Reputation: 2590
Is it possible to run the StringToWordVector filter in Weka from the command line and get a processed output file? I'd like to pre-process my data separately before feeding it back into Weka for training. So I'm trying to run the filter, get an output file, and then do the rest. I am using a high-end GPU virtual machine with SSH-only access, so I can't use the Weka GUI, only the command line.
Upvotes: 0
Views: 378
Reputation: 979
See this
java weka.filters.unsupervised.attribute.StringToWordVector -O -L -tokenizer "weka.core.tokenizers.WordTokenizer -delimiters \" \\r\\n\\t.,;:\\\"\\'()?!-¿¡+*&#$%\\\\/=<>[]_`@\"" -W 10000000 -b -i input-train.arff -o output-train-vector.arff -r input-test.arff -s output-test-vector.arff
Upvotes: 1