Image recognition with Deep Learning

The training covers an implementation of an end to end deep learning project for image recognition using CNN-based architectures, including data preparation and explainability of such systems. During the classes, we are going to cover the theoretical background of Convolutional Neural Networks, discuss the state-of-the-art architectures, train a network from scratch and finally apply transfer learning for the real time problem.

MEHR LESEN
Duration: 2 days
Level: intermediate

Image recognition with Deep Learning

1800
PLN NET
Nearest term:
- ,  2019 Kraków

Group Discounts:

1+2 friends
1260.00PLN
1800.00PLN   30% off
1+1 friend
1440.00PLN
1800.00PLN   20% off
Individual measure - Kontaktiere uns

    • Prerequisites

      • Good knowledge of Python programming
      • Some experience with Docker
      • At least basic familiarity with neural networks and ML concepts

    • Outcomes

      • Access to exclusive materials covering the scope of the workshops
      • Experience in designing the CNN-based networks for image recognition

    • Agenda

      • Data augmentation
      • A theoretical background of CNNs
      • End to end CNN based project
      • Transfer learning
      • Performance tuning
      • CNNs and interpretability

Trainer

Trainer image

Kacper Łukawski

Data Science Lead

Kacper Łukawski is Data Engineer and Tech Lead at Codete. Currently involved in Big Data projects and internal research in the field of Machine Learning. An enthusiast of applying data science in various sectors.

Workshop Schedule

- ,  2019
Kraków
- ,  2019
Warszawa

Kontakt

CONTACT PERSONS

contact person

Olga Sroka

Workshop Advisor

mail icon
                  width=   olga.sroka@codete.com

phone icon width=   +48 888 011 177

contact person

Maciej Szczepański

Workshop Advisor

mail icon
                  width=   maciej.szczepanski@codete.com

phone icon width=   +48 606 42 14 76

This value should not be blank.
This value should not be blank.
Form has been submitted.
Failed to send message - please try mail@codete.com.