Hi, Im new to the community, mahout and machine learning. Upto now i've been experimenting the Gensim libraries to perform LDA analysis on a large document corpus. I do, however, have an hadoop cluster to hand and would like to explore the capabilities of Mahout. I have my files in a proprietary format - structured text effectively where a text file contains multiple document entries. I understand i need to produce sparse vectors for the LDA analysis - im happy to write a document parser and have setup my IntelliJ/maven env with the necessary dependencies. Could someone please share some code from a similar requirement to get me started - an example reading a csv file for example. All help appreciated.
Thanks Dale -- View this message in context: http://lucene.472066.n3.nabble.com/Generating-vectors-from-custom-source-tp3592300p3592300.html Sent from the Mahout User List mailing list archive at Nabble.com.
