def run_yaml(path):
import sys
sys.path.append('..')
from pylearn2.config import yaml_parse
from utils.common import notify
with open(path) as yaml_file:
yaml = yaml_file.read()
network = yaml_parse.load(yaml)
network.main_loop()
notify()
# create dictionary with hyper parameters
hyper_params = {'model': yp.parse_to_yaml(mod), 'path': yp.parse_to_yaml(path),
'weight_decay_coeffs': weight_decay_coeffs, 'pkl_filename': pkl_filename}
# fill the yaml skelton with hyperparameters
yaml_string = default_string % hyper_params
# saving the yaml for later analysis
yaml_path = join(config.path_for_storing, timestamp+'generated_'+str(i)+'.yaml')
with open(yaml_path, 'w') as g:
g.write(yaml_string)
try:
# create the model based on a yaml
network = yaml_parse.load(yaml_string)
print t.bold_magenta('NETWORK'), type(network)
# train the model
network.main_loop()
except BaseException as e:
# if exception was thrown save yaml of a model that generated that exception
with open('0000'+str(i)+'.yaml', 'w') as YAML_FILE:
YAML_FILE.write(yaml_string)
# write down errors description to a file
with open('0000'+str(i)+'_error', 'w') as ERROR_FILE:
ERROR_FILE.write(traceback.format_exc())
# play a melody so everyone knows the testing has finished
notify()
