tensorflowで作成したtfliteモデルをflutterで利用する

import tensorflow as tf

# load model
model = tf.keras.models.load_model('./models/model.hdf5')

# convert
converter = tf.lite.TFLiteConverter.from_keras_model(model)
tflite_model = converter.convert()

# save
open("./models/model.tflite", "wb").write(tflite_model)

...
flutter:
  assets:
    - assets/label.txt
    - assets/model.tflite
...

dog
cat

    await Tflite.loadModel(model: "assets/model.tflite",
    labels: "assets/label.txt");
    // TODO imageバイナリで渡す方法でやってみたい
    // https://pub.dev/packages/tflite
    var output =  await Tflite.runModelOnImage(
      path: imagePath,
       );

    results = output;

  Future imageConnect(ImageSource source) async {
    final PickedFile img = await picker.getImage(source: source);
    if (img != null) {
      imagePath = img.path;
      image = File(img.path);
      loading = true;

      setState(() {});
      classifyImage();
    }
  }