Hi,
Thanks for the API, i am enjoying it. So i have a bug about the practical example entitle: Building a simple regression model with TensorFlow.js. I applied it to new example and i use the same data preprocessing file. The problem is: when i used the full dataset with
let scaler = new dfd.MinMaxScaler()
scaler.fit(Xtrain)
Xtrain = scaler.transform(Xtrain)
i get the good result as in the example and when i want to use it during the prediction to predict one line only, from the raw data not like in the book by passing through the preprocessing method i am always get this result

[ [
NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN,
NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN,
NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN,
NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN,
NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN,
NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN,
NaN, NaN
] ]

and i canot send this to my model to predict, so i try it by the way and get the following result: The probability of winning is: [ [ NaN ].

When i comment the line "Xtrain = scaler.transform(Xtrain)" i get different results as:

[ [
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 49, 0, 41, 10, 0, 0, 0, 1, 0, 0,
0, 1, 1, 0, 0, 0
] ]
The probability of winning is:
[ [ -1.6790392398834229 ] ]

I just want to use the raw data.csv with one example to make prediction.
can you help me please to fix that bug, Thanks in advance

Sincerely