TensorFlow.NET/src/TensorFlowNET.Core/Tensors/tensor_util.cs at master · feelsyt/TensorFlow.NET · GitHub

History

725 lines (654 loc) · 27.2 KB

Raw

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

512

513

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541

542

543

544

545

546

547

548

549

550

551

552

553

554

555

556

557

558

559

560

561

562

563

564

565

566

567

568

569

570

571

572

573

574

575

576

577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

594

595

596

597

598

599

600

601

602

603

604

605

606

607

608

609

610

611

612

613

614

615

616

617

618

619

620

621

622

623

624

625

626

627

628

629

630

631

632

633

634

635

636

637

638

639

640

641

642

643

644

645

646

647

648

649

650

651

652

653

654

655

656

657

658

659

660

661

662

663

664

665

666

667

668

669

670

671

672

673

674

675

676

677

678

679

680

681

682

683

684

685

686

687

688

689

690

691

692

693

694

695

696

697

698

699

700

701

702

703

704

705

706

707

708

709

710

711

712

713

714

715

716

717

718

719

720

721

722

723

724

725

/*****************************************************************************

Licensed under the Apache License, Version 2.0 (the "License");

you may not use this file except in compliance with the License.

You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software

distributed under the License is distributed on an "AS IS" BASIS,

WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.

See the License for the specific language governing permissions and

limitations under the License.

******************************************************************************/

using Tensorflow.NumPy;

using System;

using System.Collections.Generic;

using System.Linq;

using System.Text;

using Tensorflow.Eager;

using Tensorflow.Graphs;

using static Tensorflow.Binding;

using System.Diagnostics;

namespace Tensorflow

{

public static class tensor_util

{

/// <summary>

/// Returns the constant value of the given tensor, if efficiently calculable.

/// </summary>

/// <param name="tensor"></param>

/// <param name="partial"></param>

/// <returns></returns>

public static NDArray constant_value(Tensor tensor, bool partial = false)

{

if (tensor is NDArray nd)

return nd;

else if (tensor is EagerTensor)

return tensor.numpy();

NDArray ret = _ConstantValue(tensor, partial);

if (!(ret is null))

tensor.graph.prevent_feeding(tensor);

return ret;

}

private static NDArray _ConstantValue(Tensor tensor, bool partial)

{

switch (tensor.op.type)

{

case "Const":

return MakeNdarray(tensor.op.get_attr("value") as TensorProto);

default:

return null;

}

public static NDArray MakeNdarray(TensorProto tensor)

{

var shape = new Shape(tensor.TensorShape.Dim.Select(x => x.Size).ToArray());

var num_elements = shape.size;

var tensor_dtype = tensor.Dtype.as_tf_dtype();

T[] ExpandArrayToSize<T>(IList<T> src)

{

if (src.Count == 0)

{

return new T[0];

}

var pad_count = num_elements - src.Count;

var pre = pad_count / 2;

var after = pad_count - pre;

var first_elem = src[0];

var last_elem = src[src.Count - 1];

T[] res = new T[num_elements];

for (long i = 0; i < num_elements; i++)

{

if (i < pre) res[i] = first_elem;

else if (i >= num_elements - after) res[i] = last_elem;

else res[i] = src[(int)(i - pre)];

}

return res;

}

if (shape.ndim > 0 && tensor.TensorContent.Length > 0)

{

return np.frombuffer(tensor.TensorContent.ToByteArray(), shape, tensor_dtype);

}

NDArray values;

if (tensor.Dtype == DataType.DtHalf || tensor.Dtype == DataType.DtBfloat16)

{

values = np.array(ExpandArrayToSize(tensor.HalfVal));

}

else if (tensor.Dtype == DataType.DtFloat)

{

values = np.array(ExpandArrayToSize(tensor.FloatVal));

}

else if (new DataType[] { DataType.DtInt32, DataType.DtUint8 }.Contains(tensor.Dtype))

{

values = np.array(ExpandArrayToSize(tensor.IntVal));

}

else if (new DataType[] { DataType.DtInt64 }.Contains(tensor.Dtype))

{

values = np.array(ExpandArrayToSize(tensor.Int64Val));

}

else if (new DataType[] { DataType.DtUint64 }.Contains(tensor.Dtype))

{

values = np.array(ExpandArrayToSize(tensor.Uint64Val));

}

else if (tensor.Dtype == DataType.DtBool)

{

values = np.array(ExpandArrayToSize(tensor.BoolVal));

}

else

{

throw new TypeError($"Unsupported tensor type: {tensor.Dtype}. See " +

$"https://www.tensorflow.org/api_docs/python/tf/dtypes for supported TF dtypes.");

}

if (values.size == 0)

{

return np.zeros(shape, tensor_dtype);

}

return values.reshape(shape);

}

private static readonly TF_DataType[] quantized_types = new TF_DataType[]

{

TF_DataType.TF_QINT8, TF_DataType.TF_QUINT8, TF_DataType.TF_QINT16, TF_DataType.TF_QUINT16,

TF_DataType.TF_QINT32

};

private static Array ConvertArray<TOut>(Array inputArray, Func<object, TOut> converter)

{

if (inputArray == null)

throw new ArgumentNullException(nameof(inputArray));

var elementType = typeof(TOut);

var lengths = new int[inputArray.Rank];

for (var i = 0; i < inputArray.Rank; i++)

{

lengths[i] = inputArray.GetLength(i);

}

var outputArray = Array.CreateInstance(elementType, lengths);

FillArray(inputArray, outputArray, converter, new int[inputArray.Rank], 0);

return outputArray;

}

private static void FillArray<TIn, TOut>(Array inputArray, Array outputArray, Func<TIn, TOut> converter, int[] indices, int dimension)

{

if (dimension == inputArray.Rank - 1)

{

for (int i = 0; i < inputArray.GetLength(dimension); i++)

{

indices[dimension] = i;

var inputValue = (TIn)inputArray.GetValue(indices);

var convertedValue = converter(inputValue);

outputArray.SetValue(convertedValue, indices);

}

else

{

for (int i = 0; i < inputArray.GetLength(dimension); i++)

{

indices[dimension] = i;

FillArray(inputArray, outputArray, converter, indices, dimension + 1);

}

/// <summary>

/// Create a TensorProto, invoked in graph mode

/// </summary>

/// <param name="values"></param>

/// <param name="dtype"></param>

/// <param name="shape"></param>

/// <param name="verify_shape"></param>

/// <param name="allow_broadcast"></param>

/// <returns></returns>

public static TensorProto make_tensor_proto(object values, TF_DataType dtype = TF_DataType.DtInvalid, Shape? shape = null, bool verify_shape = false, bool allow_broadcast = false)

{

if (allow_broadcast && verify_shape)

throw new ValueError("allow_broadcast and verify_shape are not both allowed.");

if (values is TensorProto tp)

return tp;

var origin_dtype = values.GetDataType();

if (dtype == TF_DataType.DtInvalid)

dtype = origin_dtype;

else if (origin_dtype != dtype)

{

var new_system_dtype = dtype.as_system_dtype();

if (dtype != TF_DataType.TF_STRING && dtype != TF_DataType.TF_VARIANT && dtype != TF_DataType.TF_RESOURCE)

{

if (values is Array arrayValues)

{

values = dtype switch

{

TF_DataType.TF_INT32 => ConvertArray(arrayValues, Convert.ToInt32),

TF_DataType.TF_FLOAT => ConvertArray(arrayValues, Convert.ToSingle),

TF_DataType.TF_DOUBLE => ConvertArray(arrayValues, Convert.ToDouble),

_ => values,

};

} else

{

values = Convert.ChangeType(values, new_system_dtype);

}

} else

{

}

dtype = values.GetDataType();

}

shape = shape ?? values.GetShape();

var tensor_proto = new TensorProto

{

Dtype = dtype.as_datatype_enum(),

TensorShape = shape.as_shape_proto()

};

if (values is NDArray nd)

{

// scalar

if (nd.shape.IsScalar)

{

switch (nd.dtype)

{

case TF_DataType.TF_BOOL:

tensor_proto.BoolVal.AddRange(nd.ToArray<bool>());

break;

case TF_DataType.TF_UINT8:

tensor_proto.IntVal.AddRange(nd.ToArray<byte>().Select(x => (int)x).ToArray());

break;

case TF_DataType.TF_INT32:

tensor_proto.IntVal.AddRange(nd.ToArray<int>());

break;

case TF_DataType.TF_INT64:

tensor_proto.Int64Val.AddRange(nd.ToArray<long>());

break;

case TF_DataType.TF_FLOAT:

tensor_proto.FloatVal.AddRange(nd.ToArray<float>());

break;

case TF_DataType.TF_DOUBLE:

tensor_proto.DoubleVal.AddRange(nd.ToArray<double>());

break;

default:

throw new Exception("make_tensor_proto Not Implemented");

}

else

{

var len = nd.dtypesize * nd.size;

byte[] bytes = nd.ToByteArray();

tensor_proto.TensorContent = Google.Protobuf.ByteString.CopyFrom(bytes);

}

else if (dtype == TF_DataType.TF_STRING && !(values is NDArray))

{

if (values is string str)

tensor_proto.StringVal.Add(Google.Protobuf.ByteString.CopyFromUtf8(str));

else if (values is string[] str_values)

tensor_proto.StringVal.AddRange(str_values.Select(x => Google.Protobuf.ByteString.CopyFromUtf8(x)));

else if (values is byte[] byte_values)

tensor_proto.TensorContent = Google.Protobuf.ByteString.CopyFrom(byte_values);

}

else if (values is Array array)

{

// array

var len = dtype.get_datatype_size() * (int)shape.size;

byte[] bytes = new byte[len];

System.Buffer.BlockCopy(array, 0, bytes, 0, len);

tensor_proto.TensorContent = Google.Protobuf.ByteString.CopyFrom(bytes);

}

else

{

switch (values)

{

case Axis val:

tensor_proto.IntVal.AddRange(val.axis);

break;

case Shape val:

tensor_proto.Int64Val.AddRange(val.dims);

break;

case bool val:

tensor_proto.BoolVal.AddRange(new[] { val });

break;

case sbyte val:

tensor_proto.IntVal.AddRange(new[] { (int)val });

break;

case byte val:

tensor_proto.IntVal.AddRange(new[] { (int)val });

break;

case int val:

tensor_proto.IntVal.AddRange(new[] { val });

break;

case long val:

tensor_proto.Int64Val.AddRange(new[] { val });

break;

case float val:

tensor_proto.FloatVal.AddRange(new[] { val });

break;

case double val:

tensor_proto.DoubleVal.AddRange(new[] { val });

break;

default:

throw new Exception($"make_tensor_proto Not Implemented {values.GetType().Name}");

}

return tensor_proto;

}

public static Shape constant_value_as_shape(Tensor tensor)

{

bool hasattr(Graph property, string attr)

{

var t = property.GetType().GetProperties();

foreach (System.Reflection.PropertyInfo pi in t)

{

if (pi.Name == attr)

return true;

}

return false;

}

if (tensor is EagerTensor eagerTensor)

{

if (tensor.dtype == tf.int64)

return new Shape(tensor.ToArray<long>());

else

return new Shape(tensor.ToArray<int>());

}

if (tensor.shape.ndim == 0)

{

var value_ = constant_value(tensor);

if (value_ == null)

throw new ValueError(

@"Received a scalar with unknown value as shape; require a statically

known scalar with value '-1' to describe an unknown shape.");

if ((int)value_ != -1)

throw new ValueError(

String.Format(@"Received a scalar value {0} as shape; require a statically known

scalar with value '-1' to describe an unknown shape.", value_));

return tensor.shape.unknown_shape(-1);

}

var shape = tensor.shape.with_rank(1);

if (shape == new Shape(new int[] { 1 }))

{

return new Shape(new int[] { });

}

else if (tensor.op.type == "Cast")

{

var pre_cast = constant_value_as_shape(tensor.op.inputs[0]);

if (pre_cast.dims == null)

return pre_cast;

var cast_dtype = dtypes.as_tf_dtype((Type)tensor.op.get_attr("DstT"));

if (!Array.Exists(new[] { dtypes.int32, dtypes.int64 }, cast_dtype_ => cast_dtype_ == cast_dtype))

return tensor.shape.unknown_shape((int)shape.dims[0]);

long[] x_ = { };

foreach (var x in pre_cast.dims)

if (x != -1)

x_[x_.Length] = x;

else

x_[x_.Length] = -1;

var dest_dtype_shape_array = np.array(x_).astype(cast_dtype);

long[] y_ = { };

foreach (int y in dest_dtype_shape_array.ToArray<int>())

if (y >= 0)

y_[y_.Length] = y;

else

y_[y_.Length] = -1;

return new Shape(y_);

}

else if (tensor.op.type == "Shape")

{

return tensor.op.inputs[0].shape;

}

else if (tensor.op.type == "Pack")

{

var ret_ = new Shape(new int[] { });

if ((int)tensor.op.get_attr("axis") != 0)

throw new ValueError(String.Format(

@"Since rank 1 inputs are expected, Pack's axis: {0} must be 0, otherwise it

would not be rank 1.", tensor.op.get_attr("axis")));

foreach (Tensor pack_input in tensor.op.inputs)

{

var pack_input_val = (int)constant_value(pack_input);

Dimension new_dim;

if (pack_input_val < 0)

{

new_dim = new Dimension(-1);

}

else if (pack_input_val == null)

{

new_dim = new Dimension(-1);

}

else

{

new_dim = new Dimension(pack_input_val);

}

ret_ = ret_.concatenate(new long[] { new_dim });

}

return ret_;

}

else if (tensor.op.type == "Concat")

{

var ret_ = new Shape(new int[] { });

var inputlist_ = new ArraySegment<Tensor>(tensor.op.inputs, 1,

tensor.op.inputs.Length - 1);

foreach (var concat_input in inputlist_)

{

ret_ = ret_.concatenate(constant_value_as_shape(concat_input));

}

return ret_;

}

else if (tensor.op.type == "StridedSlice")

{

try

{

var begin = constant_value(tensor.op.inputs[1]);

var end = constant_value(tensor.op.inputs[2]);

var strides = constant_value(tensor.op.inputs[3]);

if (new[] { begin, end, strides }.All(x => x == null))

{

begin = begin[0];

end = end[0];

strides = strides[0];

var begin_mask = tensor.op.get_attr("begin_mask");

if ((int)begin_mask == 1)

{

begin = null;

}

var end_mask = tensor.op.get_attr("end_mask");

if ((int)end_mask == 1)

{

end = null;

}

var ellipsis_mask = tensor.op.get_attr("ellipsis_mask");

var new_axis_mask = tensor.op.get_attr("new_axis_mask");

var shrink_axis_mask = tensor.op.get_attr("shrink_axis_mask");

bool valid_attributes;

if (!(bool)ellipsis_mask && !(bool)new_axis_mask &&

!(bool)shrink_axis_mask && !((bool)begin_mask || (int)begin_mask == 1) &&

!((bool)end_mask || (int)end_mask == 1))

{

valid_attributes = true;

}

else { valid_attributes = false; }

if (valid_attributes)

{

// sorry for the mess here, but this hacky solution was the best way

// i could come up with to implement the things done in python in c#

var prev_ = constant_value_as_shape(tensor.op.inputs[0]).dims;

var prev = prev_.Skip((int)begin).Take((int)end - (int)begin).ToArray();

// 100 being the comparison doesn't really matter here; it's going to break anyway

for (int iter = 0; iter != 100; iter = iter + (int)strides)

{

prev[prev.Length] = prev_[iter];

if ((iter + (int)strides) > prev_.Length)

break;

}

var ret_ = new Shape(prev);

return ret_;

}

catch (Exception ex)

{

if (ex is ValueError || ex is TypeError) { }

}

else if (tensor.op.type == "Placeholder" &&

tensor.op.graph.building_function &&

tensor.op.graph is FuncGraph func_graph)

{

int i = 0;

foreach (Tensor capture in func_graph.internal_captures)

{

if (capture.GetType() == typeof(Tensor))

{

var external_capture = func_graph.external_captures[i];

return constant_value_as_shape(external_capture);

}

i++;

}

var ret = tensor.shape.unknown_shape((int)shape.dims[0]);

var value = constant_value(tensor);

if (value is not null)

{

var d_ = new int[value.size];

foreach (var (index, d) in enumerate(value.ToArray<int>()))

d_[index] = d >= 0 ? d : -1;

ret = ret.merge_with(new Shape(d_));

}

return ret;

}

public static TensorShapeProto as_shape<T>(T[] dims)

{

TensorShapeProto shape = new TensorShapeProto();

for (int i = 0; i < dims.Length; i++)

{

var dim = new TensorShapeProto.Types.Dim();

switch (dims[i])

{

case int n:

dim.Size = n;

break;

case long l:

dim.Size = l;

break;

default:

throw new NotImplementedException("as_shape Not Implemented");

}

// dim.Name = $"dim_{i}";

shape.Dim.Add(dim);

}

return shape;

}

public static Shape to_shape(long[] dims)

{

return new Shape(dims.Select(x => (int)x).ToArray());

}

public static Shape to_shape(int[] dims)

{

return new Shape(dims);

}

public static TensorShapeProto as_shape_proto(this Shape tshape)

{

TensorShapeProto shape = new TensorShapeProto();

for (int i = 0; i < tshape.ndim; i++)

{

var dim = new TensorShapeProto.Types.Dim();

dim.Size = tshape.dims[i];

//dim.Name = $"dim_{i}";

shape.Dim.Add(dim);

}

return shape;

}

public static Shape reshape(this Shape shape, int[] dims)

{

return new Shape(dims);

}

public static TensorShapeProto as_proto(this Shape tshape)

{

TensorShapeProto shape = new TensorShapeProto();

for (int i = 0; i < tshape.ndim; i++)

{

var dim = new TensorShapeProto.Types.Dim();

dim.Size = tshape.dims[i];

//dim.Name = $"dim_{i}";

shape.Dim.Add(dim);

}

return shape;

}

public static Tensor shape_tensor(int[] shape)

{

return ops.convert_to_tensor(shape, dtype: TF_DataType.TF_INT32, name: "shape");

}

public static ParsedSliceArgs ParseSlices(Slice[] slices)

{

var begin = new List<int>();

var end = new List<int>();

var strides = new List<int>();

var index = 0;

var (new_axis_mask, shrink_axis_mask) = (0, 0);

var (begin_mask, end_mask) = (0, 0);

var ellipsis_mask = 0;

foreach (var s in slices)

{

if (s.IsNewAxis)

{

begin.Add(0);

end.Add(0);

strides.Add(1);

new_axis_mask |= (1 << index);

}

else if (s.IsEllipsis)

{

begin.Add(0);

end.Add(0);

strides.Add(1);

ellipsis_mask |= (1 << index);

}

else

{

if (s.Start.HasValue)

{

begin.Add(s.Start.Value);

}

else

{

begin.Add(0);

begin_mask |= (1 << index);

}

if (s.Stop.HasValue)

{

end.Add(s.Stop.Value);

}

else

{

end.Add(0);

end_mask |= (1 << index);

}

strides.Add(s.Step);

if (s.IsIndex)

shrink_axis_mask |= (1 << index);

}

index += 1;

}

return new ParsedSliceArgs

{

Begin = begin.ToArray(),

End = end.ToArray(),

Strides = strides.ToArray(),

BeginMask = begin_mask,

EndMask = end_mask,

EllipsisMask = ellipsis_mask,

ShrinkAxisMask = shrink_axis_mask,

NewAxisMask = new_axis_mask

};

}

public static ParsedSliceArgs ParseSlices(Tensor start, Tensor stop = null, Tensor step = null)

{

var begin = new List<Tensor>();

var end = new List<Tensor>();

var strides = new List<Tensor>();

var index = 0;

var (new_axis_mask, shrink_axis_mask) = (0, 0);

var (begin_mask, end_mask) = (0, 0);

var ellipsis_mask = 0;

begin.Add(start);

if (stop == null)

end.Add(start + 1);

else

end.Add(stop);

shrink_axis_mask |= (1 << index);

if (step == null)

strides.Add(tf.constant(1, dtype: start.dtype));

else

strides.Add(step);

return new ParsedSliceArgs

{

PackedBegin = array_ops.stack(begin),

PackedEnd = array_ops.stack(end),

PackedStrides = array_ops.stack(strides),

BeginMask = begin_mask,

EndMask = end_mask,

EllipsisMask = ellipsis_mask,

ShrinkAxisMask = shrink_axis_mask,

NewAxisMask = new_axis_mask

};

}

/// <summary>

/// Warning: this method is an extremely dangerous method. It directly changes the dtype inside the tensor

/// and security is not guaranteed at all. Currently this method is only used for some conditions to reuse

/// the existing memory. Any other usage should be prevented. If you are sure you want to use it when

/// developing tensorflow.net, please ask @Oceanic2018 or @AsakusaRinne first.

/// </summary>

/// <param name="handle"></param>

/// <param name="dtype"></param>

internal static unsafe void DangerousManuallySetTensorDType(SafeTensorHandle handle, TF_DataType dtype)

{

long tf_tensor_address = handle.DangerousGetHandle().ToInt64();

long interface_address = *(long*)(tf_tensor_address);

long tensor_shape_address = interface_address + 8;

long tensor_dtype_address = tensor_shape_address + 13;

byte* dtype_pointer = (byte*)tensor_dtype_address;

*dtype_pointer = (byte)dtype;

Debug.Assert(c_api.TF_TensorType(handle) == dtype);

}

TensorFlow.NET/src/TensorFlowNET.Core/Tensors/tensor_util.cs at master · feelsyt/TensorFlow.NET · GitHub

Expand file tree

Breadcrumbs

tensor_util.cs

Latest commit

History

Breadcrumbs

tensor_util.cs

File metadata and controls