1
$\begingroup$

The Blue dots represent the required function and the Black ones represent the predicted function

The Blue dots represent the required function and the Black ones represent the predicted function

I am using keras with the following code:-

# Set callback functions to early stop training and save the best model so far
from keras.callbacks import EarlyStopping, ModelCheckpoint
callbacks = [EarlyStopping(monitor='val_loss',verbose=1, patience=5,min_delta=0.002,mode='auto',restore_best_weights=False),
             ModelCheckpoint(filepath='best_model.hdf5', monitor='val_loss', save_best_only=True)]

##The Architecture
np.random.seed(0)
model = Sequential()
model.add(Dense(500,kernel_initializer='normal',activation='tanh',input_dim=1))

model.add(Dense(300, kernel_initializer='normal',activation='relu'))

model.add(Dense(500, kernel_initializer='normal',activation='relu'))

model.add(Dense(1, kernel_initializer='normal',activation='linear'))

model.compile(optimizer='adam',loss='mean_absolute_error',  metrics=['mae','accuracy'])

his = model.fit(X_train,y_train,epochs=150,batch_size=10,validation_data=
(X_val,y_val), verbose=1,callbacks=callbacks)

I tried various architecture but still not getting the result please guide

$\endgroup$
1
$\begingroup$

I see you use are sequential model. Could be a task for an LSTM architecture. https://machinelearningmastery.com/time-series-prediction-lstm-recurrent-neural-networks-python-keras/

Maybe you are overengineering the thing. Judging from the dataplot, a simple OLS regression splines model may also fit well. https://www.analyticsvidhya.com/blog/2018/03/introduction-regression-splines-python-codes/

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.