I've just examined using Apache Mahout for building an Latent dirichlet allocation model on the corpus of 30 documents. I didn't have Hadoop placed on the machine for this reason a nearby execution from the Mahout produced the resulting model. I must access the model parameters, as with the believed alpha, beta, Phi, Theta
How do i access these?
/mahout lda -i /tf-vectors -o -k 4-v 27
I can tell that has folder for every iteration(i presume) from the learning formula. Each includes a single file part-r-0000 which I don't understand how to access.
Any assistance will be appreciated.
I can not assist you to off-hands except to express that those who will help you all browse the mahout-user subscriber list and respond to questions there. It is best to repeat your question there.
See here to learn more concerning how to subscribe: https://cwiki.apache.org/MAHOUT/mailing-lists.html