Examples with DeltaSet com.simiacryptus.mindseye.lang.DeltaSet used on opensource projects

Example 36 with DeltaSet

use of com.simiacryptus.mindseye.lang.DeltaSet in project MindsEye by SimiaCryptus.

the class TensorListTrainable method eval.

/**
 * Eval point sample.
 *
 * @param list    the list
 * @param monitor the monitor
 * @return the point sample
 */
@Nonnull
protected PointSample eval(@Nonnull final TensorList[] list, @Nullable final TrainingMonitor monitor) {
    int inputs = data.length;
    assert 0 < inputs;
    int items = data[0].length();
    assert 0 < items;
    @Nonnull final TimedResult<PointSample> timedResult = TimedResult.time(() -> {
        final Result[] nnContext = TensorListTrainable.getNNContext(list, mask);
        final Result result = network.eval(nnContext);
        for (@Nonnull Result nnResult : nnContext) {
            nnResult.getData().freeRef();
            nnResult.freeRef();
        }
        final TensorList resultData = result.getData();
        final DoubleSummaryStatistics statistics = resultData.stream().flatMapToDouble(x -> {
            double[] array = Arrays.stream(x.getData()).toArray();
            x.freeRef();
            return Arrays.stream(array);
        }).summaryStatistics();
        final double sum = statistics.getSum();
        @Nonnull final DeltaSet<Layer> deltaSet = new DeltaSet<Layer>();
        @Nonnull PointSample pointSample;
        try {
            result.accumulate(deltaSet, 1.0);
            // log.info(String.format("Evaluated to %s delta buffers, %s mag", DeltaSet<LayerBase>.getMap().size(), DeltaSet<LayerBase>.getMagnitude()));
            @Nonnull StateSet<Layer> stateSet = new StateSet<>(deltaSet);
            pointSample = new PointSample(deltaSet, stateSet, sum, 0.0, items);
            stateSet.freeRef();
        } finally {
            resultData.freeRef();
            result.freeRef();
            deltaSet.freeRef();
        }
        return pointSample;
    });
    if (null != monitor && verbosity() > 0) {
        monitor.log(String.format("Device completed %s items in %.3f sec", items, timedResult.timeNanos / 1e9));
    }
    @Nonnull PointSample normalize = timedResult.result.normalize();
    timedResult.result.freeRef();
    return normalize;
}

Example 37 with DeltaSet

use of com.simiacryptus.mindseye.lang.DeltaSet in project MindsEye by SimiaCryptus.

the class PerformanceTester method testPerformance.

/**
 * Test learning performance double statistics.
 *
 * @param component      the component
 * @param inputPrototype the input prototype
 * @return the double statistics
 */
@Nonnull
protected Tuple2<Double, Double> testPerformance(@Nonnull final Layer component, final Tensor... inputPrototype) {
    final Tensor[][] data = IntStream.range(0, batches).mapToObj(x -> x).flatMap(x -> Stream.<Tensor[]>of(inputPrototype)).toArray(i -> new Tensor[i][]);
    @Nonnull TimedResult<Result> timedEval = TimedResult.time(() -> {
        Result[] input = ConstantResult.batchResultArray(data);
        @Nullable Result result;
        try {
            result = component.eval(input);
        } finally {
            for (@Nonnull Result nnResult : input) {
                nnResult.freeRef();
                nnResult.getData().freeRef();
            }
        }
        return result;
    });
    final Result result = timedEval.result;
    @Nonnull final DeltaSet<Layer> buffer = new DeltaSet<Layer>();
    try {
        long timedBackprop = TimedResult.time(() -> {
            @Nonnull TensorArray tensorArray = TensorArray.wrap(result.getData().stream().map(x -> {
                return x.mapAndFree(v -> 1.0);
            }).toArray(i -> new Tensor[i]));
            result.accumulate(buffer, tensorArray);
            assert tensorArray.currentRefCount() == 0;
            return buffer;
        }).timeNanos;
        return new Tuple2<>(timedEval.timeNanos / 1e9, timedBackprop / 1e9);
    } finally {
        buffer.freeRef();
        result.freeRef();
        result.getData().freeRef();
    }
}

Example 38 with DeltaSet

use of com.simiacryptus.mindseye.lang.DeltaSet in project MindsEye by SimiaCryptus.

the class BatchDerivativeTester method testUnFrozen.

/**
 * Test un frozen.
 *
 * @param component      the component
 * @param inputPrototype the input prototype
 */
public void testUnFrozen(@Nonnull final Layer component, final Tensor[] inputPrototype) {
    @Nonnull final AtomicBoolean reachedInputFeedback = new AtomicBoolean(false);
    @Nonnull final Layer frozen = component.copy().setFrozen(false);
    @Nullable final Result eval = frozen.eval(new Result(TensorArray.create(inputPrototype), (@Nonnull final DeltaSet<Layer> buffer, @Nonnull final TensorList data) -> {
        reachedInputFeedback.set(true);
    }) {

        @Override
        public boolean isAlive() {
            return true;
        }
    });
    @Nonnull final DeltaSet<Layer> buffer = new DeltaSet<Layer>();
    TensorList data = eval.getData();
    eval.accumulate(buffer, data);
    @Nullable final List<double[]> stateList = frozen.state();
    final List<Delta<Layer>> deltas = stateList.stream().map(doubles -> {
        return buffer.stream().filter(x -> x.target == doubles).findFirst().orElse(null);
    }).filter(x -> x != null).collect(Collectors.toList());
    if (deltas.isEmpty() && !stateList.isEmpty()) {
        throw new AssertionError("Nonfrozen component not listed in delta. Deltas: " + deltas);
    }
    if (!reachedInputFeedback.get()) {
        throw new RuntimeException("Nonfrozen component did not pass input backwards");
    }
}

Example 39 with DeltaSet

use of com.simiacryptus.mindseye.lang.DeltaSet in project MindsEye by SimiaCryptus.

the class BatchDerivativeTester method getLearningGradient.

@Nonnull
private Tensor getLearningGradient(@Nonnull final Layer component, final int layerNum, @Nonnull final Tensor outputPrototype, final Tensor... inputPrototype) {
    component.setFrozen(false);
    final double[] stateArray = component.state().get(layerNum);
    final int stateLen = stateArray.length;
    @Nonnull final Tensor gradient = new Tensor(stateLen, outputPrototype.length());
    for (int j = 0; j < outputPrototype.length(); j++) {
        final int j_ = j;
        @Nonnull final DeltaSet<Layer> buffer = new DeltaSet<Layer>();
        @Nonnull final Tensor data = new Tensor(outputPrototype.getDimensions()).set((k) -> k == j_ ? 1 : 0);
        @Nullable final Result eval = component.eval(ConstantResult.singleResultArray(new Tensor[][] { inputPrototype }));
        eval.getData().get(0);
        @Nonnull TensorArray tensorArray = TensorArray.wrap(data);
        eval.accumulate(buffer, tensorArray);
        final DoubleBuffer<Layer> deltaFlushBuffer = buffer.getMap().values().stream().filter(x -> x.target == stateArray).findFirst().orElse(null);
        if (null != deltaFlushBuffer) {
            for (int i = 0; i < stateLen; i++) {
                gradient.set(new int[] { i, j_ }, deltaFlushBuffer.getDelta()[i]);
            }
        }
    }
    return gradient;
}

Example 40 with DeltaSet

use of com.simiacryptus.mindseye.lang.DeltaSet in project MindsEye by SimiaCryptus.

the class SumMetaLayer method eval.

@Nullable
@Override
public Result eval(@Nonnull final Result... inObj) {
    final Result input = inObj[0];
    Arrays.stream(inObj).forEach(nnResult -> nnResult.addRef());
    final int itemCnt = input.getData().length();
    if (null == lastResult || minBatches < itemCnt) {
        @Nonnull final ToDoubleFunction<Coordinate> f = (c) -> IntStream.range(0, itemCnt).mapToDouble(dataIndex -> input.getData().get(dataIndex).get(c)).sum();
        lastResult = input.getData().get(0).mapCoords(f);
    }
    return new Result(TensorArray.wrap(lastResult), (@Nonnull final DeltaSet<Layer> buffer, @Nonnull final TensorList data) -> {
        if (input.isAlive()) {
            @Nullable final Tensor delta = data.get(0);
            @Nonnull final Tensor[] feedback = new Tensor[itemCnt];
            Arrays.parallelSetAll(feedback, i -> new Tensor(delta.getDimensions()));
            @Nonnull final ToDoubleFunction<Coordinate> f = (inputCoord) -> {
                for (int inputItem = 0; inputItem < itemCnt; inputItem++) {
                    feedback[inputItem].add(inputCoord, delta.get(inputCoord));
                }
                return 0;
            };
            delta.mapCoords(f);
            @Nonnull TensorArray tensorArray = TensorArray.wrap(feedback);
            input.accumulate(buffer, tensorArray);
        }
    }) {

        @Override
        protected void _free() {
            Arrays.stream(inObj).forEach(nnResult -> nnResult.freeRef());
        }

        @Override
        public boolean isAlive() {
            return input.isAlive();
        }
    };
}