This section will help you learn to create a simple regression prediction model with the help of MPL. To represent Regression MPL, you need to follow the steps described below:
Step – 1: Import required modules
The necessary modules required to create a regression prediction model are as follows:
import keras
from keras.datasets import boston_housing
from sklearn.preprocessing import scale
from keras.layers import Dense
from keras.callbacks import EarlyStopping
from sklearn import preprocessing
from keras.optimizers import RMSprop
Step – 2: Load input data
Here we will import our dataset:
(model_train, model2_train), (model_test, model2_test) = data_set.load_data()
Step – 3: Process the loaded data
Now, to process the loaded data, you can refer to the following code snippet:
scaled_model_train = preprocessing.scale(model_train)
scaler_data = preprocessing.StandardScaler().fit(model_train)
scaled_model_test = scaler.transform(model_test)
Step – 4: Create and compile the model
First, we will create the actual model with the following code:
model = Sequential()
model.add(Dense(64, kernel_initializer = 'normal', activation = 'relu',
input_shape = (8,)))
model.add(Dense(64, activation = 'relu')) model.add(Dense(1))
Now, we will compile the model using the code below:
model.compile(
loss = 'mse',
optimizer = RMSprop(),
metrics = ['mean_absolute_error']
)
Step – 5: Now train the model
history = model.fit(
scaled_model_train, model_train,
batch_size = 128,
epochs = 350,
verbose = 1,
validation_split = 0.2,
callbacks = [EarlyStopping(monitor = 'val_loss', patience = 20)]
)
history = model.fit(
scaled_model_train, model_train,
batch_size = 128,
epochs = 350,
verbose = 1,
validation_split = 0.2,
callbacks = [EarlyStopping(monitor = 'val_loss', patience = 20)]
)
Step – 6: Model Evaluation
score = model.evaluate(scaled_model_test, model2_test, verbose = 0)
print('The total test loss is:', score[0])
print('The total test accuracy is:', score[1])
Step – 7: Predict from the model
prediction = model.predict(scaled_model_test)
print(prediction.flatten())
print(model2_test)
This way you can implement CNN with the help of these steps and create regression prediction.