Snowflakes

Nano-scale snowflake from Gallium solvent. Credit: Waipapa Taumata Rau, University of AucklandAdventures in Nanotech: Growing a Metallic SnowflakeScientists working at the level of atoms are manipulating metals, opening up possibilities for creating new materials.Working at the level of atoms, scientists in New Zealand and Australia created something unexpected: tiny metallic snowflakes.Why’s that significant? Because coaxing individual atoms to cooperate is leading to a revolution in engineering and technology via…

Researchers in New Zealand and Australia have grown tiny metal snowflakes, cubes, rods and other shapes. The nanocrystals form like snow out of a liquid metal, demonstrated an intriguing new potential method for manufacturing nanoscale structures.The team ran a series of experiments using gallium, an intriguing metal with a low melting point of just under 30 °C (86 °F), meaning the temperature on a summer’s day – or even just holding it in your hand – is enough to liquefy it. The researchers made alloys by mixing gallium…

Get the lowdown on Snowflake’s PII detection featurePhoto by AbsolutVision on UnsplashIn today’s world, data collection and processing are regulated and organizations have no choice but to comply with these regulations. In consequence, companies started to rethink the way they design their information systems, data stores, and business processes with privacy in mind.One foundational element of implementing data protection principles is data classification.Data classification is often defined as the process of organizing…

When SQL turns to be fancier than you thought…Picture by Frans van Heerden on PexelsHave you ever had the need of creating data out of nothing?If you did, you’ve probably been diverted to tutorials about how to create fake data with Python, JAVA, and other programming languages.However, it turns out that SQL is not only a great language to query (DQL) and manipulate data (DML), but can also be used to create synthetic rows.In Snowflake, this can be achieved by leveraging the GENERATOR() function that, as stated by the…

How to train and predict all within a UDF uploaded to SnowparkImage from Anthi K via UnsplashIn Part 1, I showed how to train a local model, wrap it in a Python udf, push it to Snowflake using Snowpark, and use Snowpark or Snowflake Sql to make predictions using that udf.This guide, on the other hand, will show you how to make a Python udf that builds, trains, and predicts on a model all using Snowpark and Snowflake compute. We will use Regression here, while Part 1 used Classification.This article will also highlight…

Photo from Alessio Soggetti via UnsplashThis guide will show you how to use Snowflake’s Snowpark with Python UDF’s, to leverage Snowflake’s compute power to run Machine Learning models using Python.Import Librariesimport snowflake.connectorimport pandas as pdimport numpy as npfrom sqlalchemy import create_enginefrom snowflake.sqlalchemy import URLfrom snowflake.connector.pandas_tools import write_pandasfrom snowflake.snowpark.functions import udffrom snowflake.snowpark.types import IntegerType, StringType, StructType,…