Closed marfox closed 9 years ago
Hi, I am trying this task and not clear about training data format due to the Italian sample. Can you explain the features you selected to be trained?
Hi, don't worry about the features, it is a syntactic transformation we are performing here. You just need to comply with the sample output syntax, this is language independent. Imagine that instead of tokens and lemmas in Italian, there could be any language (supported by TreeTagger) there.
On 3/18/15 7:24 AM, Jerry King wrote:
Hi, I am trying this task and not clear about training data format due to the Italian sample. Can you explain the features you selected to be trained?
— Reply to this email directly or view it on GitHub https://github.com/dbpedia/fact-extractor/issues/7#issuecomment-82774990.
Just a short question, is this issue still up to date? It looks like except the parameterization everything is already done.
It's still open for you to get acquainted with the code and the concepts behind
Implement step 4.i of the workflow, as per the README. You should review the related script and make it more robust. In other words:
argparse
moduleUse the following samples as input: