Deep Learning with Keras

Saving and loading the weights and the architecture of a model

Model architectures can be easily saved and loaded as follows:

# save as JSON json_string = model.to_json()# save as YAML yaml_string = model.to_yaml() # model reconstruction from JSON: from keras.models import model_from_json model = model_from_json(json_string) # model reconstruction from YAML model = model_from_yaml(yaml_string)

Model parameters (weights) can be easily saved and loaded as follows:

from keras.models import load_model model.save('my_model.h5')# creates a HDF5 file 'my_model.h5' del model# deletes the existing model# returns a compiled model# identical to the previous one model = load_model('my_model.h5')