns a tensor of the same type with all dimensions of size 1 removed. If you don't want to remove all size 1 dimensions, you can remove specific size 1 dimensions by specifying squeeze_dims .

Misalnya:

# 't' is a tensor of shape [1, 2, 1, 3, 1, 1]
shape(squeeze(t)) ==> [2, 3]

Or, to remove specific size 1 dimensions:

# 't' is a tensor of shape [1, 2, 1, 3, 1, 1]
shape(squeeze(t, [2, 4])) ==> [1, 2, 3, 1]

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.strided_slice (TFL::StridedSliceOp)

StridedSlice Op

Return a strided slice from input .

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.sub (TFL::SubOp)

Subtraction operator

Element-wise subtraction operation.

Traits: ::mlir::OpTrait::TFLRuntimeOpTrait , AlwaysSpeculatableImplTrait , QuantizableResult , ResultsBroadcastableShape

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflArithmeticCountOpInterface , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.sum (TFL::SumOp)

Sum operator

Computes the sum reduction along the specified axes

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Attributes:

Operands:

Hasil:

tfl.svdf (TFL::SVDFOp)

Single value decomposition filter operator

The SVDF op is a decomposition of a densely connected op into low rank filters. For details: https://research.google.com/pubs/pub43813.html https://arxiv.org/abs/1812.02802

Traits: QuantizableResult , quant::AccumulatorUniformScale<3, 2, 4>

Interfaces: DynamicRangeQuantizedOpInterface , TFL_StatefulOp , TflRuntimeVerifyOpInterface

Attributes:

Operands:

Hasil:

tfl.tanh (TFL::TanhOp)

Hyperbolic tangent operator

Computes element-wise Hyperbolic tangent of input

Traits: AlwaysSpeculatableImplTrait , QuantizableResult , SameOperandsAndResultShape

Interfaces: ConditionallySpeculatable , FixedOutputRangeInterface , NoMemoryEffect (MemoryEffectOpInterface) , TflArithmeticCountOpInterface , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.tile (TFL::TileOp)

Tile operator.

Constructs a tensor by tiling a given tensor.

This operation creates a new tensor by replicating input multiples times. The output tensor's i'th dimension has input.dims(i) * multiples[i] elements, and the values of input are replicated multiples[i] times along the 'i'th dimension. For example, tiling [abcd] by [2] produces [abcdabcd].

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Results:

tfl.topk_v2 (TFL::TopKV2Op)

TopK operator

Returns the top k largest element along each last dimensional slice of input and the indices of values within the last dimension of the input tensor.

Results are always sorted in the descending order.

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Hasil:

tfl.transpose (TFL::TransposeOp)

Transpose operator

Returns the Transpose of x

Traits: AlwaysSpeculatableImplTrait , QuantizableResult

Interfaces: ConditionallySpeculatable , NoMemoryEffect (MemoryEffectOpInterface) , SameOperandsAndResultsScale , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Operands:

Results:

tfl.transpose_conv (TFL::TransposeConvOp)

Transpose convolution operator

Performs transpose convolution operation on input.

Traits: AlwaysSpeculatableImplTrait , QuantizableResult , quant::AccumulatorUniformScale<3, 1, 2> , quant::AffineOpCoefficient<0, 1>

Interfaces: AffineQuantizedOpInterface , ConditionallySpeculatable , DynamicRangeQuantizedOpInterface , NoMemoryEffect (MemoryEffectOpInterface) , TFL_SparseOp , TflArithmeticCountOpInterface , TflRuntimeVerifyOpInterface

Effects: MemoryEffects::Effect{}

Atribut:

Operands: