Using fastText word vectors

You can train the machine learning services model, by using the pre-trained word vectors trained by using fastText. This is a prerequisite task for the machine learning services model.

Proceed as follows to use the fastText word vectors.

 On-prem deployment 
  1. Download and extract the crawl-300d-2M.vec.zip file.
  2. Create the data/training_vectors folder in the $TOP/mdmui/machinelearning folder.
  3. Copy the crawl-300d-2M.vec file to the $TOP/mdmui/machinelearning/data/training_vectors folder.
 Containerized deployment 
  1. Download and extract the crawl-300d-2M.vec.zip file.
  2. View ML pod details by using the following command.
    oc get pods
  3. Go to the ML pod by using the following command.
    oc rsh <ML pod name>
  4. Create the data/training_vectors folder in the $TOP/mdmui/machinelearning folder by using the following command.
    cd $TOP/mdmui/machinelearning
mkdir -p data/training_vectors
  5. Exit the ML pod by using the following command.
    exit
  6. Copy the crawl-300d-2M.vec file to the $TOP/mdmui/machinelearning/data/training_vectors folder by using the following command.
    oc cp <path of .vec file on the server> <ML pod name>:$TOP/mdmui/machinelearning/data/training_vectors

    Example

    oc cp /opt/crawl-300d-2M.vec productmaster-ml-0: $TOP/mdmui/machinelearning/data/training_vectors