# setting parameters of MLP model mlp = models.MLP() mlp.layers = layers mlp.input_space = some_space # define weight decay parameters. They depend on the number of layers (there is one parameter fo each layer) weight_decay_coeffs = yp.parse_weight_decay(mlp) # define data paths path = config.data_path # generate a filename to store the best model pkl_filename = join(config.path_for_storing, current_time+"example_best.pkl") # create dictionary with hyper parameters hyper_params = {'model': yp.parse_to_yaml(mlp), 'path': yp.parse_to_yaml(path), 'weight_decay_coeffs': weight_decay_coeffs, 'pkl_filename': pkl_filename} # obtaining the yaml skelton with open(config.yaml_skelton_path) as f: default_string = f.read() # filling the yaml skelton with hyperparameters yaml_string = default_string % hyper_params generated_yaml_path = join(config.path_for_storing, current_time+'generated_yaml.yaml') with open(generated_yaml_path, 'w') as g: g.write(yaml_string) # creating the model based on a yaml model = yaml_parse.load(yaml_string)
def t3(): tup = (models.MLP(), models.Linear(), models.TanhConvNonlinearity(), models.SoftmaxPool()) return parse_to_yaml(tup)
def t1(): l = [models.MLP(), models.Linear(), models.TanhConvNonlinearity(), models.SoftmaxPool()] return parse_to_yaml(l)
def t2(): d = {'a': models.MLP(), 'b': models.Linear(), 'c': models.TanhConvNonlinearity(), 'd': models.SoftmaxPool()} return parse_to_yaml(d)