Function sparseToDense

tfjs-backend-cpu/src/kernels/SparseToDense.ts:23–80

(args: {
  inputs: SparseToDenseInputs,
  backend: MathBackendCPU,
  attrs: SparseToDenseAttrs
})

import {scatterImpl} from './Scatter_impl';

export function sparseToDense(args: {
  inputs: SparseToDenseInputs,
  backend: MathBackendCPU,
  attrs: SparseToDenseAttrs
}): TensorInfo {
  const {inputs, backend, attrs} = args;
  const {sparseIndices, sparseValues, defaultValue} = inputs;
  const {outputShape} = attrs;

  const {sliceRank, numUpdates, sliceSize, strides, outputSize} =
      backend_util.calculateShapes(sparseValues, sparseIndices, outputShape);
  const sumDupeIndices = false;

  const indicesBuf = backend.bufferSync<Rank, 'int32'>(sparseIndices);

  let outBuf;
  switch (sparseValues.dtype) {
    case 'bool': {
      const updatesBuf = backend.bufferSync<Rank, 'bool'>(sparseValues);
      const $defaultValue =
          Boolean(backend.data.get(defaultValue.dataId).values[0]);
      outBuf = scatterImpl(
          indicesBuf, updatesBuf, outputShape, outputSize, sliceSize,
          numUpdates, sliceRank, strides, $defaultValue, sumDupeIndices);
      break;
    }
    case 'float32': {
      const updatesBuf = backend.bufferSync<Rank, 'float32'>(sparseValues);
      const $defaultValue =
          backend.data.get(defaultValue.dataId).values[0] as number;
      outBuf = scatterImpl(
          indicesBuf, updatesBuf, outputShape, outputSize, sliceSize,
          numUpdates, sliceRank, strides, $defaultValue, sumDupeIndices);
      break;
    }
    case 'int32': {
      const updatesBuf = backend.bufferSync<Rank, 'int32'>(sparseValues);
      const $defaultValue =
          backend.data.get(defaultValue.dataId).values[0] as number;
      outBuf = scatterImpl(
          indicesBuf, updatesBuf, outputShape, outputSize, sliceSize,
          numUpdates, sliceRank, strides, $defaultValue, sumDupeIndices);
      break;
    }
    case 'string': {
      const updatesBuf = backend.bufferSync<Rank, 'string'>(sparseValues);
      const $defaultValue = util.decodeString(
          backend.data.get(defaultValue.dataId).values[0] as Uint8Array);
      outBuf = scatterImpl(
          indicesBuf, updatesBuf, outputShape, outputSize, sliceSize,
          numUpdates, sliceRank, strides, $defaultValue, sumDupeIndices);
      break;
    }
    default:
      throw new Error(`Unsupported type ${sparseValues.dtype}`);
  }
  return backend.makeTensorInfo(outputShape, outBuf.dtype, outBuf.values);
}