Invitación a Charla de Divulgación “The Machine Learning workflow, with help from Python”
- 09 Nov 2017
La Charla de Divulgación titulada “The Machine Learning workflow, with help from Python”, se dictará en el aula de Posgrado 2 del Departamento de Informática, el día 13 de noviembre del presente año de 19 a 21hs.
La charla mencionada estará a cargo del Ing. Rodolfo BONNIN (https://www.linkedin.com/in/rodolfobonnin/) y es libre y gratuita.
A quien le interese figurar en la protocolización de asistencia de la misma, se le agradeceré que complete el siguiente formulario: https://goo.gl/forms/fViF2vf9o2Oia4VA2
Si bien el resumen y título de la charla están en inglés, la misma se dictará en castellano.
Resumen:
During the current year, Python has taken a leadership role in the Machine Learning realm, even surpassing domain specific specialized languages, like R. In this talk, a walkthrough of the Machine Learning problem solving workflow will be given, with references to the most useful and well known libraries for each stage. The stages to be covered, are:
Exploratory analysis: When solving machine learning problems, it’s important to take the time to analyze both the data and work ramifications beforehand. This preliminary step is flexible and less formal than all the subsequent steps we’ll cover. (Libraries: Numpy, Pandas)
Dataset Definition and Retrieving: Once we have identified the data sources, the next task is to gather all the tuples or records as a homogeneous set. (Libraries: Numpy, Pandas, OpenCV)
Enriching and Pruning the Data Features space and data cleaning. (Libraries: Numpy, Pandas)
Making Data Tractable: Dataset Preprocessing Adapting the dataset to highlight useful data. (Libraries: Numpy, Pandas, sklearn)
Model Definition: We build with machine learning are abstractions or models representing and simplifying the reality, allowing us to solve real world problems, based on a model, which we trained accordingly. (Libraries: Tensorflow, Keras)
Model Fitting: In this part of the machine learning process we have the model and data ready, and we proceed to train and validate our model. (Libraries: Tensorflow, Keras)
Model Implementation (Libraries: Tensorflow, Keras)
Results evaluation (Libraries: Tensorflow, Keras, Flask)