Neural Networks

Our workflow should be like that of Keras and not like the present one -

Here is the Sequential model:

from keras.models import Sequential

model = Sequential()

Stacking layers is as easy as .add():

from keras.layers import Dense

model.add(Dense(units=64, activation='relu', input_dim=100))
model.add(Dense(units=10, activation='softmax'))

Once your model looks good, configure its learning process with .compile():

model.compile(loss='categorical_crossentropy',
              optimizer='sgd',
              metrics=['accuracy'])

(taken from its readme)

References

Larger batch sizes are good but taking absurdly large batch size will result in fewer weight updates. So don't be absurd.

TODO: