Hi Diego,



I'm sorry, but I'm not sure what's wrong. I would try cleaning out all the 
relevant directories (including /tmp/cerebro2) and following the instructions 
in the READMEs, starting from cerebro2.server README.




Sorry I can't be of more help,

Chetan

On Wed, Sep 24, 2014 at 8:33 AM, Matthew Taylor <[email protected]> wrote:

> Discussion about installing Cerebro2 below. I'm forwarding to the list
> in case it could help others. Diego and Chetan will continue the
> discussion on the list from this point.
> ---------
> Matt Taylor
> OS Community Flag-Bearer
> Numenta
> ---------- Forwarded message ----------
> From: Diego Martinez <[email protected]>
> Date: Wed, Sep 24, 2014 at 2:55 AM
> Subject: Re: Installing Cerebro2 (debugging tool)
> To: Chetan Surpur <[email protected]>
> Cc: Matthew Taylor <[email protected]>
> Hello Chetan,
> Yes i ran the three files together but i have also ran them
> individually with the very same results. I just did it again with
> demo_tp.py just to be sure and i received the very same error i sent
> you above.
> @Matthew: no problem from my side.
> D.
> 2014-09-23 19:35 GMT+02:00 Chetan Surpur <[email protected]>:
>>
>> Hi Diego,
>>
>> From your previous email, it looks like you ran the following command:
>>
>> python demo_coordinate_encoder.py && python demo_sp.py && python demo_tp.py
>>
>> I suggest running just python demo_tp.py and trying again. Let me know if 
>> you tried that and it didn't work.
>>
>> - Chetan
>>
>>
>>
>> On Tue, Sep 23, 2014 at 3:05 AM, Diego Martinez <[email protected]> wrote:
>>>
>>> Hello Chetan,
>>>
>>> As i stated in my previous e-mail when running only demo_sp.py i get the 
>>> same error....
>>>
>>> D.
>>>
>>> 2014-09-22 19:22 GMT+02:00 Chetan Surpur <[email protected]>:
>>>>
>>>> Try running just demo_sp.py or demo_tp.py and then opening Cerebro 2 in 
>>>> your browser.
>>>>
>>>>
>>>>
>>>> On Mon, Sep 22, 2014 at 4:56 AM, Diego Martinez <[email protected]> wrote:
>>>>>
>>>>> Hello Chetan,
>>>>>
>>>>> Thanks for your quick answer. For the moment i only want to run one of 
>>>>> the demos. Once is clear for me how it works i will apply it to my models.
>>>>>
>>>>> I ran this on the terminal ~/nupic.cerebro2.server$ python 
>>>>> demo_coordinate_encoder.py && python demo_sp.py && python demo_tp.py. I 
>>>>> can see some information loading + iterations till 100 like this:
>>>>>
>>>>> **btw: not sure i need to run all the three files at the same time, i 
>>>>> also ran them individually but i had the same results**
>>>>>
>>>>> ===== [-149   24] / 3 =====
>>>>> [  0  57  58 131 200 208 245 284 373 379 424 448 449 515 528 558 592 664
>>>>>  763 853 908 963 978 987 998]
>>>>> .
>>>>> .
>>>>>
>>>>> Ran iteration: 0
>>>>> Ran iteration: 1
>>>>> Ran iteration: 2
>>>>> .
>>>>> .
>>>>>
>>>>> Then i run python server.py 9090 and the server is started successfully. 
>>>>> I open a new terminal and go to cerebro2-client, cd into static folder 
>>>>> and run command: python -m SimpleHTTPServer 8000 and everything worked 
>>>>> fine. However, when i go to http://localhost:8000 the page is completely 
>>>>> in blank and i can see an error in the server terminal where i started 
>>>>> port 9090. Here below i am sending you the error for your reference.
>>>>>
>>>>> 127.0.0.1:52077 - - [22/Sep/2014 13:37:57] "HTTP/1.1 GET 
>>>>> /_model/num_iterations" - 200 OK
>>>>> Traceback (most recent call last):
>>>>>   File "/usr/local/lib/python2.7/dist-packages/web/application.py", line 
>>>>> 239, in process
>>>>>     return self.handle()
>>>>>   File "/usr/local/lib/python2.7/dist-packages/web/application.py", line 
>>>>> 230, in handle
>>>>>     return self._delegate(fn, self.fvars, args)
>>>>>   File "/usr/local/lib/python2.7/dist-packages/web/application.py", line 
>>>>> 420, in _delegate
>>>>>     return handle_class(cls)
>>>>>   File "/usr/local/lib/python2.7/dist-packages/web/application.py", line 
>>>>> 396, in handle_class
>>>>>     return tocall(*args)
>>>>>   File "/home/diego/nupic.cerebro2.server/api/model_api.py", line 46, in 
>>>>> GET
>>>>>     return jsonResponse(readJSON(filepath))
>>>>>   File "/home/diego/nupic.cerebro2.server/api/model_api.py", line 68, in 
>>>>> readJSON
>>>>>     raise error
>>>>> IOError: [Errno 2] No such file or directory: 
>>>>> u'/tmp/cerebro2/model/dimensions/input.json'
>>>>>
>>>>> 127.0.0.1:52078 - - [22/Sep/2014 13:37:57] "HTTP/1.1 GET 
>>>>> /_model/dimensions/output" - 200 OK
>>>>> Traceback (most recent call last):
>>>>>   File "/usr/local/lib/python2.7/dist-packages/web/application.py", line 
>>>>> 239, in process
>>>>>     return self.handle()
>>>>>   File "/usr/local/lib/python2.7/dist-packages/web/application.py", line 
>>>>> 230, in handle
>>>>>     return self._delegate(fn, self.fvars, args)
>>>>>   File "/usr/local/lib/python2.7/dist-packages/web/application.py", line 
>>>>> 420, in _delegate
>>>>>     return handle_class(cls)
>>>>>   File "/usr/local/lib/python2.7/dist-packages/web/application.py", line 
>>>>> 396, in handle_class
>>>>>     return tocall(*args)
>>>>>   File "/home/diego/nupic.cerebro2.server/api/model_api.py", line 46, in 
>>>>> GET
>>>>>     return jsonResponse(readJSON(filepath))
>>>>>   File "/home/diego/nupic.cerebro2.server/api/model_api.py", line 68, in 
>>>>> readJSON
>>>>>     raise error
>>>>> IOError: [Errno 2] No such file or directory: 
>>>>> u'/tmp/cerebro2/model/encoders.json'
>>>>>
>>>>> 127.0.0.1:52077 - - [22/Sep/2014 13:37:58] "HTTP/1.1 GET 
>>>>> /_model/dimensions/input" - 500 Internal Server Error
>>>>> 127.0.0.1:52078 - - [22/Sep/2014 13:37:58] "HTTP/1.1 GET 
>>>>> /_model/encoders" - 500 Internal Server Error
>>>>>
>>>>> Thanks for your support.
>>>>>
>>>>> Kindly regards,
>>>>> Diego
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> 2014-09-22 1:31 GMT+02:00 Chetan Surpur <[email protected]>:
>>>>>>
>>>>>> Diego,
>>>>>>
>>>>>> You would use the patchCLAModel function to patch a CLAModel instance. 
>>>>>> In your case, you would do this at the place in code where 
>>>>>> run_opf_experiment.py creates a CLAModel instance.
>>>>>>
>>>>>> The demo_sp, demo_tp, etc. files all create their own CLAModel's and 
>>>>>> patch them. If you want to use OPF experiment instead, you can ignore 
>>>>>> those demo files.
>>>>>>
>>>>>> - Chetan
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Sun, Sep 21, 2014 at 1:25 PM, Diego Martinez <[email protected]> wrote:
>>>>>>>
>>>>>>> Hello Matt Hello Chetan,
>>>>>>>
>>>>>>> I am following the Mailing List Etiquette format. I hope there is no 
>>>>>>> issues this time.
>>>>>>>
>>>>>>> Suppose i want to run one of the demos that come in cerebro2.server. So 
>>>>>>> i need to define: model,sp and tp to feed the patchCLAModel funtion. My 
>>>>>>> question is: what file should i use as model? The way i am 
>>>>>>> instantiating the model is by running run_opf_experiment.py using the 
>>>>>>> best parameters found after the swarm stored at 
>>>>>>> $NUPIC/examples/swarm/simple/model_0/, finally some csv and json files 
>>>>>>> are created. So, should i patch the patchCLAModel function inside the 
>>>>>>> run_opf_experiment.py file or inside a different file under model_0 or 
>>>>>>> somewhere else? Same question goes to what file to use as sp and tp. I 
>>>>>>> have created one new file called patcher_function.py that looks like 
>>>>>>> this (pretty much is what's written in the usage section):
>>>>>>>
>>>>>>> from cerebro2.patcher import Patcher
>>>>>>> import demo_sp //this is the demo_sp.py file
>>>>>>> import demo_tp /this is the demo_tp.py file
>>>>>>> import demo_coordinate_encoder /this is the coordinate encoder demo
>>>>>>>
>>>>>>> Patcher().patchCLAModel(demo_coordinate_encoder) // i was just 
>>>>>>> testing/supposing this could be used as model
>>>>>>> Patcher().patchCLAModel(demo_sp.run) // i was just testing/supposing 
>>>>>>> this could be used as sp
>>>>>>> Patcher().patchCLAModel(demo_tp.run) // i was just testing/supposing 
>>>>>>> this could be used as tp
>>>>>>>
>>>>>>> Then i execute the file from the terminal: python patcher_function.py 
>>>>>>> and i get an error that the model does not contain the attribute 
>>>>>>> model._getSPRegion and in fact i couldn't find a different file that 
>>>>>>> contained this class. When setting sp = demo_sp.py i was able to find 
>>>>>>> sp = SpatialPooler(inputDimensions, columnDimensions etc) inside 
>>>>>>> demo_sp.py.
>>>>>>>
>>>>>>> Thanks-
>>>>>>> D.
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>

Reply via email to