Clovaを支える技術 機械学習配信基盤のご紹介
•
4 likes•6,021 views
【京都開催】MACHINE LEARNING Meetup KANSAI #1 - connpass https://mlm-kansai.connpass.com/event/81815/ での講演資料です
Clovaを支える技術 機械学習配信基盤のご紹介
- 1. Clova Keigo Hattori, Clova Center
- 2. @keigohtr Keigo Hattori keigohtr Who I am Keigo Hattori Software Engineer 2009 2009~2017 2017 ~ LINE x
- 4. 1. Clova 2. Clova ML ( ) 3. LINE (We are hiring. Welcome ML engineer!)
- 5. Clova Clova
- 7. Unique Feature • LINE • • • (WAVE )
- 9. Clova
- 10. SVM Deep LearningMachine Learning Keras Chainer Caffe scikit-learn gensim logistic regression Random Forest Neural Network Perceptron libsvm liblinear Theano
- 15. Three features of Drucker • Easy to serve • Easy to management • Easy to integrate
- 16. Drucker: Easy to serve Write your algorithm on Drucker template Features • Do not need to write proto (grpc proto) • Input/Output format is determined but flexible • Any algorithms are availble
- 17. Drucker: Easy to management Use Drucker web interface Features • Upload the latest model • Switch the models • Evaluate the performance
- 18. Drucker: Easy to integrate Integrate in the same manner Features • Do not need to write proto (grpc proto) • Input/Output format is determined but flexible
- 20. Drucker on Kubernetes Kubernetes • Rolling update • Alive monitoring Cloud Storage for ML model • Auto replication • Sync files on all servers Cloud DB for model name • Save model filename for each environment (dev/prod)
- 21. @keigohtr Keigo Hattori keigohtr Keigo Hattori @line_clova #Clova LINE Clova http://clova-blog.line.me/ja/
- 22. THANK YOU