clickmodels

clickmodels copied to clipboard

How to generate DBN_output.log

1

I was trying to generate DBN_output.log using below command. python2 bin/run_inference.py < content/stats.log I have used default config file. Content of stats.log is below, how can we generate DBN_output.log. 1c2aa70e-b31b-43c9-a4e5-769377682d2b...

ganesh-gupta-clapself

Problem About Parameter Estimation of UBM

1

Thanks for sharing the code. When I tested the UBM model, I encountered a question. We all know that UBM uses the EM algorithm to estimate the parameter. For example,...

jimmy-walker

the difference about your dbn model and other dbn model in github？

1

hi i find another version of dbn model。 it has so much difference with your verison。 can you explain whoes is right？ another version address： https://github.com/markovi/PyClick

idle-tmac

output explain?

1

what's output fields means?

rivenwang

problem about parameter estimation of Au

1

def _getSessionEstimate(self, positionRelevances, layout, clicks, intent): .... gamma = self.getGamma(self.gammas, k, layout, intent) # E_k_multiplier --- P(S_k = 0 | C_k) P(C_k | E_k = 1) if C_k == 0:...

idle-tmac

We have lots of class methods, but only "test" a subset of them by running `./inference.py < data/click_log_sample.tsv 2>inference.log` We need a proper unit-testing.

varepsilon

Currently all the new code has to support the intent probabilities. This requires thinking about correct probabilities and marginalization all the time and just makes the code complex. We should...

varepsilon