我正在尝试在 Java（JSE 而不是 Android）中运行 Tensorflow 进行图像识别（分类）。

我正在使用这里和这里的代码。

它适用于 Inceptionv3 模型，以及从 Inceptionv3 重新训练的模型。

但是对于 MobileNet 模型，它不起作用，（比如下面这篇文章）。

该代码有效，但给出了错误的结果（错误的分类标签）。从 Java 中使用 MobileNet 需要哪些代码/设置？

适用于 Inceptionv3 的代码是

try (Tensor image = Tensor.create(imageBytes)) {
    float[] labelProbabilities = executeInceptionGraph(graphDef, image);
    int bestLabelIdx = maxIndex(labelProbabilities);
    result.setText("");
    result.setText(String.format(
        "BEST MATCH: %s (%.2f%% likely)",
        labels.get(bestLabelIdx), labelProbabilities[bestLabelIdx] * 100f));
        System.out.println(
            String.format(
                "BEST MATCH: %s (%.2f%% likely)",
                labels.get(bestLabelIdx), labelProbabilities[bestLabelIdx] * 100f));
}

这适用于 Inceptionv3 模型，但不适用于 MobileNet，会给出错误“期望 args[0] 为浮点数，但提供了字符串”

对于 MobileNet，我们尝试了代码，

try (Graph g = new Graph()) {
    GraphBuilder b = new GraphBuilder(g);
    // Some constants specific to the pre-trained model at:
    // https://storage.googleapis.com/download.tensorflow.org/models/inception5h.zip
    //
    // - The model was trained with images scaled to 224x224 pixels.
    // - The colors, represented as R, G, B in 1-byte each were converted to
    //   float using (value - Mean)/Scale.
    final int H = 224;
    final int W = 224;
    final float mean = 128f;
    final float scale = 1f;
    // Since the graph is being constructed once per execution here, we can use a constant for the
    // input image. If the graph were to be re-used for multiple input images, a placeholder would
    // have been more appropriate.
    final Output<String> input = b.constant("input", imageBytes);
    final Output<Float> output = b.div(
        b.sub(
            b.resizeBilinear(
                b.expandDims(
                    b.cast(b.decodeJpeg(input, 3), Float.class),
                    b.constant("make_batch", 0)),
                    b.constant("size", new int[] {H, W})),
                    b.constant("mean", mean)),
                    b.constant("scale", scale));
                    try (Session s = new Session(g)) {
                        return s.runner().fetch(output.op().name()).run().get(0).expect(Float.class);
                    }
}

这有效，但给出了错误的标签。