Method evaluate

tfjs-layers/src/engine/training.ts:840–864

* Returns the loss value & metrics values for the model in test mode. * * Loss and metrics are specified during `compile()`, which needs to happen * before calls to `evaluate()`. * * Computation is done in batches. * * ```js * const model = tf.sequential({ * layers: [tf.

(
      x: Tensor|Tensor[], y: Tensor|Tensor[],
      args: ModelEvaluateArgs = {})

   * @doc {heading: 'Models', subheading: 'Classes'}
   */
  evaluate(
      x: Tensor|Tensor[], y: Tensor|Tensor[],
      args: ModelEvaluateArgs = {}): Scalar|Scalar[] {
    const batchSize = args.batchSize == null ? 32 : args.batchSize;
    checkBatchSize(batchSize);

    // TODO(cais): Standardize `config.sampleWeights` as well.
    // Validate user data.
    const checkBatchAxis = true;
    const standardizedOuts =
        this.standardizeUserDataXY(x, y, checkBatchAxis, batchSize);
    try {
      // TODO(cais): If uses `useLearningPhase`, set the corresponding element
      // of the input to 0.
      const ins = standardizedOuts[0].concat(standardizedOuts[1]);
      this.makeTestFunction();
      const f = this.testFunction;
      const testOuts =
          this.testLoop(f, ins, batchSize, args.verbose, args.steps);
      return singletonOrArray(testOuts);
    } finally {
      disposeNewTensors(standardizedOuts[0], x);
      disposeNewTensors(standardizedOuts[1], y);
    }
  }

  // TODO(cais): Add code snippet below once real dataset objects are
  //   available.