Using Pilco algorithm to find a controller for few robotic problems
License: GNU General Public License v2.0
Using Pilco algorithm to find a controller for few robotic problems
This is a modified version that uses numerical integration.
https://spiral.imperial.ac.uk/bitstream/10044/1/11578/4/rss2011_revision.pdf
I am interested in learning about the PILCO algorithm and found your code to be the a good example implementation. However I am having trouble getting the cost optimization step to converge. As the code is I get the error "optimizer failed: Desired error not necessarily achieved due to precision loss." The only change I made was setting useGPU to False (I don't have a GPU). Have you gotten this code to converge? Just wondering if you have any tips for how to get it running. Thanks!
I am trying to run this example program on Google Colab, and I get the following error. I tried to find the solution online but no success sofar. Any idea how to fix this?
LogicError Traceback (most recent call last)
[<ipython-input-12-86317df6fc11>](https://localhost:8080/#) in <module>
107 start_time = datetime.datetime.now()
108 print( 'start: ', start_time )
--> 109 kernel = GPy.kern.RBF( input_dim=5, useGPU=True )
110
111 # initialise pendubot
3 frames
[/usr/local/lib/python3.8/dist-packages/paramz/parameterized.py](https://localhost:8080/#) in __call__(self, *args, **kw)
51 #import ipdb;ipdb.set_trace()
52 initialize = kw.pop('initialize', True)
---> 53 self = super(ParametersChangedMeta, self).__call__(*args, **kw)
54 #logger.debug("finished init")
55 self._in_init_ = False
[/usr/local/lib/python3.8/dist-packages/GPy/kern/src/rbf.py](https://localhost:8080/#) in __init__(self, input_dim, variance, lengthscale, ARD, active_dims, name, useGPU, inv_l)
24 super(RBF, self).__init__(input_dim, variance, lengthscale, ARD, active_dims, name, useGPU=useGPU)
25 if self.useGPU:
---> 26 self.psicomp = PSICOMP_RBF_GPU()
27 else:
28 self.psicomp = PSICOMP_RBF()
[/usr/local/lib/python3.8/dist-packages/GPy/kern/src/psi_comp/rbf_psi_gpucomp.py](https://localhost:8080/#) in __init__(self, threadnum, blocknum, GPU_direct)
245 self.threadnum = threadnum
246 self.blocknum = blocknum
--> 247 module = SourceModule("#define THREADNUM "+str(self.threadnum)+"\n"+gpu_code)
248 self.g_psi1computations = module.get_function('psi1computations')
249 self.g_psi1computations.prepare('PPdPPPPiii')
[/usr/local/lib/python3.8/dist-packages/pycuda/compiler.py](https://localhost:8080/#) in __init__(self, source, nvcc, options, keep, no_extern_c, arch, code, cache_dir, include_dirs)
367 from pycuda.driver import module_from_buffer
368
--> 369 self.module = module_from_buffer(cubin)
370
371 self._bind_module()
LogicError:
Optimizer failed: Desired error not necessarily achieved due to precision loss.
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft
Open source projects and samples from Microsoft.
Google
Google ❤️ Open Source for everyone.
Alibaba
Alibaba Open Source for everyone
D3
Data-Driven Documents codes.
Tencent
China tencent open source team.