In TFJS, model.predict() is undefined - tensorflow

Question

I used the Keras model I made by converting it to tensorflow.js. I called the model in js. I tried to predict the new value to the model. However, model.predict() is undefined.
The value of model.predict().dataSync()[0] is also undefined.
If you look at the model.summary() in js, you can see that it is a model that I made. I've put in tensor and [[]] in predict() arg'. It was all to no avail.
Attached is my model and the corresponding partial code below. I wish you could help me. If you need more code, please leave a comment. Thank you for your help.
model.json
{"format": "layers-model", "generatedBy": "keras v2.4.0", "convertedBy": "TensorFlow.js Converter v3.8.0", "modelTopology": {"keras_version": "2.4.0", "backend": "tensorflow", "model_config": {"class_name": "Sequential", "config": {"name": "sequential_2", "layers": [{"class_name": "InputLayer", "config": {"batch_input_shape": [null, 34], "dtype": "float32", "sparse": false, "ragged": false, "name": "dense_5_input"}}, {"class_name": "Dense", "config": {"name": "dense_5", "trainable": true, "dtype": "float32", "units": 256, "activation": "relu", "use_bias": true, "kernel_initializer": {"class_name": "GlorotUniform", "config": {"seed": null}}, "bias_initializer": {"class_name": "Zeros", "config": {}}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}}, {"class_name": "Dense", "config": {"name": "dense_6", "trainable": true, "dtype": "float32", "units": 128, "activation": "relu", "use_bias": true, "kernel_initializer": {"class_name": "GlorotUniform", "config": {"seed": null}}, "bias_initializer": {"class_name": "Zeros", "config": {}}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}}, {"class_name": "Dense", "config": {"name": "dense_7", "trainable": true, "dtype": "float32", "units": 1000, "activation": "softmax", "use_bias": true, "kernel_initializer": {"class_name": "GlorotUniform", "config": {"seed": null}}, "bias_initializer": {"class_name": "Zeros", "config": {}}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}}]}}, "training_config": {"loss": "sparse_categorical_crossentropy", "metrics": [[{"class_name": "MeanMetricWrapper", "config": {"name": "accuracy", "dtype": "float32", "fn": "sparse_categorical_accuracy"}}]], "weighted_metrics": null, "loss_weights": null, "optimizer_config": {"class_name": "Adam", "config": {"name": "Adam", "learning_rate": 0.0010000000474974513, "decay": 0.0, "beta_1": 0.8999999761581421, "beta_2": 0.9990000128746033, "epsilon": 1e-07, "amsgrad": false}}}}, "weightsManifest": [{"paths": ["group1-shard1of1.bin"], "weights": [{"name": "dense_5/kernel", "shape": [34, 256], "dtype": "float32"}, {"name": "dense_5/bias", "shape": [256], "dtype": "float32"}, {"name": "dense_6/kernel", "shape": [256, 128], "dtype": "float32"}, {"name": "dense_6/bias", "shape": [128], "dtype": "float32"}, {"name": "dense_7/kernel", "shape": [128, 1000], "dtype": "float32"}, {"name": "dense_7/bias", "shape": [1000], "dtype": "float32"}]}]}
predictPose
const predictPose = (arr) => {
// Arr is array with 34 numbers as elements.
// ex) [234,123,324,.....,222]
const prediction, model;
model = loadModel();
const p = async () => {
prediction = model().predict([arr]).dataSync()[0];
//pedicttion = model().predict(tf.tensor([arr])).dataSync()[0];
};
console.log(prediction); // The result is undefined.
};

Answer 1

const prediction, model; will throw an error actually, you can't assign const to nothing, you can do that for let in this case as you intend to reassign the variable name later on but its unnecessary. Another issue is that you aren't even loading in any model. Based on your model.json you would need to use tf.loadLayersModel(). The code below should work.
const predictPose = async (arr) => {
const model = await tf.loadLayersModel("path to model.json");
const prediction = model.predict([arr]).dataSync()[0];
console.log(prediction);
};