ChanServ changed the topic of #mlpack to: "mlpack: a fast, flexible machine learning library :: We don't always respond instantly, but we will respond; please be patient :: Logs at http://www.mlpack.org/irc/
r0bin_h00d[m] has joined #mlpack
< AnushKiniGitter[>
Had a doubt related to the implementation of activation functions in mlpack.
< AnushKiniGitter[>
Specifically, in some files the ```Deriv``` function seems to calculate derivatives from the activation of the input. This convention has been followed in files like ```methods/ann/activation_functions/tanh_function.hpp``` and ```methods/ann/activation_functions/logistic_function.hpp```. In other cases, the activation is assumed to be the input and the derivative is calculated. This can be seen in files like
< AnushKiniGitter[>
```methods/ann/activation_functions/lisht_function.hpp``` and ```methods/ann/activation_functions/spline_function.hpp```. This is done because there are cases where the derivative ```f'(x)``` cannot be expressed in terms of the activation ```f(x)```. Is this right or am I missing something here?
< AnushKiniGitter[>
(edited) ... Is this right ... => ... Is my understanding right ...
ImQ009 has joined #mlpack
VanshBansalGitte has joined #mlpack
< kartikdutt18Gitt>
Hey @vanshbansal1505, Glad to hear that you are interested. You can try out some good first issues in mlpack repo to get familiar with the code base. Also for models repo in particular, Feel free to implement any model or take up an issue from the repo.
< AyushSingh[m]>
Should I create 2 issues for the following :
< AyushSingh[m]>
1. The bug I reported on Feb 10 9:56 pm.
< AyushSingh[m]>
2. In using ToLower function from mlpack/core/util/to_lower.hpp, often a blank string is returned.
< AyushSingh[m]>
(edited) ... 1. The bug I reported on ... => ... 1. A probable bug I mentioned on ...