Giter VIP home page Giter VIP logo

nervenet's People

Contributors

wilsonwangthu avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar

nervenet's Issues

NotImplementedError for the gym environments

Hi, I have installed gym_extensions and all the tests in the gym_extensions package work. However while I attempt to run the same environment in the original code with

python main.py --task CentipedeEight-v1 --use_gnn_as_policy 1 --network_shape 64,64 --lr 0.0003 --num_threads 4 --lr_schedule adaptive --max_timesteps 1000000 --use_gnn_as_value 0 --gnn_embedding_option noninput_shared --root_connection_option nN,Rn,uE --gnn_node_option nG,nB

I get the following error:
[0130 16:05:39 @main.py:21] New environments available : ['WalkersHopperzero-v1', 'WalkersHalfhumanoidzero-v1', 'WalkersHalfcheetahzero-v1', 'WalkersFullcheetahzero-v1', 'WalkersOstrichzero-v1', 'WalkersHopperone-v1', 'WalkersHalfhumanoidone-v1', 'WalkersHalfcheetahone-v1', 'WalkersFullcheetahone-v1', 'WalkersOstrichone-v1', 'WalkersHoppertwo-v1', 'WalkersHalfhumanoidtwo-v1', 'WalkersHalfcheetahtwo-v1', 'WalkersFullcheetahtwo-v1', 'WalkersOstrichtwo-v1', 'WalkersHopperthree-v1', 'WalkersHalfhumanoidthree-v1', 'WalkersHalfcheetahthree-v1', 'WalkersFullcheetahthree-v1', 'WalkersOstrichthree-v1', 'WalkersHopperfour-v1', 'WalkersHalfhumanoidfour-v1', 'WalkersHalfcheetahfour-v1', 'WalkersFullcheetahfour-v1', 'WalkersOstrichfour-v1', 'WalkersHopperfive-v1', 'WalkersHalfhumanoidfive-v1', 'WalkersHalfcheetahfive-v1', 'WalkersFullcheetahfive-v1', 'WalkersOstrichfive-v1', 'CentipedeThree-v1', 'CentipedeFour-v1', 'CentipedeFive-v1', 'CentipedeSix-v1', 'CentipedeSeven-v1', 'CentipedeEight-v1', 'CentipedeTen-v1', 'CentipedeTwelve-v1', 'CentipedeFourteen-v1', 'CentipedeTwenty-v1', 'CentipedeThirty-v1', 'CentipedeForty-v1', 'CentipedeFifty-v1', 'CpCentipedeThree-v1', 'CpCentipedeFour-v1', 'CpCentipedeFive-v1', 'CpCentipedeSix-v1', 'CpCentipedeSeven-v1', 'CpCentipedeEight-v1', 'CpCentipedeTen-v1', 'CpCentipedeTwelve-v1', 'CpCentipedeFourteen-v1', 'ReacherZero-v1', 'ReacherOne-v1', 'ReacherTwo-v1', 'ReacherThree-v1', 'ReacherFour-v1', 'ReacherFive-v1', 'ReacherSix-v1', 'ReacherSeven-v1', 'SnakeThree-v1', 'SnakeFour-v1', 'SnakeFive-v1', 'SnakeSix-v1', 'SnakeSeven-v1', 'SnakeEight-v1', 'SnakeNine-v1', 'SnakeTen-v1', 'SnakeTwenty-v1', 'SnakeForty-v1'] Traceback (most recent call last): File "/home/josyula/Programs/NerveNet/tool/main.py", line 26, in <module> learner_env = gym.make(args.task) File "/home/josyula/miniconda3/lib/python3.9/site-packages/gym/envs/registration.py", line 235, in make return registry.make(id, **kwargs) File "/home/josyula/miniconda3/lib/python3.9/site-packages/gym/envs/registration.py", line 129, in make env = spec.make(**kwargs) File "/home/josyula/miniconda3/lib/python3.9/site-packages/gym/envs/registration.py", line 90, in make env = cls(**_kwargs) File "/home/josyula/Programs/NerveNet/tool/../environments/multitask_env/walkers.py", line 374, in __init__ super(WalkersHalfcheetahfiveEnv, self).__init__(num=5) File "/home/josyula/Programs/NerveNet/tool/../environments/multitask_env/walkers.py", line 190, in __init__ mujoco_env.MujocoEnv.__init__(self, xml_path, 4) File "/home/josyula/miniconda3/lib/python3.9/site-packages/gym/envs/mujoco/mujoco_env.py", line 71, in __init__ observation, _reward, done, _info = self.step(action) File "/home/josyula/miniconda3/lib/python3.9/site-packages/gym/core.py", line 58, in step raise NotImplementedError NotImplementedError

same environment works in gym_extensions:
python nervenet_env_single_test.py
Can you please help regarding this issue? Thank you.

Python version

Really like your work and that you shared the code.
But maybe you should add which Python version you used.
To me it seems like you're using 2.x, rather than 3.x

matrix's rank wrong

Environment : Windows, tf1.2.1-cpu, python 3.6.2, mujoco1.3

ValueError: Shape must be rank 2 but is rank 4 for 'MLP_embedding_node_type_root/layer_0/MatMul' (op: 'MatMul') with input shapes: [?,5], [0,1075052544,0,1077936128].

When I use NerveNet to test, I will get this problem. But MLPAA is right. So How can I solve this problem?

TypeError: unsupported operand type(s) for -: 'NoneType' and 'NoneType'

I installed tf=1.13.1,gym=0.7.4.
Running the following code is ok.
"python main.py --task CentipedeEight-v1 --use_gnn_as_policy 0 --num_threads 4 --ckpt_name ../checkpoint/centipede/fc/6 --mlp_raw_transfer 1 --transfer_env CentipedeSix2CentipedeEight --test 100"
But running the train will report the following bug.
"python main.py --task ReacherOne-v1 --use_gnn_as_policy 1 --network_shape 64,64 --lr 0.0003 --num_threads 4 --lr_schedule adaptive --max_timesteps 1000000 --use_gnn_as_value 0 --gnn_embedding_option noninput_shared --root_connection_option nN,Rn,uE --gnn_node_option nG,nB
"

Bug Details:
File "/data1/zhihaocheng/NerveNet/tool/../agent/optimization_agent.py", line 287, in build_ppo_update_op
self.grads = tf.gradients(self.loss, self.tvars)
File "/home/zhihangcheng/anaconda3/envs/Py27/lib/python2.7/site-packages/tensorflow/python/ops/gradients_impl.py", line 664, in gradients
unconnected_gradients)
File "/home/zhihangcheng/anaconda3/envs/Py27/lib/python2.7/site-packages/tensorflow/python/ops/gradients_impl.py", line 965, in _GradientsHelper
lambda: grad_fn(op, *out_grads))
File "/home/zhihangcheng/anaconda3/envs/Py27/lib/python2.7/site-packages/tensorflow/python/ops/gradients_impl.py", line 420, in _MaybeCompile
return grad_fn() # Exit early
File "/home/zhihangcheng/anaconda3/envs/Py27/lib/python2.7/site-packages/tensorflow/python/ops/gradients_impl.py", line 965, in
lambda: grad_fn(op, *out_grads))
File "/home/zhihangcheng/anaconda3/envs/Py27/lib/python2.7/site-packages/tensorflow/python/ops/math_grad.py", line 328, in _UnsortedSegmentSumGrad
return _GatherDropNegatives(grad, op.inputs[1])[0], None, None
File "/home/zhihangcheng/anaconda3/envs/Py27/lib/python2.7/site-packages/tensorflow/python/ops/math_grad.py", line 296, in _GatherDropNegatives
for _ in range(gathered.shape.ndims - is_positive.shape.ndims):
TypeError: unsupported operand type(s) for -: 'NoneType' and 'NoneType'

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.