Step 3: Package the model service

step_3_package_service.py

# 1) import the custom BentoService defined above
from sentiment_analysis_service import SKSentimentAnalysis

# 2) `pack` it with required artifacts, i.e. the trained model from step 1
bento_service = SKSentimentAnalysis()
bento_service.pack('model', sentiment_lr)

# 3) save your BentoSerivce to file archive
saved_path = bento_service.save()

# 4) Start a REST API model server with the BentoService saved above to serve the model
!bentoml serve SKSentimentAnalysis:latest --port=5000