Scaling with PML

To process and analyze data for typical analytics or machine learning use cases, it is common to hit storage and processing power issues. If the dataset is too big to be stored on your laptop drive, if it cannot be entirely loaded into memory, you must find practical solutions. It is, of course, possible to Read more about Scaling with PML[…]

Punch Data Science Meetup

Punch Data Science Meetup When: on the 15th of October. Where: in Rennes at the Google atelier numérique. Organised as part of the Rennes Machine Learning Meetup /French Tech Speaker: Simon Grah from Thales Theresis Research Lab Topic: Maritime Traffic Anomaly Detection We are very happy to have Simon animate our next meetup in Rennes. Many thanks to  Camille Saumard (from lumenai), to invite Read more about Punch Data Science Meetup[…]

Punchplatform Machine Learning (PML) for platform monitoring

Punchplatform periodically collects and stores data characterizing the health of the platform (metrics). It gathers both system metrics (CPU, RAM) and applicative metrics like the tuple travel time through a Storm topology. Since its last version, Punchplatform contains a specific module dedicated to machine learning based on Apache Spark: Punchplatform Machine Learning (check out this Read more about Punchplatform Machine Learning (PML) for platform monitoring[…]

Introducing PML

The PunchPlatform Machine Learning (PML) sdk lets you design and run arbitrary spark applications on your platform data. Why would you do that ? Punchplatforms process large quantities of data, parsed, normalized then indexed in ElasticSeach in real time. As a result you have powerful search capabilities through Kibana dashboards. In turn, spark applications let Read more about Introducing PML[…]