Add curriculum code to notebook and simplify

python/PPO.ipynb

     "summary_freq = 10000 # Frequency at which to save training statistics.\n",
     "save_freq = 50000 # Frequency at which to save model.\n",
     "env_name = \"environment\" # Name of the training environment file.\n",
+    "curriculum_file = None\n",
     "\n",
     "### Algorithm-specific parameters for tuning\n",
     "gamma = 0.99 # Reward discount rate.\n",
    "metadata": {},
    "outputs": [],
    "source": [
-    "env = UnityEnvironment(file_name=env_name)\n",
+    "env = UnityEnvironment(file_name=env_name, curriculum=curriculum_file)\n",
     "print(str(env))\n",
     "brain_name = env.brain_names[0]"
    ]
    "source": [
     "tf.reset_default_graph()\n",
     "\n",
+    "if curriculum_file == \"None\":\n",
+    "    curriculum_file = None\n",
+    "\n",
+    "\n",
+    "def get_progress():\n",
+    "    if curriculum_file is not None:\n",
+    "        if env._curriculum.measure_type == \"progress\":\n",
+    "            return steps / max_steps\n",
+    "        elif env._curriculum.measure_type == \"reward\":\n",
+    "            return last_reward\n",
+    "        else:\n",
+    "            return None\n",
+    "    else:\n",
+    "        return None\n",
+    "\n",
     "# Create the Tensorflow model graph\n",
     "ppo_model = create_agent_model(env, lr=learning_rate,\n",
     "                               h_size=hidden_units, epsilon=epsilon,\n",
     "        saver.restore(sess, ckpt.model_checkpoint_path)\n",
     "    else:\n",
     "        sess.run(init)\n",
-    "    steps = sess.run(ppo_model.global_step)\n",
+    "    steps, last_reward = sess.run([ppo_model.global_step, ppo_model.last_reward])    \n",
-    "    info = env.reset(train_mode=train_model)[brain_name]\n",
+    "    info = env.reset(train_mode=train_model, progress=get_progress())[brain_name]\n",
-    "            info = env.reset(train_mode=train_model)[brain_name]\n",
+    "            info = env.reset(train_mode=train_model, progress=get_progress())[brain_name]\n",
     "        # Decide and take an action\n",
     "        new_info = trainer.take_action(info, env, brain_name)\n",
     "        info = new_info\n",
     "            save_model(sess, model_path=model_path, steps=steps, saver=saver)\n",
     "        steps += 1\n",
     "        sess.run(ppo_model.increment_step)\n",
+    "        if len(trainer.stats['cumulative_reward']) > 0:\n",
+    "            mean_reward = np.mean(trainer.stats['cumulative_reward'])\n",
+    "            sess.run(ppo_model.update_reward, feed_dict={ppo_model.new_reward: mean_reward})\n",
+    "            last_reward = sess.run(ppo_model.last_reward)\n",
     "    # Final save Tensorflow model\n",
     "    if steps != 0 and train_model:\n",
     "        save_model(sess, model_path=model_path, steps=steps, saver=saver)\n",

python/ppo.py

 
 Options:
   --help                     Show this message.
-  --curriculum               Whether to use curriculum for training (requires curriculum json) [default: False]
-  --curriculum-path=<path>   Path to curriculum json file for environment [default: curriculum.json]
+  --curriculum=<file>        Curriculum json file for environment [default: None]
   --max-steps=<n>            Maximum number of steps to run environment [default: 1e6].
   --run-path=<path>          The sub-directory name for model and summary statistics [default: ppo].
   --load                     Whether to load the model or randomly initialize [default: False].
 env_name = options['<env>']
 keep_checkpoints = int(options['--keep-checkpoints'])
 worker_id = int(options['--worker-id'])
-use_curriculum = options['--curriculum']
-if use_curriculum:
-    curriculum_path = str(options['--curriculum-path'])
-else:
-    curriculum_path = None
+curriculum_file = str(options['--curriculum'])
+if curriculum_file == "None":
+    curriculum_file = None
 
 # Algorithm-specific parameters for tuning
 gamma = float(options['--gamma'])
 hidden_units = int(options['--hidden-units'])
 batch_size = int(options['--batch-size'])
 
-env = UnityEnvironment(file_name=env_name, worker_id=worker_id, curriculum=curriculum_path)
+env = UnityEnvironment(file_name=env_name, worker_id=worker_id, curriculum=curriculum_file)
 print(str(env))
 brain_name = env.brain_names[0]
 
 
 
 def get_progress():
-    if use_curriculum:
+    if curriculum_file is not None:
         if env._curriculum.measure_type == "progress":
             return steps / max_steps
         elif env._curriculum.measure_type == "reward":
     else:
         return None
+
 
 with tf.Session() as sess:
     # Instantiate model parameters

python/unityagents/environment.py

         :return: A Data structure corresponding to the initial reset state of the environment.
         """
         old_lesson = self._curriculum.get_lesson_number()
-        config = self._curriculum.get_lesson(progress) if config is None else config 
+        config = self._curriculum.get_lesson(progress) if config is None else config
         if old_lesson != self._curriculum.get_lesson_number():
             logger.info("\nLesson changed. Now in Lesson {0} : \t{1}"
                 .format(self._curriculum.get_lesson_number(),