Why are my loss and accuracy number are too much in neural net?

Machine learning newbie here, i am currently learning on how to do neural net with tensorflow. I have this dataset below:

out of curiosity, i wonder how will neural net perform on this dataset instead of using Linear Regression,so i typed this code below:

x = file.drop(columns= "Height").values
y = file.drop(columns= "Weight").values
y = np.reshape(y, (-1, 1))
x = np.reshape(x, (-1, 1))

from sklearn.preprocessing import StandardScaler
ss = StandardScaler()
x = ss.fit_transform(x)

from sklearn.model_selection import train_test_split
x_train, x_test, y_train, y_test = train_test_split(x, y, test_size = 0.2)


import tensorflow as tf
model = tf.keras.Sequential([
    tf.keras.layers.Dense(16, activation = "relu"),
    tf.keras.layers.Dense(16, activation = "relu"),
    tf.keras.layers.Dense(1, activation = "sigmoid")
])

model.compile(optimizer= tf.keras.optimizers.Adam(),
loss = tf.keras.losses.BinaryCrossentropy(), metrics= "accuracy")

model.fit(x_train, y_train, epochs = 10, batch_size = 16)#I actually dont fully know what batch size is, im adding that just for kicks

i scaled the data with the basic knowledge of by scaling it, the accuracy of neural net will increase slightly. However, when i run the code above, i get this "ugly looking" Accuracy and losses:

What can i do to make the accuracy and loss better?

For applying regression on the data,any linear activation should be used in the output layer. Remove the sigmoid activation as it reduces the continuous values in the height. To further improve accuracy units inside "tf.keras.layers.Dense()" could also be increased.

I am not entirely sure what is making that high of a loss but i would suggest you increase your dataset size that may help. I mean a lot bigger. Why dont you try using some survey reports or even kaggle(altough the risk is upto you because the data from kaggle isnt authenticated)