""" Basic task of the Python-script: Creates user-defined runscripts for training, set ups a user-defined target directory and allows for full control on the setting of hyperparameters. """ __email__ = "b.gong@fz-juelich.de" __authors__ = "Bing Gong, Scarlet Stadtler,Michael Langguth" __date__ = "2020-10-27" # import modules import sys, os, glob import numpy as np import datetime as dt import json as js import metadata sys.path.append(path.abspath('../video_prediction/')) from models import get_model_class known_architectures = ["savp","convLSTM","vae","mcnet"] if not (model in known_architectures): # start script def main(): # get required information from the user by keyboard interaction # path to preprocessed data exp_dir = input("Enter the path to the preprocessed data (directory where tf-records files are located):\n") exp_dir = os.path.join(exp_dir,"train") # sanity check (does preprocessed data exist?) if not (os.path.isdir(exp_dir)): raise NotADirectoryError("Passed path to preprocessed data '"+exp_dir+"' does not exist!") file_list = glob.glob(os.path.join(exp_dir,"sequence*.tfrecords")) if len(file_list) == 0: raise FileNotFoundError("Passed path to preprocessed data '"+exp_dir+"' exists,"+\ "but no tfrecord-files can be found therein") # path to virtual environment to be used venv_name = input("Enter the name of the virtual environment which should be used:\n") # sanity check (does virtual environment exist?) if not (os.path.isfile("../",venv_name,"bin","activate")): raise FileNotFoundError("Could not find a virtual environment named "+venv_name) # model # experimental ID exp_id = input("Enter your desired experimental id:\n") # also get current timestamp and user-name timestamp = dt.datetime.now().strftime("%Y%m%dT%H%M%S") user_name = os.environ["USER"]