Learn how to train a model with a train dataset in Keras
Source file needed
Dataset
model.py, trainmodel.py
Preparation
Dataset
Download directly from Keras
CNN model
See other tutorials of how to build a model
Template code for my
#Importing library
import keras
from keras.models import Sequential
from keras.layers import Dense, Activation, Dropout, Flatten, Conv2D, MaxPooling2D
from keras.layers.normalization import BatchNormalization
# Define your model
def MYMODEL(weights_path=None):
model = Sequential()
# architecture goes here
model.add(Conv2D(input_shape=(224,224,3),filters=64,kernel_size=(3,3),padding="same", activation="relu"))
if weights_path:
model.load_weights(weights_path)
return model
Train model
Read dataset
Preprocessing for correct input size
Train with an optimizer
from keras.optimizers import Adam
opt = Adam(lr=0.001)
model.compile(optimizer=opt, loss=keras.losses.categorical_crossentropy, metrics=['accuracy'])