Click here to Skip to main content
14,266,990 members

Standardscaler is not working for ANN but minmax does

jameskm69 asked:

Open original thread
I have a dataset with multiple features and a target. I am using ANN to predict. When I scale the features using MInMax, everything works fine and during the compile and fix I get good loss and accuracy.
HOWEVER, when I use StandardScaler(). The prediction is off and then I noticed during the Compile/fit I get acc:0.00000 through out the whole epoch.
Just wondering why. I thought I can use any (specially StandardScaler).
Any suggestion?
Thank you

What I have tried:

Here is the code
dataset = pd.read_csv('somedata.csv')

X=dataset.iloc[:,0:13]
y=dataset.iloc[:,13].values
#StandardScaler
from sklearn.preprocessing import StandardScaler
sc2= StandardScaler()
X= sc2.fit_transform(X)
y= y.reshape(-1,1)
y=sc2.fit_transform(y)

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)

from keras import Sequential
from keras.layers import Dense

regressor = Sequential()
regressor.add(Dense(units=13, input_dim=13))
regressor.add(Dense(units=1))
regressor.compile(optimizer='adam', loss='mean_squared_error',  metrics=['accuracy'])
regressor.fit(X_train,y_train, epochs=100, batch_size=32, verbose=1)
Tags: Python, MACHINE_LEARNING

Preview



When answering a question please:
  1. Read the question carefully.
  2. Understand that English isn't everyone's first language so be lenient of bad spelling and grammar.
  3. If a question is poorly phrased then either ask for clarification, ignore it, or edit the question and fix the problem. Insults are not welcome.
  4. Don't tell someone to read the manual. Chances are they have and don't get it. Provide an answer or move on to the next question.
Let's work to help developers, not make them feel stupid.
Please note that all posts will be submitted under the The Code Project Open License (CPOL).




CodeProject, 503-250 Ferrand Drive Toronto Ontario, M3C 3G8 Canada +1 416-849-8900 x 100