I am training an LSTM model for binary classification and have two CSV files that has the same underlying data. When I train it with "file1" I get model accuracy close