def _log_policy(self):
save_path = os.path.join(self._log_dir, 'models')
filesystem.mkdir(save_path)
weights = self._policy.get_weights()
data = {'policy_weights': weights}
full_path = os.path.join(save_path, 'policy_{}.pkl'.format(self._total_timestep))
print('Saving policy to: {}'.format(full_path))
pickle.dump(data, open(full_path, 'wb'))
def _log_model(self):
save_path = os.path.join(self._log_dir, 'models')
filesystem.mkdir(save_path)
print('Saving model to: {}'.format(save_path))
self._model.save(save_path, self._total_timestep)
