Updated ray dependency and specified package versions

README.md

@@ -7,7 +7,7 @@ This is the official research code for the paper Kirsch et al. 2019:
 
 Install the following dependencies (in a virtualenv preferably)
 ```bash
-pip3 install ray gym[all] tensorflow-gpu scipy numpy
+pip3 install ray[tune]==0.7.7 gym[all] mujoco_py>=2 tensorflow-gpu==1.13.2 scipy numpy
 ```
 
 This code base uses [ray](https://github.com/ray-project/ray), if you would like to use multiple machines,

ray_experiments.py

@@ -17,7 +17,7 @@
 
 np.warnings.filterwarnings('ignore')
 logger = logging.getLogger(__name__)
-
+tf.logging.set_verbosity(tf.logging.ERROR)
 
 class LLFSExperiment(ExtendedTrainable):
 
@@ -178,9 +178,9 @@ def _restore(self, checkpoint_data):
 # noinspection PyProtectedMember
 def count_required_gpus(config):
     if config['agent_count'] > 1:
-        return math.ceil(config['agent_count'] * ray_workers.AgentWorker._num_gpus + ray_workers.ObjectiveServer._num_gpus)
+        return math.ceil(config['agent_count'] * ray_workers.AgentWorker.__ray_metadata__.num_gpus + ray_workers.ObjectiveServer.__ray_metadata__.num_gpus)
     else:
-        return ray_workers.AgentWorker._num_gpus
+        return ray_workers.AgentWorker.__ray_metadata__.num_gpus
 
 
 def init_ray(redis_address=None):
@@ -191,7 +191,7 @@ def init_ray(redis_address=None):
         ray.init(object_store_memory=mem, redis_max_memory=mem, temp_dir='/tmp/metagenrl/ray')
 
 
-def run(config, run_name='metagenrl', timesteps=300 * 1000, samples=1):
+def run(config, run_name='metagenrl', timesteps=700 * 1000, samples=1):
     tune.register_trainable(run_name, LLFSExperiment)
     trial_gpus = count_required_gpus(config)
     print(f'Requiring {trial_gpus} extra gpus.')
@@ -226,14 +226,17 @@ def test(args):
     Performs meta-test training
     """
     assert isinstance(args.objective, str)
-    config = configs.test(args.objective)
+    config = configs.test(args.objective, chkp=args.chkp)
     config.update({
+        'name': args.name,
         'env_name': tune.grid_search([
             'Hopper-v2',
+            'HalfCheetah-v2',
+            'LunarLanderContinuous-v2',
         ]),
     })
 
-    run(config, run_name='test-public-CheetahLunar')
+    run(config, run_name=f'test-public-{args.name}-chkp{args.chkp}', samples=1)
 
 
 if __name__ == '__main__':
@@ -244,6 +247,8 @@ def test(args):
     parser.add_argument('command', choices=FUNCTION_MAP.keys())
     parser.add_argument('--redis', dest='redis_address', action='store', type=str)
     parser.add_argument('--objective', action='store', type=str)
+    parser.add_argument('--name', action='store', type=str)
+    parser.add_argument('--chkp', action='store', type=int, default=-1)
     parsed_args = parser.parse_args()
     init_ray(parsed_args.redis_address)
     func = FUNCTION_MAP[parsed_args.command]

ray_workers.py

@@ -511,4 +511,4 @@ def compute_objective_gradients(self, t, grad_oid):
             self.objective_vars_oid = None
 
         self.feed_dict[self.locals.plasma_grad_oid] = grad_oid
-        self.sess.run(self.locals.plasma_write_grads, self.feed_dict)
+        self.sess.run(self.locals.plasma_write_grads, self.feed_dict)

utils.py

@@ -89,7 +89,7 @@ def get_store_socket():
     :return: socket file path
     """
     try:
-        return ray.worker.global_worker.plasma_client.store_socket_name
+        return ray.worker.global_worker.node.plasma_store_socket_name
     except AttributeError:
         return ''
 
@@ -222,4 +222,4 @@ def get_vars(scope, trainable_only=True):
 
 def count_vars(scope):
     v = get_vars(scope)
-    return sum([np.prod(var.shape.as_list()) for var in v])
+    return sum([np.prod(var.shape.as_list()) for var in v])