Python
In this training science of data we start with the grip of working tools: Jupyter Notebook, Spyder, among others. Then we start with Numpy (numerical Python). We insist on calculation functions: maths, stats, fi nancial, etc. The heart of the training is Pandas. This module allows you to sort, filter, select, cut, truncate, categorize, group, aggregate, rotate, stack, cross, and so on. Pandas does what a spreadsheet and SQL table do and more. Pandas allows you to apply Numpy or other functions. In addition, Pandas can automate procedures and work in volume (parallel computing on multiple processors, on graphics card or on a cluster). Pandas is often the machine learning model foundation, because you have to prepare the data before using it. Finally, Pandas makes it easy to visualize data with modules like Matplotlib and Seaborn. There is also a GeoPandas extension for geomatics to enhance map visualization.
Some specializations and pointed topics not covered are available extra. We will revise them quickly.
Updated on
A data scientist, Hugues explores data collected within organizations and develops analyses and forecasts, or collaborates on projects aimed at extracting value from collected data. His projects include, among others, customer segmentation, map-based data visualization, transaction matching, and subscription lifetime…
View full profile$1,274 / participant
Public bodies and non-profits · plus taxes
$1,499 / participant
Businesses and individuals · plus taxes
Train your whole team as a private group, on your own documents, at your premises or in a virtual class, on the dates that suit you.
97% satisfaction (end-of-course evaluations, 120 evaluations) · Read all testimonials
