Anyhow, there you have the code. ODDT has another repo with code snippets:
- snippet #2 [
http://nbviewer.ipython.org/github/oddt/jcheminf/blob/master/Snippet_2.ipynb]
- train various models (RF, SVM, NN, MLR) on RFScore descriptors
- snippet #3 [
http://nbviewer.ipython.org/github/oddt/jcheminf/blob/master/Snippet_3.ipynb]
- train RF using many fingerprints (OpenBabel's and RDKit's)
----
Pozdrawiam, | Best regards,
Maciek Wójcikowski
[email protected]
2015-02-20 13:52 GMT+01:00 Igor Filippov <[email protected]>:
> Oops, sorry, got into a wrong branch! I am just repeating Maciek's answer
> it looks like!
>
> Igor
>
> On Fri, Feb 20, 2015 at 7:50 AM, Igor Filippov <[email protected]>
> wrote:
>
>> Maciek,
>>
>> I think scikit-learn is using numpy arrays and not plain Python lists.
>> They look very similar, but are not quite the same thing.
>> Maybe post a bit more complete code sample for people to play with?
>>
>> Igor
>>
>> On Fri, Feb 20, 2015 at 4:06 AM, Maciek Wójcikowski <
>> [email protected]> wrote:
>>
>>> Hello,
>>>
>>> If I can remember correctly coefficients are Numpy array. You can try
>>> model.coef_.flatten() to get flat Numpy Array. If you really want a
>>> python list, then you probably should wrap it up with list(model.
>>> coef_.flatten()).
>>>
>>> The main reason, why the vector is nested is that you can have many
>>> output values for one feature vector.
>>>
>>> PS.
>>> I could also recommend my "Open Drug Discovery Toolkit" for playing
>>> around with RDKit and scikit-learn.
>>> https://github.com/oddt/oddt
>>>
>>> ----
>>> Pozdrawiam, | Best regards,
>>> Maciek Wójcikowski
>>> [email protected]
>>>
>>> 2015-02-20 7:29 GMT+01:00 Greg Landrum <[email protected]>:
>>>
>>>>
>>>>
>>>> On Thu, Feb 19, 2015 at 11:59 PM, Matthew Lardy <[email protected]>
>>>> wrote:
>>>>
>>>>>
>>>>> I have been able to build models via scikit-learn with the RDKit
>>>>> python wrappers. That all works beautifully!
>>>>>
>>>>
>>>> It's a nice combination, isn't it?
>>>>
>>>>
>>>>> What I am struggling to get are the weights, or scalers, applied to
>>>>> each bit position. For a SVM regression model (SVR) I think that the
>>>>> values I seek are in the coef_ (if the model is created via the linear
>>>>> kernel). But, all I get is something like this when I print that out:
>>>>>
>>>>> [[-0. -0.87146158 -0.46331996 ..., 0.31076767 -0.
>>>>> -0.81882195]]
>>>>>
>>>>>
>>>> I don't really know the SVM regression approach particularly well, but
>>>> it looks like that's a vector of vectors. Is the length of the inner vector
>>>> the same as the length of the fingerprint/descriptor vector you are
>>>> providing?
>>>>
>>>> -greg
>>>>
>>>>
>>>>
>>>> ------------------------------------------------------------------------------
>>>> Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
>>>> from Actuate! Instantly Supercharge Your Business Reports and Dashboards
>>>> with Interactivity, Sharing, Native Excel Exports, App Integration &
>>>> more
>>>> Get technology previously reserved for billion-dollar corporations, FREE
>>>>
>>>> http://pubads.g.doubleclick.net/gampad/clk?id=190641631&iu=/4140/ostg.clktrk
>>>> _______________________________________________
>>>> Rdkit-discuss mailing list
>>>> [email protected]
>>>> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
>>>>
>>>>
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
>>> from Actuate! Instantly Supercharge Your Business Reports and Dashboards
>>> with Interactivity, Sharing, Native Excel Exports, App Integration & more
>>> Get technology previously reserved for billion-dollar corporations, FREE
>>>
>>> http://pubads.g.doubleclick.net/gampad/clk?id=190641631&iu=/4140/ostg.clktrk
>>> _______________________________________________
>>> Rdkit-discuss mailing list
>>> [email protected]
>>> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
>>>
>>>
>>
>
------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=190641631&iu=/4140/ostg.clktrk
_______________________________________________
Rdkit-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
