Packages

class PythonBigDL[T] extends Serializable

Implementation of Python API for BigDL

Linear Supertypes
Serializable, Serializable, AnyRef, Any
Known Subclasses
PythonBigDLKeras, PythonBigDLOnnx, PythonBigDLValidator
Ordering
  1. Alphabetic
  2. By Inheritance
Inherited
  1. PythonBigDL
  2. Serializable
  3. Serializable
  4. AnyRef
  5. Any
  1. Hide All
  2. Show All
Visibility
  1. Public
  2. All

Instance Constructors

  1. new PythonBigDL()(implicit arg0: ClassTag[T], ev: TensorNumeric[T])

Value Members

  1. final def !=(arg0: Any): Boolean
    Definition Classes
    AnyRef → Any
  2. final def ##(): Int
    Definition Classes
    AnyRef → Any
  3. final def ==(arg0: Any): Boolean
    Definition Classes
    AnyRef → Any
  4. def activityToJTensors(outputActivity: Activity): List[JTensor]
  5. def addScheduler(seq: SequentialSchedule, scheduler: LearningRateSchedule, maxIteration: Int): SequentialSchedule
  6. final def asInstanceOf[T0]: T0
    Definition Classes
    Any
  7. def batching(dataset: DataSet[dataset.Sample[T]], batchSize: Int): DataSet[MiniBatch[T]]
  8. def clone(): AnyRef
    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @native() @throws( ... )
  9. def createAbs(): Abs[T]
  10. def createAbsCriterion(sizeAverage: Boolean = true): AbsCriterion[T]
  11. def createActivityRegularization(l1: Double, l2: Double): ActivityRegularization[T]
  12. def createAdadelta(decayRate: Double = 0.9, Epsilon: Double = 1e-10): Adadelta[T]
  13. def createAdagrad(learningRate: Double = 1e-3, learningRateDecay: Double = 0.0, weightDecay: Double = 0.0): Adagrad[T]
  14. def createAdam(learningRate: Double = 1e-3, learningRateDecay: Double = 0.0, beta1: Double = 0.9, beta2: Double = 0.999, Epsilon: Double = 1e-8): Adam[T]
  15. def createAdamax(learningRate: Double = 0.002, beta1: Double = 0.9, beta2: Double = 0.999, Epsilon: Double = 1e-38): Adamax[T]
  16. def createAdd(inputSize: Int): Add[T]
  17. def createAddConstant(constant_scalar: Double, inplace: Boolean = false): AddConstant[T]
  18. def createAspectScale(scale: Int, scaleMultipleOf: Int, maxSize: Int, resizeMode: Int = 1, useScaleFactor: Boolean = true, minScale: Double = -1): FeatureTransformer
  19. def createAttention(hiddenSize: Int, numHeads: Int, attentionDropout: Float): Attention[T]
  20. def createBCECriterion(weights: JTensor = null, sizeAverage: Boolean = true): BCECriterion[T]
  21. def createBatchNormalization(nOutput: Int, eps: Double = 1e-5, momentum: Double = 0.1, affine: Boolean = true, initWeight: JTensor = null, initBias: JTensor = null, initGradWeight: JTensor = null, initGradBias: JTensor = null): BatchNormalization[T]
  22. def createBiRecurrent(merge: AbstractModule[Table, Tensor[T], T] = null): BiRecurrent[T]
  23. def createBifurcateSplitTable(dimension: Int): BifurcateSplitTable[T]
  24. def createBilinear(inputSize1: Int, inputSize2: Int, outputSize: Int, biasRes: Boolean = true, wRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null): Bilinear[T]
  25. def createBilinearFiller(): BilinearFiller.type
  26. def createBinaryThreshold(th: Double, ip: Boolean): BinaryThreshold[T]
  27. def createBinaryTreeLSTM(inputSize: Int, hiddenSize: Int, gateOutput: Boolean = true, withGraph: Boolean = true): BinaryTreeLSTM[T]
  28. def createBottle(module: AbstractModule[Activity, Activity, T], nInputDim: Int = 2, nOutputDim1: Int = Int.MaxValue): Bottle[T]
  29. def createBrightness(deltaLow: Double, deltaHigh: Double): Brightness
  30. def createBytesToMat(byteKey: String): BytesToMat
  31. def createCAdd(size: List[Int], bRegularizer: Regularizer[T] = null): CAdd[T]
  32. def createCAddTable(inplace: Boolean = false): CAddTable[T, T]
  33. def createCAveTable(inplace: Boolean = false): CAveTable[T]
  34. def createCDivTable(): CDivTable[T]
  35. def createCMaxTable(): CMaxTable[T]
  36. def createCMinTable(): CMinTable[T]
  37. def createCMul(size: List[Int], wRegularizer: Regularizer[T] = null): CMul[T]
  38. def createCMulTable(): CMulTable[T]
  39. def createCSubTable(): CSubTable[T]
  40. def createCategoricalCrossEntropy(): CategoricalCrossEntropy[T]
  41. def createCenterCrop(cropWidth: Int, cropHeight: Int, isClip: Boolean): CenterCrop
  42. def createChannelNormalize(meanR: Double, meanG: Double, meanB: Double, stdR: Double = 1, stdG: Double = 1, stdB: Double = 1): FeatureTransformer
  43. def createChannelOrder(): ChannelOrder
  44. def createChannelScaledNormalizer(meanR: Int, meanG: Int, meanB: Int, scale: Double): ChannelScaledNormalizer
  45. def createClamp(min: Int, max: Int): Clamp[T]
  46. def createClassNLLCriterion(weights: JTensor = null, sizeAverage: Boolean = true, logProbAsInput: Boolean = true): ClassNLLCriterion[T]
  47. def createClassSimplexCriterion(nClasses: Int): ClassSimplexCriterion[T]
  48. def createColorJitter(brightnessProb: Double = 0.5, brightnessDelta: Double = 32, contrastProb: Double = 0.5, contrastLower: Double = 0.5, contrastUpper: Double = 1.5, hueProb: Double = 0.5, hueDelta: Double = 18, saturationProb: Double = 0.5, saturationLower: Double = 0.5, saturationUpper: Double = 1.5, randomOrderProb: Double = 0, shuffle: Boolean = false): ColorJitter
  49. def createConcat(dimension: Int): Concat[T]
  50. def createConcatTable(): ConcatTable[T]
  51. def createConstInitMethod(value: Double): ConstInitMethod
  52. def createContiguous(): Contiguous[T]
  53. def createContrast(deltaLow: Double, deltaHigh: Double): Contrast
  54. def createConvLSTMPeephole(inputSize: Int, outputSize: Int, kernelI: Int, kernelC: Int, stride: Int = 1, padding: Int = -1, activation: TensorModule[T] = null, innerActivation: TensorModule[T] = null, wRegularizer: Regularizer[T] = null, uRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null, cRegularizer: Regularizer[T] = null, withPeephole: Boolean = true): ConvLSTMPeephole[T]
  55. def createConvLSTMPeephole3D(inputSize: Int, outputSize: Int, kernelI: Int, kernelC: Int, stride: Int = 1, padding: Int = -1, wRegularizer: Regularizer[T] = null, uRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null, cRegularizer: Regularizer[T] = null, withPeephole: Boolean = true): ConvLSTMPeephole3D[T]
  56. def createCosine(inputSize: Int, outputSize: Int): Cosine[T]
  57. def createCosineDistance(): CosineDistance[T]
  58. def createCosineDistanceCriterion(sizeAverage: Boolean = true): CosineDistanceCriterion[T]
  59. def createCosineEmbeddingCriterion(margin: Double = 0.0, sizeAverage: Boolean = true): CosineEmbeddingCriterion[T]
  60. def createCosineProximityCriterion(): CosineProximityCriterion[T]
  61. def createCropping2D(heightCrop: List[Int], widthCrop: List[Int], dataFormat: String = "NCHW"): Cropping2D[T]
  62. def createCropping3D(dim1Crop: List[Int], dim2Crop: List[Int], dim3Crop: List[Int], dataFormat: String = Cropping3D.CHANNEL_FIRST): Cropping3D[T]
  63. def createCrossEntropyCriterion(weights: JTensor = null, sizeAverage: Boolean = true): CrossEntropyCriterion[T]
  64. def createCrossProduct(numTensor: Int = 0, embeddingSize: Int = 0): CrossProduct[T]
  65. def createDLClassifier(model: Module[T], criterion: Criterion[T], featureSize: ArrayList[Int], labelSize: ArrayList[Int]): DLClassifier[T]
  66. def createDLClassifierModel(model: Module[T], featureSize: ArrayList[Int]): DLClassifierModel[T]
  67. def createDLEstimator(model: Module[T], criterion: Criterion[T], featureSize: ArrayList[Int], labelSize: ArrayList[Int]): DLEstimator[T]
  68. def createDLImageTransformer(transformer: FeatureTransformer): DLImageTransformer
  69. def createDLModel(model: Module[T], featureSize: ArrayList[Int]): DLModel[T]
  70. def createDatasetFromImageFrame(imageFrame: ImageFrame): DataSet[ImageFeature]
  71. def createDefault(): Default
  72. def createDenseToSparse(): DenseToSparse[T]
  73. def createDetectionCrop(roiKey: String, normalized: Boolean): DetectionCrop
  74. def createDetectionOutputFrcnn(nmsThresh: Float = 0.3f, nClasses: Int, bboxVote: Boolean, maxPerImage: Int = 100, thresh: Double = 0.05): DetectionOutputFrcnn
  75. def createDetectionOutputSSD(nClasses: Int, shareLocation: Boolean, bgLabel: Int, nmsThresh: Double, nmsTopk: Int, keepTopK: Int, confThresh: Double, varianceEncodedInTarget: Boolean, confPostProcess: Boolean): DetectionOutputSSD[T]
  76. def createDiceCoefficientCriterion(sizeAverage: Boolean = true, epsilon: Float = 1.0f): DiceCoefficientCriterion[T]
  77. def createDistKLDivCriterion(sizeAverage: Boolean = true): DistKLDivCriterion[T]
  78. def createDistriOptimizer(model: AbstractModule[Activity, Activity, T], trainingRdd: JavaRDD[Sample], criterion: Criterion[T], optimMethod: Map[String, OptimMethod[T]], endTrigger: Trigger, batchSize: Int): Optimizer[T, MiniBatch[T]]
  79. def createDistriOptimizerFromDataSet(model: AbstractModule[Activity, Activity, T], trainDataSet: DataSet[ImageFeature], criterion: Criterion[T], optimMethod: Map[String, OptimMethod[T]], endTrigger: Trigger, batchSize: Int): Optimizer[T, MiniBatch[T]]
  80. def createDistributedImageFrame(imageRdd: JavaRDD[JTensor], labelRdd: JavaRDD[JTensor]): DistributedImageFrame
  81. def createDotProduct(): DotProduct[T]
  82. def createDotProductCriterion(sizeAverage: Boolean = false): DotProductCriterion[T]
  83. def createDropout(initP: Double = 0.5, inplace: Boolean = false, scale: Boolean = true): Dropout[T]
  84. def createELU(alpha: Double = 1.0, inplace: Boolean = false): ELU[T]
  85. def createEcho(): Echo[T]
  86. def createEuclidean(inputSize: Int, outputSize: Int, fastBackward: Boolean = true): Euclidean[T]
  87. def createEveryEpoch(): Trigger
  88. def createExp(): Exp[T]
  89. def createExpand(meansR: Int = 123, meansG: Int = 117, meansB: Int = 104, minExpandRatio: Double = 1.0, maxExpandRatio: Double = 4.0): Expand
  90. def createExpandSize(targetSizes: List[Int]): ExpandSize[T]
  91. def createExponential(decayStep: Int, decayRate: Double, stairCase: Boolean = false): Exponential
  92. def createFPN(in_channels_list: List[Int], out_channels: Int, top_blocks: Int = 0, in_channels_of_p6p7: Int = 0, out_channels_of_p6p7: Int = 0): FPN[T]
  93. def createFeedForwardNetwork(hiddenSize: Int, filterSize: Int, reluDropout: Float): FeedForwardNetwork[T]
  94. def createFiller(startX: Double, startY: Double, endX: Double, endY: Double, value: Int = 255): Filler
  95. def createFixExpand(eh: Int, ew: Int): FixExpand
  96. def createFixedCrop(wStart: Double, hStart: Double, wEnd: Double, hEnd: Double, normalized: Boolean, isClip: Boolean): FixedCrop
  97. def createFlattenTable(): FlattenTable[T]
  98. def createFtrl(learningRate: Double = 1e-3, learningRatePower: Double = -0.5, initialAccumulatorValue: Double = 0.1, l1RegularizationStrength: Double = 0.0, l2RegularizationStrength: Double = 0.0, l2ShrinkageRegularizationStrength: Double = 0.0): Ftrl[T]
  99. def createGRU(inputSize: Int, outputSize: Int, p: Double = 0, activation: TensorModule[T] = null, innerActivation: TensorModule[T] = null, wRegularizer: Regularizer[T] = null, uRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null): GRU[T]
  100. def createGaussianCriterion(): GaussianCriterion[T]
  101. def createGaussianDropout(rate: Double): GaussianDropout[T]
  102. def createGaussianNoise(stddev: Double): GaussianNoise[T]
  103. def createGaussianSampler(): GaussianSampler[T]
  104. def createGradientReversal(lambda: Double = 1): GradientReversal[T]
  105. def createHFlip(): HFlip
  106. def createHardShrink(lambda: Double = 0.5): HardShrink[T]
  107. def createHardSigmoid: HardSigmoid[T]
  108. def createHardTanh(minValue: Double = -1, maxValue: Double = 1, inplace: Boolean = false): HardTanh[T]
  109. def createHighway(size: Int, withBias: Boolean, activation: TensorModule[T] = null, wRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null): Graph[T]
  110. def createHingeEmbeddingCriterion(margin: Double = 1, sizeAverage: Boolean = true): HingeEmbeddingCriterion[T]
  111. def createHitRatio(k: Int = 10, negNum: Int = 100): ValidationMethod[T]
  112. def createHue(deltaLow: Double, deltaHigh: Double): Hue
  113. def createIdentity(): Identity[T]
  114. def createImageFeature(data: JTensor = null, label: JTensor = null, uri: String = null): ImageFeature
  115. def createImageFrameToSample(inputKeys: List[String], targetKeys: List[String], sampleKey: String): ImageFrameToSample[T]
  116. def createIndex(dimension: Int): Index[T]
  117. def createInferReshape(size: List[Int], batchMode: Boolean = false): InferReshape[T]
  118. def createInput(): ModuleNode[T]
  119. def createJoinTable(dimension: Int, nInputDims: Int): JoinTable[T]
  120. def createKLDCriterion(sizeAverage: Boolean): KLDCriterion[T]
  121. def createKullbackLeiblerDivergenceCriterion: KullbackLeiblerDivergenceCriterion[T]
  122. def createL1Cost(): L1Cost[T]
  123. def createL1HingeEmbeddingCriterion(margin: Double = 1): L1HingeEmbeddingCriterion[T]
  124. def createL1L2Regularizer(l1: Double, l2: Double): L1L2Regularizer[T]
  125. def createL1Penalty(l1weight: Int, sizeAverage: Boolean = false, provideOutput: Boolean = true): L1Penalty[T]
  126. def createL1Regularizer(l1: Double): L1Regularizer[T]
  127. def createL2Regularizer(l2: Double): L2Regularizer[T]
  128. def createLBFGS(maxIter: Int = 20, maxEval: Double = Double.MaxValue, tolFun: Double = 1e-5, tolX: Double = 1e-9, nCorrection: Int = 100, learningRate: Double = 1.0, verbose: Boolean = false, lineSearch: LineSearch[T] = null, lineSearchOptions: Map[Any, Any] = null): LBFGS[T]
  129. def createLSTM(inputSize: Int, hiddenSize: Int, p: Double = 0, activation: TensorModule[T] = null, innerActivation: TensorModule[T] = null, wRegularizer: Regularizer[T] = null, uRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null): LSTM[T]
  130. def createLSTMPeephole(inputSize: Int, hiddenSize: Int, p: Double = 0, wRegularizer: Regularizer[T] = null, uRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null): LSTMPeephole[T]
  131. def createLayerNormalization(hiddenSize: Int): LayerNormalization[T]
  132. def createLeakyReLU(negval: Double = 0.01, inplace: Boolean = false): LeakyReLU[T]
  133. def createLinear(inputSize: Int, outputSize: Int, withBias: Boolean, wRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null, initWeight: JTensor = null, initBias: JTensor = null, initGradWeight: JTensor = null, initGradBias: JTensor = null): Linear[T]
  134. def createLocalImageFrame(images: List[JTensor], labels: List[JTensor]): LocalImageFrame
  135. def createLocalOptimizer(features: List[JTensor], y: JTensor, model: AbstractModule[Activity, Activity, T], criterion: Criterion[T], optimMethod: Map[String, OptimMethod[T]], endTrigger: Trigger, batchSize: Int, localCores: Int): Optimizer[T, MiniBatch[T]]
  136. def createLocallyConnected1D(nInputFrame: Int, inputFrameSize: Int, outputFrameSize: Int, kernelW: Int, strideW: Int = 1, propagateBack: Boolean = true, wRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null, initWeight: JTensor = null, initBias: JTensor = null, initGradWeight: JTensor = null, initGradBias: JTensor = null): LocallyConnected1D[T]
  137. def createLocallyConnected2D(nInputPlane: Int, inputWidth: Int, inputHeight: Int, nOutputPlane: Int, kernelW: Int, kernelH: Int, strideW: Int = 1, strideH: Int = 1, padW: Int = 0, padH: Int = 0, propagateBack: Boolean = true, wRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null, initWeight: JTensor = null, initBias: JTensor = null, initGradWeight: JTensor = null, initGradBias: JTensor = null, withBias: Boolean = true, dataFormat: String = "NCHW"): LocallyConnected2D[T]
  138. def createLog(): Log[T]
  139. def createLogSigmoid(): LogSigmoid[T]
  140. def createLogSoftMax(): LogSoftMax[T]
  141. def createLookupTable(nIndex: Int, nOutput: Int, paddingValue: Double = 0, maxNorm: Double = Double.MaxValue, normType: Double = 2.0, shouldScaleGradByFreq: Boolean = false, wRegularizer: Regularizer[T] = null): LookupTable[T]
  142. def createLookupTableSparse(nIndex: Int, nOutput: Int, combiner: String = "sum", maxNorm: Double = -1, wRegularizer: Regularizer[T] = null): LookupTableSparse[T]
  143. def createLoss(criterion: Criterion[T]): ValidationMethod[T]
  144. def createMAE(): ValidationMethod[T]
  145. def createMM(transA: Boolean = false, transB: Boolean = false): MM[T]
  146. def createMSECriterion: MSECriterion[T]
  147. def createMV(trans: Boolean = false): MV[T]
  148. def createMapTable(module: AbstractModule[Activity, Activity, T] = null): MapTable[T]
  149. def createMarginCriterion(margin: Double = 1.0, sizeAverage: Boolean = true, squared: Boolean = false): MarginCriterion[T]
  150. def createMarginRankingCriterion(margin: Double = 1.0, sizeAverage: Boolean = true): MarginRankingCriterion[T]
  151. def createMaskedSelect(): MaskedSelect[T]
  152. def createMasking(maskValue: Double): Masking[T]
  153. def createMatToFloats(validHeight: Int = 300, validWidth: Int = 300, validChannels: Int = 3, outKey: String = ImageFeature.floats, shareBuffer: Boolean = true): MatToFloats
  154. def createMatToTensor(toRGB: Boolean = false, tensorKey: String = ImageFeature.imageTensor): MatToTensor[T]
  155. def createMax(dim: Int = 1, numInputDims: Int = Int.MinValue): Max[T]
  156. def createMaxEpoch(max: Int): Trigger
  157. def createMaxIteration(max: Int): Trigger
  158. def createMaxScore(max: Float): Trigger
  159. def createMaxout(inputSize: Int, outputSize: Int, maxoutNumber: Int, withBias: Boolean = true, wRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null, initWeight: Tensor[T] = null, initBias: Tensor[T] = null): Maxout[T]
  160. def createMean(dimension: Int = 1, nInputDims: Int = -1, squeeze: Boolean = true): Mean[T]
  161. def createMeanAbsolutePercentageCriterion: MeanAbsolutePercentageCriterion[T]
  162. def createMeanAveragePrecision(k: Int, classes: Int): ValidationMethod[T]
  163. def createMeanAveragePrecisionObjectDetection(classes: Int, iou: Float, useVoc2007: Boolean, skipClass: Int): ValidationMethod[T]
  164. def createMeanSquaredLogarithmicCriterion: MeanSquaredLogarithmicCriterion[T]
  165. def createMin(dim: Int = 1, numInputDims: Int = Int.MinValue): Min[T]
  166. def createMinLoss(min: Float): Trigger
  167. def createMixtureTable(dim: Int = Int.MaxValue): MixtureTable[T]
  168. def createModel(input: List[ModuleNode[T]], output: List[ModuleNode[T]]): Graph[T]
  169. def createModelPreprocessor(preprocessor: AbstractModule[Activity, Activity, T], trainable: AbstractModule[Activity, Activity, T]): Graph[T]
  170. def createMsraFiller(varianceNormAverage: Boolean = true): MsraFiller
  171. def createMul(): Mul[T]
  172. def createMulConstant(scalar: Double, inplace: Boolean = false): MulConstant[T]
  173. def createMultiCriterion(): MultiCriterion[T]
  174. def createMultiLabelMarginCriterion(sizeAverage: Boolean = true): MultiLabelMarginCriterion[T]
  175. def createMultiLabelSoftMarginCriterion(weights: JTensor = null, sizeAverage: Boolean = true): MultiLabelSoftMarginCriterion[T]
  176. def createMultiMarginCriterion(p: Int = 1, weights: JTensor = null, margin: Double = 1.0, sizeAverage: Boolean = true): MultiMarginCriterion[T]
  177. def createMultiRNNCell(cells: List[Cell[T]]): MultiRNNCell[T]
  178. def createMultiStep(stepSizes: List[Int], gamma: Double): MultiStep
  179. def createNDCG(k: Int = 10, negNum: Int = 100): ValidationMethod[T]
  180. def createNarrow(dimension: Int, offset: Int, length: Int = 1): Narrow[T]
  181. def createNarrowTable(offset: Int, length: Int = 1): NarrowTable[T]
  182. def createNegative(inplace: Boolean): Negative[T]
  183. def createNegativeEntropyPenalty(beta: Double): NegativeEntropyPenalty[T]
  184. def createNode(module: AbstractModule[Activity, Activity, T], x: List[ModuleNode[T]]): ModuleNode[T]
  185. def createNormalize(p: Double, eps: Double = 1e-10): Normalize[T]
  186. def createNormalizeScale(p: Double, eps: Double = 1e-10, scale: Double, size: List[Int], wRegularizer: Regularizer[T] = null): NormalizeScale[T]
  187. def createOnes(): Ones.type
  188. def createPGCriterion(sizeAverage: Boolean = false): PGCriterion[T]
  189. def createPReLU(nOutputPlane: Int = 0): PReLU[T]
  190. def createPack(dimension: Int): Pack[T]
  191. def createPadding(dim: Int, pad: Int, nInputDim: Int, value: Double = 0.0, nIndex: Int = 1): Padding[T]
  192. def createPairwiseDistance(norm: Int = 2): PairwiseDistance[T]
  193. def createParallelAdam(learningRate: Double = 1e-3, learningRateDecay: Double = 0.0, beta1: Double = 0.9, beta2: Double = 0.999, Epsilon: Double = 1e-8, parallelNum: Int = Engine.coreNumber()): ParallelAdam[T]
  194. def createParallelCriterion(repeatTarget: Boolean = false): ParallelCriterion[T]
  195. def createParallelTable(): ParallelTable[T]
  196. def createPipeline(list: List[FeatureTransformer]): FeatureTransformer
  197. def createPixelBytesToMat(byteKey: String): PixelBytesToMat
  198. def createPixelNormalize(means: List[Double]): PixelNormalizer
  199. def createPlateau(monitor: String, factor: Float = 0.1f, patience: Int = 10, mode: String = "min", epsilon: Float = 1e-4f, cooldown: Int = 0, minLr: Float = 0): Plateau
  200. def createPoissonCriterion: PoissonCriterion[T]
  201. def createPoly(power: Double, maxIteration: Int): Poly
  202. def createPooler(resolution: Int, scales: List[Double], sampling_ratio: Int): Pooler[T]
  203. def createPower(power: Double, scale: Double = 1, shift: Double = 0): Power[T]
  204. def createPriorBox(minSizes: List[Double], maxSizes: List[Double] = null, aspectRatios: List[Double] = null, isFlip: Boolean = true, isClip: Boolean = false, variances: List[Double] = null, offset: Float = 0.5f, imgH: Int = 0, imgW: Int = 0, imgSize: Int = 0, stepH: Float = 0, stepW: Float = 0, step: Float = 0): PriorBox[T]
  205. def createProposal(preNmsTopN: Int, postNmsTopN: Int, ratios: List[Double], scales: List[Double], rpnPreNmsTopNTrain: Int = 12000, rpnPostNmsTopNTrain: Int = 2000): Proposal
  206. def createRMSprop(learningRate: Double = 1e-2, learningRateDecay: Double = 0.0, decayRate: Double = 0.99, Epsilon: Double = 1e-8): RMSprop[T]
  207. def createRReLU(lower: Double = 1.0 / 8, upper: Double = 1.0 / 3, inplace: Boolean = false): RReLU[T]
  208. def createRandomAlterAspect(min_area_ratio: Float, max_area_ratio: Int, min_aspect_ratio_change: Float, interp_mode: String, cropLength: Int): RandomAlterAspect
  209. def createRandomAspectScale(scales: List[Int], scaleMultipleOf: Int = 1, maxSize: Int = 1000): RandomAspectScale
  210. def createRandomCrop(cropWidth: Int, cropHeight: Int, isClip: Boolean): RandomCrop
  211. def createRandomCropper(cropWidth: Int, cropHeight: Int, mirror: Boolean, cropperMethod: String, channels: Int): RandomCropper
  212. def createRandomNormal(mean: Double, stdv: Double): RandomNormal
  213. def createRandomResize(minSize: Int, maxSize: Int): RandomResize
  214. def createRandomSampler(): FeatureTransformer
  215. def createRandomTransformer(transformer: FeatureTransformer, prob: Double): RandomTransformer
  216. def createRandomUniform(): InitializationMethod
  217. def createRandomUniform(lower: Double, upper: Double): InitializationMethod
  218. def createReLU(ip: Boolean = false): ReLU[T]
  219. def createReLU6(inplace: Boolean = false): ReLU6[T]
  220. def createRecurrent(): Recurrent[T]
  221. def createRecurrentDecoder(outputLength: Int): RecurrentDecoder[T]
  222. def createReplicate(nFeatures: Int, dim: Int = 1, nDim: Int = Int.MaxValue): Replicate[T]
  223. def createReshape(size: List[Int], batchMode: Boolean = null): Reshape[T]
  224. def createResize(resizeH: Int, resizeW: Int, resizeMode: Int = Imgproc.INTER_LINEAR, useScaleFactor: Boolean): Resize
  225. def createResizeBilinear(outputHeight: Int, outputWidth: Int, alignCorner: Boolean, dataFormat: String): ResizeBilinear[T]
  226. def createReverse(dimension: Int = 1, isInplace: Boolean = false): Reverse[T]
  227. def createRnnCell(inputSize: Int, hiddenSize: Int, activation: TensorModule[T], isInputWithBias: Boolean = true, isHiddenWithBias: Boolean = true, wRegularizer: Regularizer[T] = null, uRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null): RnnCell[T]
  228. def createRoiAlign(spatial_scale: Double, sampling_ratio: Int, pooled_h: Int, pooled_w: Int): RoiAlign[T]
  229. def createRoiHFlip(normalized: Boolean = true): RoiHFlip
  230. def createRoiNormalize(): RoiNormalize
  231. def createRoiPooling(pooled_w: Int, pooled_h: Int, spatial_scale: Double): RoiPooling[T]
  232. def createRoiProject(needMeetCenterConstraint: Boolean): RoiProject
  233. def createRoiResize(normalized: Boolean): RoiResize
  234. def createSGD(learningRate: Double = 1e-3, learningRateDecay: Double = 0.0, weightDecay: Double = 0.0, momentum: Double = 0.0, dampening: Double = Double.MaxValue, nesterov: Boolean = false, leaningRateSchedule: LearningRateSchedule = SGD.Default(), learningRates: JTensor = null, weightDecays: JTensor = null): SGD[T]
  235. def createSReLU(shape: ArrayList[Int], shareAxes: ArrayList[Int] = null): SReLU[T]
  236. def createSaturation(deltaLow: Double, deltaHigh: Double): Saturation
  237. def createScale(size: List[Int]): Scale[T]
  238. def createSelect(dimension: Int, index: Int): Select[T]
  239. def createSelectTable(dimension: Int): SelectTable[T]
  240. def createSequenceBeamSearch(vocabSize: Int, beamSize: Int, alpha: Float, decodeLength: Int, eosId: Float, paddingValue: Float, numHiddenLayers: Int, hiddenSize: Int): SequenceBeamSearch[T]
  241. def createSequential(): Container[Activity, Activity, T]
  242. def createSequentialSchedule(iterationPerEpoch: Int): SequentialSchedule
  243. def createSeveralIteration(interval: Int): Trigger
  244. def createSigmoid(): Sigmoid[T]
  245. def createSmoothL1Criterion(sizeAverage: Boolean = true): SmoothL1Criterion[T]
  246. def createSmoothL1CriterionWithWeights(sigma: Double, num: Int = 0): SmoothL1CriterionWithWeights[T]
  247. def createSoftMarginCriterion(sizeAverage: Boolean = true): SoftMarginCriterion[T]
  248. def createSoftMax(pos: Int = 1): SoftMax[T]
  249. def createSoftMin(): SoftMin[T]
  250. def createSoftPlus(beta: Double = 1.0): SoftPlus[T]
  251. def createSoftShrink(lambda: Double = 0.5): SoftShrink[T]
  252. def createSoftSign(): SoftSign[T]
  253. def createSoftmaxWithCriterion(ignoreLabel: Integer = null, normalizeMode: String = "VALID"): SoftmaxWithCriterion[T]
  254. def createSparseJoinTable(dimension: Int): SparseJoinTable[T]
  255. def createSparseLinear(inputSize: Int, outputSize: Int, withBias: Boolean, backwardStart: Int = -1, backwardLength: Int = -1, wRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null, initWeight: JTensor = null, initBias: JTensor = null, initGradWeight: JTensor = null, initGradBias: JTensor = null): SparseLinear[T]
  256. def createSpatialAveragePooling(kW: Int, kH: Int, dW: Int = 1, dH: Int = 1, padW: Int = 0, padH: Int = 0, globalPooling: Boolean = false, ceilMode: Boolean = false, countIncludePad: Boolean = true, divide: Boolean = true, format: String = "NCHW"): SpatialAveragePooling[T]
  257. def createSpatialBatchNormalization(nOutput: Int, eps: Double = 1e-5, momentum: Double = 0.1, affine: Boolean = true, initWeight: JTensor = null, initBias: JTensor = null, initGradWeight: JTensor = null, initGradBias: JTensor = null, dataFormat: String = "NCHW"): SpatialBatchNormalization[T]
  258. def createSpatialContrastiveNormalization(nInputPlane: Int = 1, kernel: JTensor = null, threshold: Double = 1e-4, thresval: Double = 1e-4): SpatialContrastiveNormalization[T]
  259. def createSpatialConvolution(nInputPlane: Int, nOutputPlane: Int, kernelW: Int, kernelH: Int, strideW: Int = 1, strideH: Int = 1, padW: Int = 0, padH: Int = 0, nGroup: Int = 1, propagateBack: Boolean = true, wRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null, initWeight: JTensor = null, initBias: JTensor = null, initGradWeight: JTensor = null, initGradBias: JTensor = null, withBias: Boolean = true, dataFormat: String = "NCHW"): SpatialConvolution[T]
  260. def createSpatialConvolutionMap(connTable: JTensor, kW: Int, kH: Int, dW: Int = 1, dH: Int = 1, padW: Int = 0, padH: Int = 0, wRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null): SpatialConvolutionMap[T]
  261. def createSpatialCrossMapLRN(size: Int = 5, alpha: Double = 1.0, beta: Double = 0.75, k: Double = 1.0, dataFormat: String = "NCHW"): SpatialCrossMapLRN[T]
  262. def createSpatialDilatedConvolution(nInputPlane: Int, nOutputPlane: Int, kW: Int, kH: Int, dW: Int = 1, dH: Int = 1, padW: Int = 0, padH: Int = 0, dilationW: Int = 1, dilationH: Int = 1, wRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null): SpatialDilatedConvolution[T]
  263. def createSpatialDivisiveNormalization(nInputPlane: Int = 1, kernel: JTensor = null, threshold: Double = 1e-4, thresval: Double = 1e-4): SpatialDivisiveNormalization[T]
  264. def createSpatialDropout1D(initP: Double = 0.5): SpatialDropout1D[T]
  265. def createSpatialDropout2D(initP: Double = 0.5, dataFormat: String = "NCHW"): SpatialDropout2D[T]
  266. def createSpatialDropout3D(initP: Double = 0.5, dataFormat: String = "NCHW"): SpatialDropout3D[T]
  267. def createSpatialFullConvolution(nInputPlane: Int, nOutputPlane: Int, kW: Int, kH: Int, dW: Int = 1, dH: Int = 1, padW: Int = 0, padH: Int = 0, adjW: Int = 0, adjH: Int = 0, nGroup: Int = 1, noBias: Boolean = false, wRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null): SpatialFullConvolution[T]
  268. def createSpatialMaxPooling(kW: Int, kH: Int, dW: Int, dH: Int, padW: Int = 0, padH: Int = 0, ceilMode: Boolean = false, format: String = "NCHW"): SpatialMaxPooling[T]
  269. def createSpatialSeparableConvolution(nInputChannel: Int, nOutputChannel: Int, depthMultiplier: Int, kW: Int, kH: Int, sW: Int = 1, sH: Int = 1, pW: Int = 0, pH: Int = 0, withBias: Boolean = true, dataFormat: String = "NCHW", wRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null, pRegularizer: Regularizer[T] = null): SpatialSeparableConvolution[T]
  270. def createSpatialShareConvolution(nInputPlane: Int, nOutputPlane: Int, kernelW: Int, kernelH: Int, strideW: Int = 1, strideH: Int = 1, padW: Int = 0, padH: Int = 0, nGroup: Int = 1, propagateBack: Boolean = true, wRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null, initWeight: JTensor = null, initBias: JTensor = null, initGradWeight: JTensor = null, initGradBias: JTensor = null, withBias: Boolean = true): SpatialShareConvolution[T]
  271. def createSpatialSubtractiveNormalization(nInputPlane: Int = 1, kernel: JTensor = null): SpatialSubtractiveNormalization[T]
  272. def createSpatialWithinChannelLRN(size: Int = 5, alpha: Double = 1.0, beta: Double = 0.75): SpatialWithinChannelLRN[T]
  273. def createSpatialZeroPadding(padLeft: Int, padRight: Int, padTop: Int, padBottom: Int): SpatialZeroPadding[T]
  274. def createSplitTable(dimension: Int, nInputDims: Int = -1): SplitTable[T]
  275. def createSqrt(): Sqrt[T]
  276. def createSquare(): Square[T]
  277. def createSqueeze(dim: Int = Int.MinValue, numInputDims: Int = Int.MinValue): Squeeze[T]
  278. def createStep(stepSize: Int, gamma: Double): Step
  279. def createSum(dimension: Int = 1, nInputDims: Int = -1, sizeAverage: Boolean = false, squeeze: Boolean = true): Sum[T]
  280. def createTableOperation(operationLayer: AbstractModule[Table, Tensor[T], T]): TableOperation[T]
  281. def createTanh(): Tanh[T]
  282. def createTanhShrink(): TanhShrink[T]
  283. def createTemporalConvolution(inputFrameSize: Int, outputFrameSize: Int, kernelW: Int, strideW: Int = 1, propagateBack: Boolean = true, wRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null, initWeight: JTensor = null, initBias: JTensor = null, initGradWeight: JTensor = null, initGradBias: JTensor = null): TemporalConvolution[T]
  284. def createTemporalMaxPooling(kW: Int, dW: Int): TemporalMaxPooling[T]
  285. def createThreshold(th: Double = 1e-6, v: Double = 0.0, ip: Boolean = false): Threshold[T]
  286. def createTile(dim: Int, copies: Int): Tile[T]
  287. def createTimeDistributed(layer: TensorModule[T]): TimeDistributed[T]
  288. def createTimeDistributedCriterion(critrn: TensorCriterion[T], sizeAverage: Boolean = false, dimension: Int = 2): TimeDistributedCriterion[T]
  289. def createTimeDistributedMaskCriterion(critrn: TensorCriterion[T], paddingValue: Int = 0): TimeDistributedMaskCriterion[T]
  290. def createTop1Accuracy(): ValidationMethod[T]
  291. def createTop5Accuracy(): ValidationMethod[T]
  292. def createTrainSummary(logDir: String, appName: String): TrainSummary
  293. def createTransformer(vocabSize: Int, hiddenSize: Int, numHeads: Int, filterSize: Int, numHiddenlayers: Int, postprocessDropout: Double, attentionDropout: Double, reluDropout: Double): Transformer[T]
  294. def createTransformerCriterion(criterion: AbstractCriterion[Activity, Activity, T], inputTransformer: AbstractModule[Activity, Activity, T] = null, targetTransformer: AbstractModule[Activity, Activity, T] = null): TransformerCriterion[T]
  295. def createTranspose(permutations: List[List[Int]]): Transpose[T]
  296. def createTreeNNAccuracy(): ValidationMethod[T]
  297. def createTriggerAnd(first: Trigger, others: List[Trigger]): Trigger
  298. def createTriggerOr(first: Trigger, others: List[Trigger]): Trigger
  299. def createUnsqueeze(pos: List[Int], numInputDims: Int = Int.MinValue): Unsqueeze[T]
  300. def createUpSampling1D(length: Int): UpSampling1D[T]
  301. def createUpSampling2D(size: List[Int], dataFormat: String): UpSampling2D[T]
  302. def createUpSampling3D(size: List[Int]): UpSampling3D[T]
  303. def createValidationSummary(logDir: String, appName: String): ValidationSummary
  304. def createView(sizes: List[Int], num_input_dims: Int = 0): View[T]
  305. def createVolumetricAveragePooling(kT: Int, kW: Int, kH: Int, dT: Int, dW: Int, dH: Int, padT: Int = 0, padW: Int = 0, padH: Int = 0, countIncludePad: Boolean = true, ceilMode: Boolean = false): VolumetricAveragePooling[T]
  306. def createVolumetricConvolution(nInputPlane: Int, nOutputPlane: Int, kT: Int, kW: Int, kH: Int, dT: Int = 1, dW: Int = 1, dH: Int = 1, padT: Int = 0, padW: Int = 0, padH: Int = 0, withBias: Boolean = true, wRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null): VolumetricConvolution[T]
  307. def createVolumetricFullConvolution(nInputPlane: Int, nOutputPlane: Int, kT: Int, kW: Int, kH: Int, dT: Int = 1, dW: Int = 1, dH: Int = 1, padT: Int = 0, padW: Int = 0, padH: Int = 0, adjT: Int = 0, adjW: Int = 0, adjH: Int = 0, nGroup: Int = 1, noBias: Boolean = false, wRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null): VolumetricFullConvolution[T]
  308. def createVolumetricMaxPooling(kT: Int, kW: Int, kH: Int, dT: Int, dW: Int, dH: Int, padT: Int = 0, padW: Int = 0, padH: Int = 0): VolumetricMaxPooling[T]
  309. def createWarmup(delta: Double): Warmup
  310. def createXavier(): Xavier.type
  311. def createZeros(): Zeros.type
  312. def criterionBackward(criterion: AbstractCriterion[Activity, Activity, T], input: List[_ <: AnyRef], inputIsTable: Boolean, target: List[_ <: AnyRef], targetIsTable: Boolean): List[JTensor]
  313. def criterionForward(criterion: AbstractCriterion[Activity, Activity, T], input: List[_ <: AnyRef], inputIsTable: Boolean, target: List[_ <: AnyRef], targetIsTable: Boolean): T
  314. def disableClip(optimizer: Optimizer[T, MiniBatch[T]]): Unit
  315. def distributedImageFrameRandomSplit(imageFrame: DistributedImageFrame, weights: List[Double]): Array[ImageFrame]
  316. def distributedImageFrameToImageTensorRdd(imageFrame: DistributedImageFrame, floatKey: String = ImageFeature.floats, toChw: Boolean = true): JavaRDD[JTensor]
  317. def distributedImageFrameToLabelTensorRdd(imageFrame: DistributedImageFrame): JavaRDD[JTensor]
  318. def distributedImageFrameToPredict(imageFrame: DistributedImageFrame, key: String): JavaRDD[List[Any]]
  319. def distributedImageFrameToSample(imageFrame: DistributedImageFrame, key: String): JavaRDD[Sample]
  320. def distributedImageFrameToUri(imageFrame: DistributedImageFrame, key: String): JavaRDD[String]
  321. def dlClassifierModelTransform(dlClassifierModel: DLClassifierModel[T], dataSet: DataFrame): DataFrame
  322. def dlImageTransform(dlImageTransformer: DLImageTransformer, dataSet: DataFrame): DataFrame
  323. def dlModelTransform(dlModel: DLModel[T], dataSet: DataFrame): DataFrame
  324. def dlReadImage(path: String, sc: JavaSparkContext, minParitions: Int): DataFrame
  325. final def eq(arg0: AnyRef): Boolean
    Definition Classes
    AnyRef
  326. def equals(arg0: Any): Boolean
    Definition Classes
    AnyRef → Any
  327. def evaluate(module: AbstractModule[Activity, Activity, T]): AbstractModule[Activity, Activity, T]
  328. def featureTransformDataset(dataset: DataSet[ImageFeature], transformer: FeatureTransformer): DataSet[ImageFeature]
  329. def finalize(): Unit
    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( classOf[java.lang.Throwable] )
  330. def findGraphNode(model: Graph[T], name: String): ModuleNode[T]
  331. def fitClassifier(classifier: DLClassifier[T], dataSet: DataFrame): DLModel[T]
  332. def fitEstimator(estimator: DLEstimator[T], dataSet: DataFrame): DLModel[T]
  333. def freeze(model: AbstractModule[Activity, Activity, T], freezeLayers: List[String]): AbstractModule[Activity, Activity, T]
  334. final def getClass(): Class[_]
    Definition Classes
    AnyRef → Any
    Annotations
    @native()
  335. def getContainerModules(module: Container[Activity, Activity, T]): List[AbstractModule[Activity, Activity, T]]
  336. def getEngineType(): String
  337. def getFlattenModules(module: Container[Activity, Activity, T], includeContainer: Boolean): List[AbstractModule[Activity, Activity, T]]
  338. def getHiddenState(rec: Recurrent[T]): JActivity
  339. def getNodeAndCoreNumber(): Array[Int]
  340. def getOptimizerVersion(): String
  341. def getRealClassNameOfJValue(module: AbstractModule[Activity, Activity, T]): String
  342. def getRunningMean(module: BatchNormalization[T]): JTensor
  343. def getRunningStd(module: BatchNormalization[T]): JTensor
  344. def getWeights(model: AbstractModule[Activity, Activity, T]): List[JTensor]
  345. def hashCode(): Int
    Definition Classes
    AnyRef → Any
    Annotations
    @native()
  346. def imageFeatureGetKeys(imageFeature: ImageFeature): List[String]
  347. def imageFeatureToImageTensor(imageFeature: ImageFeature, floatKey: String = ImageFeature.floats, toChw: Boolean = true): JTensor
  348. def imageFeatureToLabelTensor(imageFeature: ImageFeature): JTensor
  349. def initEngine(): Unit
  350. def isDistributed(imageFrame: ImageFrame): Boolean
  351. final def isInstanceOf[T0]: Boolean
    Definition Classes
    Any
  352. def isLocal(imageFrame: ImageFrame): Boolean
  353. def isWithWeights(module: Module[T]): Boolean
  354. def jTensorsToActivity(input: List[_ <: AnyRef], isTable: Boolean): Activity
  355. def loadBigDL(path: String): AbstractModule[Activity, Activity, T]
  356. def loadBigDLModule(modulePath: String, weightPath: String): AbstractModule[Activity, Activity, T]
  357. def loadCaffe(model: AbstractModule[Activity, Activity, T], defPath: String, modelPath: String, matchAll: Boolean = true): AbstractModule[Activity, Activity, T]
  358. def loadCaffeModel(defPath: String, modelPath: String): AbstractModule[Activity, Activity, T]
  359. def loadOptimMethod(path: String): OptimMethod[T]
  360. def loadTF(path: String, inputs: List[String], outputs: List[String], byteOrder: String, binFile: String = null, generatedBackward: Boolean = true): AbstractModule[Activity, Activity, T]
  361. def loadTorch(path: String): AbstractModule[Activity, Activity, T]
  362. def localImageFrameToImageTensor(imageFrame: LocalImageFrame, floatKey: String = ImageFeature.floats, toChw: Boolean = true): List[JTensor]
  363. def localImageFrameToLabelTensor(imageFrame: LocalImageFrame): List[JTensor]
  364. def localImageFrameToPredict(imageFrame: LocalImageFrame, key: String): List[List[Any]]
  365. def localImageFrameToSample(imageFrame: LocalImageFrame, key: String): List[Sample]
  366. def localImageFrameToUri(imageFrame: LocalImageFrame, key: String): List[String]
  367. def modelBackward(model: AbstractModule[Activity, Activity, T], input: List[_ <: AnyRef], inputIsTable: Boolean, gradOutput: List[_ <: AnyRef], gradOutputIsTable: Boolean): List[JTensor]
  368. def modelEvaluate(model: AbstractModule[Activity, Activity, T], valRDD: JavaRDD[Sample], batchSize: Int, valMethods: List[ValidationMethod[T]]): List[EvaluatedResult]
  369. def modelEvaluateImageFrame(model: AbstractModule[Activity, Activity, T], imageFrame: ImageFrame, batchSize: Int, valMethods: List[ValidationMethod[T]]): List[EvaluatedResult]
  370. def modelForward(model: AbstractModule[Activity, Activity, T], input: List[_ <: AnyRef], inputIsTable: Boolean): List[JTensor]
  371. def modelGetParameters(model: AbstractModule[Activity, Activity, T]): Map[Any, Map[Any, List[List[Any]]]]
  372. def modelPredictClass(model: AbstractModule[Activity, Activity, T], dataRdd: JavaRDD[Sample]): JavaRDD[Int]
  373. def modelPredictImage(model: AbstractModule[Activity, Activity, T], imageFrame: ImageFrame, featLayerName: String, shareBuffer: Boolean, batchPerPartition: Int, predictKey: String): ImageFrame
  374. def modelPredictRDD(model: AbstractModule[Activity, Activity, T], dataRdd: JavaRDD[Sample], batchSize: Int = -1): JavaRDD[JTensor]
  375. def modelSave(module: AbstractModule[Activity, Activity, T], path: String, overWrite: Boolean): Unit
  376. final def ne(arg0: AnyRef): Boolean
    Definition Classes
    AnyRef
  377. final def notify(): Unit
    Definition Classes
    AnyRef
    Annotations
    @native()
  378. final def notifyAll(): Unit
    Definition Classes
    AnyRef
    Annotations
    @native()
  379. def predictLocal(model: AbstractModule[Activity, Activity, T], features: List[JTensor], batchSize: Int = -1): List[JTensor]
  380. def predictLocalClass(model: AbstractModule[Activity, Activity, T], features: List[JTensor]): List[Int]
  381. def quantize(module: AbstractModule[Activity, Activity, T]): Module[T]
  382. def read(path: String, sc: JavaSparkContext, minPartitions: Int): ImageFrame
  383. def readParquet(path: String, sc: JavaSparkContext): DistributedImageFrame
  384. def redirectSparkLogs(logPath: String): Unit
  385. def saveBigDLModule(module: AbstractModule[Activity, Activity, T], modulePath: String, weightPath: String, overWrite: Boolean): Unit
  386. def saveCaffe(module: AbstractModule[Activity, Activity, T], prototxtPath: String, modelPath: String, useV2: Boolean = true, overwrite: Boolean = false): Unit
  387. def saveGraphTopology(model: Graph[T], logPath: String): Graph[T]
  388. def saveOptimMethod(method: OptimMethod[T], path: String, overWrite: Boolean = false): Unit
  389. def saveTF(model: AbstractModule[Activity, Activity, T], inputs: List[Any], path: String, byteOrder: String, dataFormat: String): Unit
  390. def saveTensorDictionary(tensors: HashMap[String, JTensor], path: String): Unit

    Save tensor dictionary to a Java hashmap object file

  391. def seqFilesToImageFrame(url: String, sc: JavaSparkContext, classNum: Int, partitionNum: Int): ImageFrame
  392. def setBatchSizeDLClassifier(classifier: DLClassifier[T], batchSize: Int): DLClassifier[T]
  393. def setBatchSizeDLClassifierModel(dlClassifierModel: DLClassifierModel[T], batchSize: Int): DLClassifierModel[T]
  394. def setBatchSizeDLEstimator(estimator: DLEstimator[T], batchSize: Int): DLEstimator[T]
  395. def setBatchSizeDLModel(dlModel: DLModel[T], batchSize: Int): DLModel[T]
  396. def setCheckPoint(optimizer: Optimizer[T, MiniBatch[T]], trigger: Trigger, checkPointPath: String, isOverwrite: Boolean): Unit
  397. def setConstantClip(optimizer: Optimizer[T, MiniBatch[T]], min: Float, max: Float): Unit
  398. def setCriterion(optimizer: Optimizer[T, MiniBatch[T]], criterion: Criterion[T]): Unit
  399. def setFeatureSizeDLClassifierModel(dlClassifierModel: DLClassifierModel[T], featureSize: ArrayList[Int]): DLClassifierModel[T]
  400. def setFeatureSizeDLModel(dlModel: DLModel[T], featureSize: ArrayList[Int]): DLModel[T]
  401. def setInitMethod(layer: Initializable, initMethods: ArrayList[InitializationMethod]): layer.type
  402. def setInitMethod(layer: Initializable, weightInitMethod: InitializationMethod, biasInitMethod: InitializationMethod): layer.type
  403. def setInputFormats(graph: StaticGraph[T], inputFormat: List[Int]): StaticGraph[T]
  404. def setL2NormClip(optimizer: Optimizer[T, MiniBatch[T]], normValue: Float): Unit
  405. def setLabel(labelMap: Map[String, Float], imageFrame: ImageFrame): Unit
  406. def setLearningRateDLClassifier(classifier: DLClassifier[T], lr: Double): DLClassifier[T]
  407. def setLearningRateDLEstimator(estimator: DLEstimator[T], lr: Double): DLEstimator[T]
  408. def setMaxEpochDLClassifier(classifier: DLClassifier[T], maxEpoch: Int): DLClassifier[T]
  409. def setMaxEpochDLEstimator(estimator: DLEstimator[T], maxEpoch: Int): DLEstimator[T]
  410. def setModelSeed(seed: Long): Unit
  411. def setOptimizerVersion(version: String): Unit
  412. def setOutputFormats(graph: StaticGraph[T], outputFormat: List[Int]): StaticGraph[T]
  413. def setRunningMean(module: BatchNormalization[T], runningMean: JTensor): Unit
  414. def setRunningStd(module: BatchNormalization[T], runningStd: JTensor): Unit
  415. def setStopGradient(model: Graph[T], layers: List[String]): Graph[T]
  416. def setTrainData(optimizer: Optimizer[T, MiniBatch[T]], trainingRdd: JavaRDD[Sample], batchSize: Int): Unit
  417. def setTrainSummary(optimizer: Optimizer[T, MiniBatch[T]], summary: TrainSummary): Unit
  418. def setValSummary(optimizer: Optimizer[T, MiniBatch[T]], summary: ValidationSummary): Unit
  419. def setValidation(optimizer: Optimizer[T, MiniBatch[T]], batchSize: Int, trigger: Trigger, xVal: List[JTensor], yVal: JTensor, vMethods: List[ValidationMethod[T]]): Unit
  420. def setValidation(optimizer: Optimizer[T, MiniBatch[T]], batchSize: Int, trigger: Trigger, valRdd: JavaRDD[Sample], vMethods: List[ValidationMethod[T]]): Unit
  421. def setValidationFromDataSet(optimizer: Optimizer[T, MiniBatch[T]], batchSize: Int, trigger: Trigger, valDataSet: DataSet[ImageFeature], vMethods: List[ValidationMethod[T]]): Unit
  422. def setWeights(model: AbstractModule[Activity, Activity, T], weights: List[JTensor]): Unit
  423. def showBigDlInfoLogs(): Unit
  424. def summaryReadScalar(summary: Summary, tag: String): List[List[Any]]
  425. def summarySetTrigger(summary: TrainSummary, summaryName: String, trigger: Trigger): TrainSummary
  426. final def synchronized[T0](arg0: ⇒ T0): T0
    Definition Classes
    AnyRef
  427. def testSample(sample: Sample): Sample
  428. def testTensor(jTensor: JTensor): JTensor
  429. def toGraph(sequential: Sequential[T]): StaticGraph[T]
  430. def toJSample(psamples: RDD[Sample]): RDD[dataset.Sample[T]]
  431. def toJSample(record: Sample): dataset.Sample[T]
  432. def toJTensor(tensor: Tensor[T]): JTensor
  433. def toPySample(sample: dataset.Sample[T]): Sample
  434. def toSampleArray(Xs: List[Tensor[T]], y: Tensor[T] = null): Array[dataset.Sample[T]]
  435. def toString(): String
    Definition Classes
    AnyRef → Any
  436. def toTensor(jTensor: JTensor): Tensor[T]
  437. def trainTF(modelPath: String, output: String, samples: JavaRDD[Sample], optMethod: OptimMethod[T], criterion: Criterion[T], batchSize: Int, endWhen: Trigger): AbstractModule[Activity, Activity, T]
  438. def transformImageFeature(transformer: FeatureTransformer, feature: ImageFeature): ImageFeature
  439. def transformImageFrame(transformer: FeatureTransformer, imageFrame: ImageFrame): ImageFrame
  440. def unFreeze(model: AbstractModule[Activity, Activity, T], names: List[String]): AbstractModule[Activity, Activity, T]
  441. def uniform(a: Double, b: Double, size: List[Int]): JTensor
  442. def updateParameters(model: AbstractModule[Activity, Activity, T], lr: Double): Unit
  443. final def wait(): Unit
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  444. final def wait(arg0: Long, arg1: Int): Unit
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  445. final def wait(arg0: Long): Unit
    Definition Classes
    AnyRef
    Annotations
    @native() @throws( ... )
  446. def writeParquet(path: String, output: String, sc: JavaSparkContext, partitionNum: Int = 1): Unit

Inherited from Serializable

Inherited from Serializable

Inherited from AnyRef

Inherited from Any

Ungrouped